Altera EPS40F1508I6ES Stratix device handbook, volume 1 Datasheet

Stratix Device Handbook, Volume 1
101 Innovation Drive
San Jose, CA 95134
(408) 544-7000
http://www.altera.com
S5V1-3.4
Copyright © 2006 Altera Corporation. All rights reserved. Altera, The Programmable Solutions Company, the stylized Altera logo, specific device designations, and all other words and logos that are identified as trademarks and/or service marks are, unless noted otherwise, the trademarks and
service marks of Altera Corporation in the U.S. and other countries. All other product or service names are the property of their respective holders. Altera products are protected under numerous U.S. and foreign patents and pending applications, maskwork rights, and copyrights. Altera warrants
performance of its semiconductor products to current specifications in accordance with Altera's standard warranty, but reserves the right to make
changes to any products and services at any time without notice. Altera assumes no responsibility or liability arising out of the application or use of any information, product, or service described herein except as expressly agreed to in writing by Altera
Corporation. Altera customers are advised to obtain the latest version of device specifications before relying on any published information and before placing orders for products or services.
ii
Altera Corporation
Contents
Chapter Revision Dates .......................................................................... vii
About This Handbook .............................................................................. ix
How to Find Information ........................................................................................................................ ix
How to Contact Altera ............................................................................................................................. ix
Typographic Conventions ........................................................................................................................ x
Section I. Stratix Device Family Data Sheet
Revision History ............................................................................................................................ Part I–1
Chapter 1. Introduction
Introduction ............................................................................................................................................ 1–1
Features ................................................................................................................................................... 1–2
Chapter 2. Stratix Architecture
Functional Description .......................................................................................................................... 2–1
Logic Array Blocks ................................................................................................................................ 2–3
LAB Interconnects ............................................................................................................................ 2–4
LAB Control Signals ......................................................................................................................... 2–5
Logic Elements ....................................................................................................................................... 2–6
LUT Chain & Register Chain .......................................................................................................... 2–8
addnsub Signal ................................................................................................................................. 2–8
LE Operating Modes ........................................................................................................................ 2–8
Clear & Preset Logic Control ........................................................................................................ 2–13
MultiTrack Interconnect ..................................................................................................................... 2–14
TriMatrix Memory ............................................................................................................................... 2–21
Memory Modes ............................................................................................................................... 2–22
Clear Signals .................................................................................................................................... 2–24
Parity Bit Support ........................................................................................................................... 2–24
Shift Register Support .................................................................................................................... 2–25
Memory Block Size ......................................................................................................................... 2–26
Independent Clock Mode .............................................................................................................. 2–44
Input/Output Clock Mode ........................................................................................................... 2–46
Read/Write Clock Mode ............................................................................................................... 2–49
Single-Port Mode ............................................................................................................................ 2–51
Multiplier Block .............................................................................................................................. 2–57
Adder/Output Blocks ................................................................................................................... 2–61
Modes of Operation ....................................................................................................................... 2–64
Altera Corporation
iii
Contents
Stratix Device Handbook, Volume 1
DSP Block Interface ........................................................................................................................ 2–70
PLLs & Clock Networks ..................................................................................................................... 2–73
Global & Hierarchical Clocking ................................................................................................... 2–73
Enhanced & Fast PLLs ................................................................................................................... 2–81
Enhanced PLLs ............................................................................................................................... 2–87
Fast PLLs ........................................................................................................................................ 2–100
I/O Structure ...................................................................................................................................... 2–104
Double-Data Rate I/O Pins ......................................................................................................... 2–111
External RAM Interfacing ........................................................................................................... 2–115
Programmable Drive Strength ................................................................................................... 2–119
Open-Drain Output ...................................................................................................................... 2–120
Slew-Rate Control ........................................................................................................................ 2–120
Bus Hold ........................................................................................................................................ 2–121
Programmable Pull-Up Resistor ................................................................................................ 2–122
Advanced I/O Standard Support .............................................................................................. 2–122
Differential On-Chip Termination ............................................................................................. 2–127
MultiVolt I/O Interface ............................................................................................................... 2–129
High-Speed Differential I/O Support ............................................................................................ 2–130
Dedicated Circuitry ...................................................................................................................... 2–137
Byte Alignment ............................................................................................................................. 2–140
Power Sequencing & Hot Socketing ............................................................................................... 2–140
Chapter 3. Configuration & Testing
IEEE Std. 1149.1 (JTAG) Boundary-Scan Support ............................................................................ 3–1
SignalTap II Embedded Logic Analyzer ............................................................................................ 3–5
Configuration ......................................................................................................................................... 3–5
Operating Modes .............................................................................................................................. 3–5
Configuring Stratix FPGAs with JRunner .................................................................................... 3–7
Configuration Schemes ................................................................................................................... 3–7
Partial Reconfiguration .................................................................................................................... 3–7
Remote Update Configuration Modes .......................................................................................... 3–8
Stratix Automated Single Event Upset (SEU) Detection ................................................................ 3–12
Custom-Built Circuitry .................................................................................................................. 3–13
Software Interface ........................................................................................................................... 3–13
Temperature Sensing Diode ............................................................................................................... 3–13
Chapter 4. DC & Switching Characteristics
Operating Conditions ........................................................................................................................... 4–1
Power Consumption ........................................................................................................................... 4–17
Timing Model ....................................................................................................................................... 4–19
Preliminary & Final Timing .......................................................................................................... 4–19
Performance .................................................................................................................................... 4–20
Internal Timing Parameters .......................................................................................................... 4–22
External Timing Parameters ......................................................................................................... 4–33
Stratix External I/O Timing .......................................................................................................... 4–36
I/O Timing Measurement Methodology .................................................................................... 4–60
External I/O Delay Parameters .................................................................................................... 4–66
iv
Altera Corporation
Contents
Contents
Maximum Input & Output Clock Rates ...................................................................................... 4–76
High-Speed I/O Specification ........................................................................................................... 4–87
PLL Specifications ................................................................................................................................ 4–94
DLL Specifications ............................................................................................................................. 4–102
Chapter 5. Reference & Ordering Information
Software .................................................................................................................................................. 5–1
Device Pin-Outs ..................................................................................................................................... 5–1
Ordering Information ........................................................................................................................... 5–1
Index
Altera Corporation
v
Contents
vi
Stratix Device Handbook, Volume 1
Altera Corporation
Chapter Revision Dates
The chapters in this book, Stratix Device Handbook, Volume 1, were revised on the following dates.
Where chapters or groups of chapters are available separately, part numbers are listed.
Chapter 1. Introduction
Revised:
Part number:
July 2005
S51001-3.2
Chapter 2. Stratix Architecture
Revised:
July 2005
Part number: S51002-3.2
Chapter 3. Configuration & Testing
Revised:
July 2005
Part number: S51003-1.3
Chapter 4. DC & Switching Characteristics
Revised:
January 2006
Part number: S51004-3.4
Chapter 5. Reference & Ordering Information
Revised:
September 2004
Part number: S51005-2.1
Altera Corporation
vii
Chapter Revision Dates
viii
Stratix Device Handbook, Volume 1
Altera Corporation
About This Handbook
This handbook provides comprehensive information about the Altera®
Stratix family of devices.
How to Find
Information
How to Contact
Altera
Information Type
Technical support
Product literature
You can find more information in the following ways:
■
The Adobe Acrobat Find feature, which searches the text of a PDF
document. Click the binoculars toolbar icon to open the Find dialog
box.
■
Acrobat bookmarks, which serve as an additional table of contents in
PDF documents.
■
Thumbnail icons, which provide miniature previews of each page,
provide a link to the pages.
■
Numerous links, shown in green text, which allow you to jump to
related information.
For the most up-to-date information about Altera products, go to the
Altera world-wide web site at www.altera.com. For technical support on
this product, go to www.altera.com/mysupport. For additional
information about Altera products, consult the sources shown below.
USA & Canada
All Other Locations
www.altera.com/mysupport/
www.altera.com/mysupport/
(800) 800-EPLD (3753)
(7:00 a.m. to 5:00 p.m. Pacific Time)
+1 408-544-8767
7:00 a.m. to 5:00 p.m. (GMT -8:00)
Pacific Time
www.altera.com
www.altera.com
Altera literature services
[email protected]
[email protected]
Non-technical customer
service
(800) 767-3753
+ 1 408-544-7000
7:00 a.m. to 5:00 p.m. (GMT -8:00)
Pacific Time
FTP site
ftp.altera.com
ftp.altera.com
Altera Corporation
ix
Typographic Conventions
Typographic
Conventions
Visual Cue
Stratix Device Handbook, Volume 1
This document uses the typographic conventions shown below.
Meaning
Bold Type with Initial
Capital Letters
Command names, dialog box titles, checkbox options, and dialog box options are
shown in bold, initial capital letters. Example: Save As dialog box.
bold type
External timing parameters, directory names, project names, disk drive names,
filenames, filename extensions, and software utility names are shown in bold
type. Examples: fMAX, \qdesigns directory, d: drive, chiptrip.gdf file.
Italic Type with Initial Capital
Letters
Document titles are shown in italic type with initial capital letters. Example: AN 75:
High-Speed Board Designs.
Italic type
Internal timing parameters and variables are shown in italic type.
Examples: tPIA, n + 1.
Variable names are enclosed in angle brackets (< >) and shown in italic type.
Example: <file name>, <project name>.pof file.
Initial Capital Letters
Keyboard keys and menu names are shown with initial capital letters. Examples:
Delete key, the Options menu.
“Subheading Title”
References to sections within a document and titles of on-line help topics are
shown in quotation marks. Example: “Typographic Conventions.”
Courier type
Signal and port names are shown in lowercase Courier type. Examples: data1,
tdi, input. Active-low signals are denoted by suffix n, e.g., resetn.
Anything that must be typed exactly as it appears is shown in Courier type. For
example: c:\qdesigns\tutorial\chiptrip.gdf. Also, sections of an
actual file, such as a Report File, references to parts of files (e.g., the AHDL
keyword SUBDESIGN), as well as logic function names (e.g., TRI) are shown in
Courier.
1., 2., 3., and
a., b., c., etc.
Numbered steps are used in a list of items when the sequence of the items is
important, such as the steps listed in a procedure.
■
Bullets are used in a list of items when the sequence of the items is not important.
●
•
v
The checkmark indicates a procedure that consists of one step only.
1
The hand points to information that requires special attention.
r
The angled arrow indicates you should press the Enter key.
f
The feet direct you to more information on a particular topic.
x
Altera Corporation
Section I. Stratix Device
Family Data Sheet
This section provides the data sheet specifications for Stratix® devices.
They contain feature definitions of the internal architecture,
configuration and JTAG boundary-scan testing information, DC
operating conditions, AC timing parameters, a reference to power
consumption, and ordering information for Stratix devices.
This section contains the following chapters:
Revision History
Chapter
1
■
Chapter 1, Introduction
■
Chapter 2, Stratix Architecture
■
Chapter 3, Configuration & Testing
■
Chapter 4, DC & Switching Characteristics
■
Chapter 5, Reference & Ordering Information
The table below shows the revision history for Chapters 1 through 5.
Date/Version
Changes Made
July 2005, v3.2
●
Minor content changes.
September 2004, v3.1
●
Updated Table 1–6 on page 1–5.
April 2004, v3.0
●
Main section page numbers changed on first page.
Changed PCI-X to PCI-X 1.0 in “Features” on page 1–2.
Global change from SignalTap to SignalTap II.
The DSP blocks in “Features” on page 1–2 provide dedicated
implementation of multipliers that are now “faster than 300 MHz.”
●
●
●
January 2004, v2.2
●
Updated -5 speed grade device information in Table 1-6.
October 2003, v2.1
●
Add -8 speed grade device information.
July 2003, v2.0
●
Format changes throughout chapter.
Altera Corporation
Section I–1
Stratix Device Family Data Sheet
Chapter
Date/Version
2
July 2005 v3.2
Stratix Device Handbook, Volume 1
Changes Made
●
●
●
September 2004, v3.1
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
April 2004, v3.0
●
●
●
●
●
●
●
●
November 2003, v2.2
●
●
●
●
●
October 2003, v2.1
●
●
●
●
●
Section I–2
Added “Clear Signals” section.
Updated “Power Sequencing & Hot Socketing” section.
Format changes.
Updated fast regional clock networks description on page 2–73.
Deleted the word preliminary from the “specification for the maximum
time to relock is 100 µs” on page 2–90.
Added information about differential SSTL and HSTL outputs in
“External Clock Outputs” on page 2–92.
Updated notes in Figure 2–55 on page 2–93.
Added information about m counter to “Clock Multiplication &
Division” on page 2–101.
Updated Note 1 in Table 2–58 on page 2–101.
Updated description of “Clock Multiplication & Division” on
page 2–88.
Updated Table 2–22 on page 2–102.
Added references to AN 349 and AN 329 to “External RAM
Interfacing” on page 2–115.
Table 2–25 on page 2–116: updated the table, updated Notes 3 and
4. Notes 4, 5, and 6, are now Notes 5, 6, and 7, respectively.
Updated Table 2–26 on page 2–117.
Added information about PCI Compliance to page 2–120.
Table 2–32 on page 2–126: updated the table and deleted Note 1.
Updated reference to device pin-outs now being available on the web
on page 2–130.
Added Notes 4 and 5 to Table 2–36 on page 2–130.
Updated Note 3 in Table 2–37 on page 2–131.
Updated Note 5 in Table 2–41 on page 2–135.
Added note 3 to rows 11 and 12 in Table 2–18.
Deleted “Stratix and Stratix GX Device PLL Availability” table.
Added I/O standards row in Table 2–28 that support max and min
strength.
Row clk [1,3,8,10] was removed from Table 2–30.
Added checkmarks in Enhanced column for LVPECL, 3.3-V PCML,
LVDS, and HyperTransport technology rows in Table 2–32.
Removed the Left and Right I/O Banks row in Table 2–34.
Changed RCLK values in Figures 2–50 and 2–51.
External RAM Interfacing section replaced.
Added 672-pin BGA package information in Table 2–37.
Removed support for series and parallel on-chip termination.
Termination Technology renamed differential on-chip termination.
Updated the number of channels per PLL in Tables 2-38 through 242.
Updated Figures 2–65 and 2–67.
Updated DDR I information.
Updated Table 2–22.
Added Tables 2–25, 2–29, 2–30, and 2–72.
Updated Figures 2–59, 2–65, and 2–67.
Updated the Lock Detect section.
Altera Corporation
Stratix Device Family Data Sheet
Chapter
Date/Version
2
July 2003, v2.0
Changes Made
●
●
●
●
●
●
●
●
●
●
●
●
3
July 2005, v1.3
●
Updated “Operating Modes” section.
Updated “Temperature Sensing Diode” section.
Updated “IEEE Std. 1149.1 (JTAG) Boundary-Scan Support” section.
Updated “Configuration” section.
January 2005, v1.2
●
Updated limits for JTAG chain of devices.
September 2004, v1.1
●
●
●
●
4
Added reference on page 2-73 to Figures 2-50 and 2-51 for RCLK
connections.
Updated ranges for EPLL post-scale and pre-scale dividers on page
2-85.
Updated PLL Reconfiguration frequency from 25 to 22 MHz on page
2-87.
New requirement to assert are set signal each PLL when it has to reacquire lock on either a new clock after loss of lock (page 2-96).
Updated max input frequency for CLK[1,3,8,10] from 462 to 500,
Table 2-24.
Renamed impedance matching to series termination throughout.
Updated naming convention for DQS pins on page 2-112 to match pin
tables.
Added DDR SDRAM Performance Specification on page 2-117.
Added external reference resistor values for terminator technology
(page 2-136).
Added Terminator Technology Specification on pages 2-137 and 2138.
Updated Tables 2-45 to 2-49 to reflect PLL cross-bank support for
high speed differential channels at full speed.
Wire bond package performance specification for “high” speed
channels was increased to 624 Mbps from 462 Mbps throughout
chapter.
●
Added new section, “Stratix Automated Single Event Upset (SEU)
Detection” on page 3–12.
Updated description of “Custom-Built Circuitry” on page 3–13.
April 2003, v1.0
●
No new changes in Stratix Device Handbook v2.0.
January 2006, v3.4
●
Added Table 4–135.
July 2005, v3.3
●
Updated Tables 4–6 and 4–30.
Updated Tables 4–103 through 4–108.
Updated Tables 4–114 through 4–124.
Updated Table 4–129.
Added Table 4–130.
●
●
●
●
Altera Corporation
Section I–3
Stratix Device Family Data Sheet
Stratix Device Handbook, Volume 1
Chapter
Date/Version
Changes Made
4
January 2005, 3.2
●
Updated rise and fall input values.
September 2004, v3.1
●
Updated Note 3 in Table 4–8 on page 4–4.
Updated Table 4–10 on page 4–6.
Updated Table 4–20 on page 4–12 through Table 4–23 on
page 4–13. Added rows VIL(AC) and VIH(AC) to each table.
Updated Table 4–26 on page 4–14 through Table 4–29 on
page 4–15.
Updated Table 4–31 on page 4–16.
Updated description of “External Timing Parameters” on page 4–33.
Updated Table 4–36 on page 4–20.
Added signals tOUTCO, TXZ, and TZX to Figure 4–4 on page 4–33.
Added rows tM512CLKENSU and tM512CLKENH to Table 4–40 on
page 4–24.
Added rows tM4CLKENSU and tM4CLKENH to Table 4–41 on page 4–24.
Updated Note 2 in Table 4–54 on page 4–35.
Added rows tMRAMCLKENSU and tMRAMCLKENH to Table 4–42 on
page 4–25.
Updated Table 4–46 on page 4–29.
Updated Table 4–47 on page 4–29.
●
●
●
●
●
●
●
●
●
●
●
●
●
Section I–4
Altera Corporation
Stratix Device Family Data Sheet
Chapter
Date/Version
4
Changes Made
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
Altera Corporation
Table 4–48 on page 4–30: added rows tM512CLKSENSU and tM512CLKENH,
and updated symbol names.
Updated power-up current (ICCINT) required to power a Stratix
device on page 4–17.
Updated Table 4–37 on page 4–22 through Table 4–43 on
page 4–27.
Table 4–49 on page 4–31: added rows tM4KCLKENSU, tM4KCLKENH,
tM4KBESU, and tM4KBEH, deleted rows tM4KRADDRASU and tM4KRADDRH, and
updated symbol names.
Table 4–50 on page 4–31: added rows tMRAMCLKENSU, tMRAMCLKENH,
tMRAMBESU, and tMRAMBEH, deleted rows tMRAMADDRASU and
tMRAMRADDRH, and updated symbol names.
Table 4–52 on page 4–34: updated table, deleted “Conditions”
column, and added rows tXZ and tZX.
Table 4–52 on page 4–34: updated table, deleted “Conditions”
column, and added rows tXZ and tZX.
Table 4–53 on page 4–34: updated table and added rows tXZPLL and
tZXPLL.
Updated Note 2 in Table 4–53 on page 4–34.
Table 4–54 on page 4–35: updated table, deleted “Conditions”
column, and added rows tXZPLL and tZXPLL.
Updated Note 2 in Table 4–54 on page 4–35.
Deleted Note 2 from Table 4–55 on page 4–36 through Table 4–66 on
page 4–41.
Updated Table 4–55 on page 4–36 through Table 4–96 on
page 4–56. Added rows TXZ, TZX, TXZPLL, and TZXPLL.
Added Note 4 to Table 4–101 on page 4–62.
Deleted Note 1 from Table 4–67 on page 4–42 through Table 4–84 on
page 4–50.
Added new section “I/O Timing Measurement Methodology” on
page 4–60.
Deleted Note 1 from Table 4–67 on page 4–42 through Table 4–84 on
page 4–50.
Deleted Note 2 from Table 4–85 on page 4–51 through Table 4–96 on
page 4–56.
Added Note 4 to Table 4–101 on page 4–62.
Table 4–102 on page 4–64: updated table and added Note 4.
Updated description of “External I/O Delay Parameters” on
page 4–66.
Added Note 1 to Table 4–109 on page 4–73 and Table 4–110 on
page 4–74.
Updated Table 4–103 on page 4–66 through Table 4–110 on
page 4–74.
Deleted Note 2 from Table 4–103 on page 4–66 through Table 4–106
on page 4–69.
Added new paragraph about output adder delays on page 4–68.
Updated Table 4–110 on page 4–74.
Added Note 1 to Table 4–111 through Table 4–113 on page 4–75.
Section I–5
Stratix Device Family Data Sheet
Chapter
Stratix Device Handbook, Volume 1
Date/Version
4
Changes Made
●
●
●
●
●
●
April 2004, v3.0
●
Table 4–129 on page 4–96: updated table and added Note 10.
Updated Table 4–131 and Table 4–132 on page 4–100.
Updated Table 4–110 on page 4–74.
Updated Table 4–123 on page 4–85.
Updated Table 4–124 on page 4–87. through Table 4–126 on
page 4–92.
Added Note 10 to Table 4–129 on page 4–96.
Moved Table 4–127 on page 4–94 to correct order in the chapter.
Updated Table 4–131 on page 4–100 through Table 4–132 on
page 4–100.
Deleted tXZ and tZX from Figure 4–4.
Waveform was added to Figure 4–6.
The minimum and maximum duty cycle values in Note 3 of Table 4–8
were moved to a new Table 4–9.
Changes were made to values in SSTL-3 Class I and II rows in
Table 4–17.
Note 1 was added to Table 4–34.
Added tSU_R and tSU_C rows in Table 4–38.
Changed Table 4–55 title from “EP1S10 Column Pin Fast Regional
Clock External I/O Timing Parameters” to “EP1S10 External I/O
Timing on Column Pins Using Fast Regional Clock Networks.”
Changed values in Tables 4–46, 4–48 to 4–51, 4–128, and 4–131.
Added tARESET row in Tables 4–127 to 4–132.
Deleted -5 Speed Grade column in Tables 4–117 to 4–119 and 4–122
to 4–123.
Fixed differential waveform in Figure 4–1.
Added “Definition of I/O Skew” section.
Added tSU and tCO_C rows and made changes to values in tPRE and
tCLKHL rows in Table 4–46.
Values changed in the tSU and tH rows in Table 4–47.
Values changed in the tM4KCLKHL row in Table 4–49.
Values changed in the tMRAMCLKHL row in Table 4–50.
Added Table 4–51 to “Internal Timing Parameters” section.
The timing information is preliminary in Tables 4–55 through 4–96.
Table 4–111 was separated into 3 tables: Tables 4–111 to 4–113.
●
Updated Tables 4–127 through 4–129.
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
November 2003, v2.2
Section I–6
Updated Table 4–123 on page 4–85 through Table 4–126 on
page 4–92.
Updated Note 3 in Table 4–123 on page 4–85.
Table 4–125 on page 4–88: moved to correct order in chapter, and
updated table.
Updated Table 4–126 on page 4–92.
Updated Table 4–127 on page 4–94.
Updated Table 4–128 on page 4–95.
Altera Corporation
Stratix Device Family Data Sheet
Chapter
Date/Version
4
October 2003, v2.1
Changes Made
●
●
●
●
●
July 2003, v2.0
●
●
●
●
●
●
●
●
●
●
●
5
Added -8 speed grade information.
Updated performance information in Table 4–36.
Updated timing information in Tables 4–55 through 4–96.
Updated delay information in Tables 4–103 through 4–108.
Updated programmable delay information in Tables 4–100 and
4–103.
Updated clock rates in Tables 4–114 through 4–123.
Updated speed grade information in the introduction on page 4-1.
Corrected figures 4-1 & 4-2 and Table 4-9 to reflect how VID and VOD
are specified.
Added note 6 to Table 4-32.
Updated Stratix Performance Table 4-35.
Updated EP1S60 and EP1S80 timing parameters in Tables 4-82 to 493. The Stratix timing models are final for all devices.
Updated Stratix IOE programmable delay chains in Tables 4-100 to 4101.
Added single-ended I/O standard output pin delay adders for loading
in Table 4-102.
Added spec for FPLL[10..7]CLK pins in Tables 4-104 and 4-107.
Updated high-speed I/O specification for J=2 in Tables 4-114 and 4115.
Updated EPLL specification and fast PLL specification in Tables 4116 to 4-120.
September 2004, v2.1
●
Updated reference to device pin-outs on page 5–1 to indicate that
device pin-outs are no longer included in this manual and are now
available on the Altera web site.
April 2003, v1.0
●
No new changes in Stratix Device Handbook v2.0.
Altera Corporation
Section I–7
Stratix Device Family Data Sheet
Section I–8
Stratix Device Handbook, Volume 1
Altera Corporation
1. Introduction
S51001-3.2
Introduction
The Stratix® family of FPGAs is based on a 1.5-V, 0.13-µm, all-layer copper
SRAM process, with densities of up to 79,040 logic elements (LEs) and up
to 7.5 Mbits of RAM. Stratix devices offer up to 22 digital signal
processing (DSP) blocks with up to 176 (9-bit × 9-bit) embedded
multipliers, optimized for DSP applications that enable efficient
implementation of high-performance filters and multipliers. Stratix
devices support various I/O standards and also offer a complete clock
management solution with its hierarchical clock structure with up to
420-MHz performance and up to 12 phase-locked loops (PLLs).
The following shows the main sections in the Stratix Device Family Data
Sheet:
Section
Page
Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–2
Functional Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–1
Logic Array Blocks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–3
Logic Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–6
MultiTrack Interconnect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–14
TriMatrix Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–21
Digital Signal Processing Block . . . . . . . . . . . . . . . . . . . . . . . . 2–52
PLLs & Clock Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–73
I/O Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–104
High-Speed Differential I/O Support. . . . . . . . . . . . . . . . . . 2–130
Power Sequencing & Hot Socketing . . . . . . . . . . . . . . . . . . . 2–140
IEEE Std. 1149.1 (JTAG) Boundary-Scan Support. . . . . . . . . . 3–1
SignalTap II Embedded Logic Analyzer . . . . . . . . . . . . . . . . . 3–5
Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–5
Temperature Sensing Diode. . . . . . . . . . . . . . . . . . . . . . . . . . . 3–13
Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–1
Power Consumption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–17
Timing Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–19
Software. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–1
Device Pin-Outs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–1
Ordering Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–1
Altera Corporation
July 2005
1–1
Features
Features
The Stratix family offers the following features:
■
■
■
■
■
■
■
■
■
■
■
■
■
■
■
■
■
10,570 to 79,040 LEs; see Table 1–1
Up to 7,427,520 RAM bits (928,440 bytes) available without reducing
logic resources
TriMatrixTM memory consisting of three RAM block sizes to
implement true dual-port memory and first-in first-out (FIFO)
buffers
High-speed DSP blocks provide dedicated implementation of
multipliers (faster than 300 MHz), multiply-accumulate functions,
and finite impulse response (FIR) filters
Up to 16 global clocks with 22 clocking resources per device region
Up to 12 PLLs (four enhanced PLLs and eight fast PLLs) per device
provide spread spectrum, programmable bandwidth, clock switchover, real-time PLL reconfiguration, and advanced multiplication
and phase shifting
Support for numerous single-ended and differential I/O standards
High-speed differential I/O support on up to 116 channels with up
to 80 channels optimized for 840 megabits per second (Mbps)
Support for high-speed networking and communications bus
standards including RapidIO, UTOPIA IV, CSIX, HyperTransportTM
technology, 10G Ethernet XSBI, SPI-4 Phase 2 (POS-PHY Level 4),
and SFI-4
Differential on-chip termination support for LVDS
Support for high-speed external memory, including zero bus
turnaround (ZBT) SRAM, quad data rate (QDR and QDRII) SRAM,
double data rate (DDR) SDRAM, DDR fast cycle RAM (FCRAM),
and single data rate (SDR) SDRAM
Support for 66-MHz PCI (64 and 32 bit) in -6 and faster speed-grade
devices, support for 33-MHz PCI (64 and 32 bit) in -8 and faster
speed-grade devices
Support for 133-MHz PCI-X 1.0 in -5 speed-grade devices
Support for 100-MHz PCI-X 1.0 in -6 and faster speed-grade devices
Support for 66-MHz PCI-X 1.0 in -7 speed-grade devices
Support for multiple intellectual property megafunctions from
Altera MegaCore® functions and Altera Megafunction Partners
Program (AMPPSM) megafunctions
Support for remote configuration updates
1–2
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Introduction
Table 1–1. Stratix Device Features — EP1S10, EP1S20, EP1S25, EP1S30
Feature
EP1S10
EP1S20
EP1S25
EP1S30
10,570
18,460
25,660
32,470
M512 RAM blocks (32 × 18 bits)
94
194
224
295
M4K RAM blocks (128 × 36 bits)
60
82
138
171
LEs
M-RAM blocks (4K × 144 bits)
1
2
2
4
920,448
1,669,248
1,944,576
3,317,184
DSP blocks
6
10
10
12
Embedded multipliers (1)
48
80
80
96
Total RAM bits
PLLs
Maximum user I/O pins
6
6
6
10
426
586
706
726
Table 1–2. Stratix Device Features — EP1S40, EP1S60, EP1S80
Feature
LEs
EP1S40
EP1S60
EP1S80
41,250
57,120
79,040
M512 RAM blocks (32 × 18 bits)
384
574
767
M4K RAM blocks (128 × 36 bits)
183
292
364
M-RAM blocks (4K × 144 bits)
Total RAM bits
4
6
9
3,423,744
5,215,104
7,427,520
DSP blocks
14
18
22
Embedded multipliers (1)
112
144
176
PLLs
12
12
12
Maximum user I/O pins
822
1,022
1,238
Note to Tables 1–1 and 1–2:
(1)
This parameter lists the total number of 9 × 9-bit multipliers for each device. For the total number of 18 × 18-bit
multipliers per device, divide the total number of 9 × 9-bit multipliers by 2. For the total number of 36 × 36-bit
multipliers per device, divide the total number of 9 × 9-bit multipliers by 8.
Altera Corporation
July 2005
1–3
Stratix Device Handbook, Volume 1
Features
Stratix devices are available in space-saving FineLine BGA® and ball-grid
array (BGA) packages (see Tables 1–3 through 1–5). All Stratix devices
support vertical migration within the same package (for example, you
can migrate between the EP1S10, EP1S20, and EP1S25 devices in the 672pin BGA package). Vertical migration means that you can migrate to
devices whose dedicated pins, configuration pins, and power pins are the
same for a given package across device densities. For I/O pin migration
across densities, you must cross-reference the available I/O pins using
the device pin-outs for all planned densities of a given package type to
identify which I/O pins are migrational. The Quartus® II software can
automatically cross reference and place all pins except differential pins
for migration when given a device migration list. You must use the pinouts for each device to verify the differential placement migration. A
future version of the Quartus II software will support differential pin
migration.
Table 1–3. Stratix Package Options & I/O Pin Counts
484-Pin
FineLine
BGA
672-Pin
FineLine
BGA
780-Pin
FineLine
BGA
345
335
345
426
EP1S20
426
361
EP1S25
473
Device
EP1S10
672-Pin
BGA
956-Pin
BGA
426
586
473
597
1,020-Pin
FineLine
BGA
1,508-Pin
FineLine
BGA
706
EP1S30
683
597
726
EP1S40
683
615
773
822
EP1S60
683
773
1,022
EP1S80
683
773
1,203
Note to Table 1–3:
(1)
All I/O pin counts include 20 dedicated clock input pins (clk[15..0]p, clk0n, clk2n, clk9n, and clk11n)
that can be used for data inputs.
Table 1–4. Stratix BGA Package Sizes
Dimension
672 Pin
956 Pin
Pitch (mm)
1.27
1.27
(mm2)
1,225
1,600
35 × 35
40 × 40
Area
Length × width (mm × mm)
1–4
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Introduction
Table 1–5. Stratix FineLine BGA Package Sizes
484 Pin
672 Pin
780 Pin
1,020 Pin
1,508 Pin
Pitch (mm)
Dimension
1.00
1.00
1.00
1.00
1.00
(mm2)
529
729
841
1,089
1,600
23 × 23
27 × 27
29 × 29
33 × 33
40 × 40
Area
Length × width
(mm × mm)
Stratix devices are available in up to four speed grades, -5, -6, -7, and -8,
with -5 being the fastest. Table 1–6 shows Stratix device speed-grade
offerings.
Table 1–6. Stratix Device Speed Grades
484-Pin
FineLine
BGA
672-Pin
FineLine
BGA
780-Pin
FineLine
BGA
-6, -7
-5, -6, -7
-6, -7
-5, -6, -7
EP1S20
-6, -7
-5, -6, -7
EP1S25
-6, -7
Device
EP1S10
672-Pin
BGA
956-Pin
BGA
-6, -7
-5, -6, -7
-6, -7, -8
-5, -6, -7
1,020-Pin
FineLine
BGA
1,508-Pin
FineLine
BGA
-5, -6, -7
EP1S30
-5, -6, -7
-5, -6, -7, -8
-5, -6, -7
EP1S40
-5, -6, -7
-5, -6, -7, -8
-5, -6, -7
EP1S60
-6, -7
-5, -6, -7
-6, -7
EP1S80
-6, -7
-5, -6, -7
-5, -6, -7
Altera Corporation
July 2005
-5, -6, -7
1–5
Stratix Device Handbook, Volume 1
Features
1–6
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
2. Stratix Architecture
S51002-3.2
Functional
Description
Stratix® devices contain a two-dimensional row- and column-based
architecture to implement custom logic. A series of column and row
interconnects of varying length and speed provide signal interconnects
between logic array blocks (LABs), memory block structures, and DSP
blocks.
The logic array consists of LABs, with 10 logic elements (LEs) in each
LAB. An LE is a small unit of logic providing efficient implementation of
user logic functions. LABs are grouped into rows and columns across the
device.
M512 RAM blocks are simple dual-port memory blocks with 512 bits plus
parity (576 bits). These blocks provide dedicated simple dual-port or
single-port memory up to 18-bits wide at up to 318 MHz. M512 blocks are
grouped into columns across the device in between certain LABs.
M4K RAM blocks are true dual-port memory blocks with 4K bits plus
parity (4,608 bits). These blocks provide dedicated true dual-port, simple
dual-port, or single-port memory up to 36-bits wide at up to 291 MHz.
These blocks are grouped into columns across the device in between
certain LABs.
M-RAM blocks are true dual-port memory blocks with 512K bits plus
parity (589,824 bits). These blocks provide dedicated true dual-port,
simple dual-port, or single-port memory up to 144-bits wide at up to
269 MHz. Several M-RAM blocks are located individually or in pairs
within the device’s logic array.
Digital signal processing (DSP) blocks can implement up to either eight
full-precision 9 × 9-bit multipliers, four full-precision 18 × 18-bit
multipliers, or one full-precision 36 × 36-bit multiplier with add or
subtract features. These blocks also contain 18-bit input shift registers for
digital signal processing applications, including FIR and infinite impulse
response (IIR) filters. DSP blocks are grouped into two columns in each
device.
Each Stratix device I/O pin is fed by an I/O element (IOE) located at the
end of LAB rows and columns around the periphery of the device. I/O
pins support numerous single-ended and differential I/O standards.
Each IOE contains a bidirectional I/O buffer and six registers for
registering input, output, and output-enable signals. When used with
Altera Corporation
July 2005
2–1
Functional Description
dedicated clocks, these registers provide exceptional performance and
interface support with external memory devices such as DDR SDRAM,
FCRAM, ZBT, and QDR SRAM devices.
High-speed serial interface channels support transfers at up to 840 Mbps
using LVDS, LVPECL, 3.3-V PCML, or HyperTransport technology I/O
standards.
Figure 2–1 shows an overview of the Stratix device.
Figure 2–1. Stratix Block Diagram
M512 RAM Blocks for
Dual-Port Memory, Shift
Registers, & FIFO Buffers
DSP Blocks for
Multiplication and Full
Implementation of FIR Filters
M4K RAM Blocks
for True Dual-Port
Memory & Other Embedded
Memory Functions
IOEs Support DDR, PCI, GTL+, SSTL-3,
SSTL-2, HSTL, LVDS, LVPECL, PCML,
HyperTransport & other I/O Standards
IOEs
IOEs
IOEs
IOEs
LABs
LABs
LABs
LABs
LABs
IOEs
LABs
IOEs
LABs
LABs
LABs
LABs
LABs
LABs
IOEs
LABs
LABs
LABs
LABs
LABs
LABs
IOEs
LABs
LABs
LABs
LABs
LABs
LABs
IOEs
LABs
LABs
LABs
LABs
IOEs
LABs
LABs
LABs
LABs
IOEs
LABs
LABs
LABs
LABs
IOEs
LABs
LABs
LABs
LABs
IOEs
LABs
LABs
LABs
LABs
IOEs
LABs
LABs
LABs
LABs
IOEs
LABs
LABs
LABs
LABs
IOEs
LABs
LABs
LABs
LABs
IOEs
LABs
LABs
LABs
LABs
IOEs
LABs
LABs
LABs
LABs
IOEs
LABs
LABs
LABs
LABs
IOEs
LABs
LABs
LABs
LABs
IOEs
LABs
LABs
LABs
LABs
M-RAM Block
LABs
LABs
DSP
Block
2–2
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
The number of M512 RAM, M4K RAM, and DSP blocks varies by device
along with row and column numbers and M-RAM blocks. Table 2–1 lists
the resources available in Stratix devices.
Table 2–1. Stratix Device Resources
Device
M512 RAM
M4K RAM
Columns/Blocks Columns/Blocks
M-RAM
Blocks
DSP Block
Columns/Blocks
LAB
Columns
LAB Rows
EP1S10
4 / 94
2 / 60
1
2/6
40
30
EP1S20
6 / 194
2 / 82
2
2 / 10
52
41
EP1S25
6 / 224
3 / 138
2
2 / 10
62
46
EP1S30
7 / 295
3 / 171
4
2 / 12
67
57
EP1S40
8 / 384
3 / 183
4
2 / 14
77
61
EP1S60
10 / 574
4 / 292
6
2 / 18
90
73
EP1S80
11 / 767
4 / 364
9
2 / 22
101
91
Logic Array
Blocks
Altera Corporation
July 2005
Each LAB consists of 10 LEs, LE carry chains, LAB control signals, local
interconnect, LUT chain, and register chain connection lines. The local
interconnect transfers signals between LEs in the same LAB. LUT chain
connections transfer the output of one LE’s LUT to the adjacent LE for fast
sequential LUT connections within the same LAB. Register chain
connections transfer the output of one LE’s register to the adjacent LE’s
register within an LAB. The Quartus® II Compiler places associated logic
within an LAB or adjacent LABs, allowing the use of local, LUT chain,
and register chain connections for performance and area efficiency.
Figure 2–2 shows the Stratix LAB.
2–3
Stratix Device Handbook, Volume 1
Logic Array Blocks
Figure 2–2. Stratix LAB Structure
Row Interconnects of
Variable Speed & Length
Direct link
interconnect from
adjacent block
Direct link
interconnect from
adjacent block
Direct link
interconnect to
adjacent block
Direct link
interconnect to
adjacent block
Local Interconnect
LAB
Three-Sided Architecture—Local
Interconnect is Driven from Either Side by
Columns & LABs, & from Above by Rows
Column Interconnects of
Variable Speed & Length
LAB Interconnects
The LAB local interconnect can drive LEs within the same LAB. The LAB
local interconnect is driven by column and row interconnects and LE
outputs within the same LAB. Neighboring LABs, M512 RAM blocks,
M4K RAM blocks, or DSP blocks from the left and right can also drive an
LAB’s local interconnect through the direct link connection. The direct
link connection feature minimizes the use of row and column
interconnects, providing higher performance and flexibility. Each LE can
drive 30 other LEs through fast local and direct link interconnects.
Figure 2–3 shows the direct link connection.
2–4
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–3. Direct Link Connection
Direct link interconnect from
left LAB, TriMatrix memory
block, DSP block, or IOE output
Direct link interconnect from
right LAB, TriMatrix memory
block, DSP block, or IOE output
Direct link
interconnect
to right
Direct link
interconnect
to left
Local
Interconnect
LAB
LAB Control Signals
Each LAB contains dedicated logic for driving control signals to its LEs.
The control signals include two clocks, two clock enables, two
asynchronous clears, synchronous clear, asynchronous preset/load,
synchronous load, and add/subtract control signals. This gives a
maximum of 10 control signals at a time. Although synchronous load and
clear signals are generally used when implementing counters, they can
also be used with other functions.
Each LAB can use two clocks and two clock enable signals. Each LAB’s
clock and clock enable signals are linked. For example, any LE in a
particular LAB using the labclk1 signal will also use labclkena1. If
the LAB uses both the rising and falling edges of a clock, it also uses both
LAB-wide clock signals. De-asserting the clock enable signal will turn off
the LAB-wide clock.
Each LAB can use two asynchronous clear signals and an asynchronous
load/preset signal. The asynchronous load acts as a preset when the
asynchronous load data input is tied high.
Altera Corporation
July 2005
2–5
Stratix Device Handbook, Volume 1
Logic Elements
With the LAB-wide addnsub control signal, a single LE can implement a
one-bit adder and subtractor. This saves LE resources and improves
performance for logic functions such as DSP correlators and signed
multipliers that alternate between addition and subtraction depending
on data.
The LAB row clocks [7..0] and LAB local interconnect generate the LABwide control signals. The MultiTrackTM interconnect’s inherent low skew
allows clock and control signal distribution in addition to data. Figure 2–4
shows the LAB control signal generation circuit.
Figure 2–4. LAB-Wide Control Signals
Dedicated
Row LAB
Clocks
8
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
Logic Elements
labclkena2
labclkena1
labclk1
labclk2
labclr2
syncload
asyncload
or labpre
labclr1
addnsub
synclr
The smallest unit of logic in the Stratix architecture, the LE, is compact
and provides advanced features with efficient logic utilization. Each LE
contains a four-input LUT, which is a function generator that can
implement any function of four variables. In addition, each LE contains a
programmable register and carry chain with carry select capability. A
single LE also supports dynamic single bit addition or subtraction mode
selectable by an LAB-wide control signal. Each LE drives all types of
interconnects: local, row, column, LUT chain, register chain, and direct
link interconnects. See Figure 2–5.
2–6
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–5. Stratix LE
Register chain
routing from
previous LE
LAB-wide
Register Bypass
Synchronous
Load
LAB-wide
Packed
Synchronous
Register Select
Clear
LAB Carry-In
Carry-In1
addnsub
Carry-In0
Programmable
Register
LUT chain
routing to next LE
data1
data2
data3
Look-Up
Table
(LUT)
Carry
Chain
Synchronous
Load and
Clear Logic
PRN/ALD
D
Q
ADATA
Row, column,
and direct link
routing
data4
ENA
CLRN
labclr1
labclr2
labpre/aload
Chip-Wide
Reset
Asynchronous
Clear/Preset/
Load Logic
Row, column,
and direct link
routing
Local Routing
Clock &
Clock Enable
Select
Register
Feedback
Register chain
output
labclk1
labclk2
labclkena1
labclkena2
Carry-Out0
Carry-Out1
LAB Carry-Out
Each LE’s programmable register can be configured for D, T, JK, or SR
operation. Each register has data, true asynchronous load data, clock,
clock enable, clear, and asynchronous load/preset inputs. Global signals,
general-purpose I/O pins, or any internal logic can drive the register’s
clock and clear control signals. Either general-purpose I/O pins or
internal logic can drive the clock enable, preset, asynchronous load, and
asynchronous data. The asynchronous load data input comes from the
data3 input of the LE. For combinatorial functions, the register is
bypassed and the output of the LUT drives directly to the outputs of the
LE.
Each LE has three outputs that drive the local, row, and column routing
resources. The LUT or register output can drive these three outputs
independently. Two LE outputs drive column or row and direct link
routing connections and one drives local interconnect resources. This
allows the LUT to drive one output while the register drives another
output. This feature, called register packing, improves device utilization
because the device can use the register and the LUT for unrelated
Altera Corporation
July 2005
2–7
Stratix Device Handbook, Volume 1
Logic Elements
functions. Another special packing mode allows the register output to
feed back into the LUT of the same LE so that the register is packed with
its own fan-out LUT. This provides another mechanism for improved
fitting. The LE can also drive out registered and unregistered versions of
the LUT output.
LUT Chain & Register Chain
In addition to the three general routing outputs, the LEs within an LAB
have LUT chain and register chain outputs. LUT chain connections allow
LUTs within the same LAB to cascade together for wide input functions.
Register chain outputs allow registers within the same LAB to cascade
together. The register chain output allows an LAB to use LUTs for a single
combinatorial function and the registers to be used for an unrelated shift
register implementation. These resources speed up connections between
LABs while saving local interconnect resources. See “MultiTrack
Interconnect” on page 2–14 for more information on LUT chain and
register chain connections.
addnsub Signal
The LE’s dynamic adder/subtractor feature saves logic resources by
using one set of LEs to implement both an adder and a subtractor. This
feature is controlled by the LAB-wide control signal addnsub. The
addnsub signal sets the LAB to perform either A + B or A – B. The LUT
computes addition, and subtraction is computed by adding the two’s
complement of the intended subtractor. The LAB-wide signal converts to
two’s complement by inverting the B bits within the LAB and setting
carry-in = 1 to add one to the least significant bit (LSB). The LSB of an
adder/subtractor must be placed in the first LE of the LAB, where the
LAB-wide addnsub signal automatically sets the carry-in to 1. The
Quartus II Compiler automatically places and uses the adder/subtractor
feature when using adder/subtractor parameterized functions.
LE Operating Modes
The Stratix LE can operate in one of the following modes:
■
■
Normal mode
Dynamic arithmetic mode
Each mode uses LE resources differently. In each mode, eight available
inputs to the LE—the four data inputs from the LAB local interconnect;
carry-in0 and carry-in1 from the previous LE; the LAB carry-in
from the previous carry-chain LAB; and the register chain connection—
are directed to different destinations to implement the desired logic
function. LAB-wide signals provide clock, asynchronous clear,
2–8
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
asynchronous preset load, synchronous clear, synchronous load, and
clock enable control for the register. These LAB-wide signals are available
in all LE modes. The addnsub control signal is allowed in arithmetic
mode.
The Quartus II software, in conjunction with parameterized functions
such as library of parameterized modules (LPM) functions, automatically
chooses the appropriate mode for common functions such as counters,
adders, subtractors, and arithmetic functions. If required, you can also
create special-purpose functions that specify which LE operating mode to
use for optimal performance.
Normal Mode
The normal mode is suitable for general logic applications and
combinatorial functions. In normal mode, four data inputs from the LAB
local interconnect are inputs to a four-input LUT (see Figure 2–6). The
Quartus II Compiler automatically selects the carry-in or the data3
signal as one of the inputs to the LUT. Each LE can use LUT chain
connections to drive its combinatorial output directly to the next LE in the
LAB. Asynchronous load data for the register comes from the data3
input of the LE. LEs in normal mode support packed registers.
Figure 2–6. LE in Normal Mode
sload
sclear
(LAB Wide) (LAB Wide)
aload
(LAB Wide)
Register chain
connection
addnsub (LAB Wide)
(1)
data1
data2
data3
cin (from cout
of previous LE)
4-Input
LUT
ALD/PRE
ADATA Q
D
Row, column, and
direct link routing
ENA
CLRN
Row, column, and
direct link routing
clock (LAB Wide)
ena (LAB Wide)
data4
Local routing
aclr (LAB Wide)
LUT chain
connection
Register
chain output
Register Feedback
Note to Figure 2–6:
(1)
This signal is only allowed in normal mode if the LE is at the end of an adder/subtractor chain.
Altera Corporation
July 2005
2–9
Stratix Device Handbook, Volume 1
Logic Elements
Dynamic Arithmetic Mode
The dynamic arithmetic mode is ideal for implementing adders, counters,
accumulators, wide parity functions, and comparators. An LE in dynamic
arithmetic mode uses four 2-input LUTs configurable as a dynamic
adder/subtractor. The first two 2-input LUTs compute two summations
based on a possible carry-in of 1 or 0; the other two LUTs generate carry
outputs for the two chains of the carry select circuitry. As shown in
Figure 2–7, the LAB carry-in signal selects either the carry-in0 or
carry-in1 chain. The selected chain’s logic level in turn determines
which parallel sum is generated as a combinatorial or registered output.
For example, when implementing an adder, the sum output is the
selection of two possible calculated sums: data1 + data2 + carry-in0
or data1 + data2 + carry-in1. The other two LUTs use the data1 and
data2 signals to generate two possible carry-out signals—one for a carry
of 1 and the other for a carry of 0. The carry-in0 signal acts as the carry
select for the carry-out0 output and carry-in1 acts as the carry select
for the carry-out1 output. LEs in arithmetic mode can drive out
registered and unregistered versions of the LUT output.
The dynamic arithmetic mode also offers clock enable, counter enable,
synchronous up/down control, synchronous clear, synchronous load,
and dynamic adder/subtractor options. The LAB local interconnect data
inputs generate the counter enable and synchronous up/down control
signals. The synchronous clear and synchronous load options are LABwide signals that affect all registers in the LAB. The Quartus II software
automatically places any registers that are not used by the counter into
other LABs. The addnsub LAB-wide signal controls whether the LE acts
as an adder or subtractor.
2–10
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–7. LE in Dynamic Arithmetic Mode
LAB Carry-In
sload
sclear
(LAB Wide) (LAB Wide)
Register chain
connection
Carry-In0
Carry-In1
addnsub
(LAB Wide)
(1)
data1
data2
data3
LUT
LUT
aload
(LAB Wide)
ALD/PRE
ADATA Q
D
Row, column, and
direct link routing
ENA
CLRN
Row, column, and
direct link routing
clock (LAB Wide)
ena (LAB Wide)
LUT
Local routing
aclr (LAB Wide)
LUT chain
connection
LUT
Register
chain output
Register Feedback
Carry-Out0
Carry-Out1
Note to Figure 2–7:
(1)
The addnsub signal is tied to the carry input for the first LE of a carry chain only.
Carry-Select Chain
The carry-select chain provides a very fast carry-select function between
LEs in arithmetic mode. The carry-select chain uses the redundant carry
calculation to increase the speed of carry functions. The LE is configured
to calculate outputs for a possible carry-in of 1 and carry-in of 0 in
parallel. The carry-in0 and carry-in1 signals from a lower-order bit
feed forward into the higher-order bit via the parallel carry chain and feed
into both the LUT and the next portion of the carry chain. Carry-select
chains can begin in any LE within an LAB.
The speed advantage of the carry-select chain is in the parallel precomputation of carry chains. Since the LAB carry-in selects the
precomputed carry chain, not every LE is in the critical path. Only the
propagation delay between LAB carry-in generation (LE 5 and LE 10) are
now part of the critical path. This feature allows the Stratix architecture to
implement high-speed counters, adders, multipliers, parity functions,
and comparators of arbitrary width.
Altera Corporation
July 2005
2–11
Stratix Device Handbook, Volume 1
Logic Elements
Figure 2–8 shows the carry-select circuitry in an LAB for a 10-bit full
adder. One portion of the LUT generates the sum of two bits using the
input signals and the appropriate carry-in bit; the sum is routed to the
output of the LE. The register can be bypassed for simple adders or used
for accumulator functions. Another portion of the LUT generates carryout bits. An LAB-wide carry in bit selects which chain is used for the
addition of given inputs. The carry-in signal for each chain, carry-in0
or carry-in1, selects the carry-out to carry forward to the carry-in
signal of the next-higher-order bit. The final carry-out signal is routed to
an LE, where it is fed to local, row, or column interconnects.
The Quartus II Compiler automatically creates carry chain logic during
design processing, or you can create it manually during design entry.
Parameterized functions such as LPM functions automatically take
advantage of carry chains for the appropriate functions.
The Quartus II Compiler creates carry chains longer than 10 LEs by
linking LABs together automatically. For enhanced fitting, a long carry
chain runs vertically allowing fast horizontal connections to TriMatrix™
memory and DSP blocks. A carry chain can continue as far as a full
column.
2–12
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–8. Carry Select Chain
LAB Carry-In
0
1
A1
B1
LE1
A2
B2
LE2
Sum1
LAB Carry-In
Carry-In0
Carry-In1
A3
B3
LE3
A4
B4
LE4
A5
B5
LE5
0
Sum2
Sum3
LUT
data1
data2
Sum
LUT
Sum4
LUT
Sum5
LUT
1
A6
B6
LE6
A7
B7
LE7
A8
B8
LE8
A9
B9
LE9
A10
B10
LE10
Sum6
Carry-Out0
Carry-Out1
Sum7
Sum8
Sum9
Sum10
LAB Carry-Out
Clear & Preset Logic Control
LAB-wide signals control the logic for the register’s clear and preset
signals. The LE directly supports an asynchronous clear and preset
function. The register preset is achieved through the asynchronous load
of a logic high. The direct asynchronous preset does not require a NOTgate push-back technique. Stratix devices support simultaneous preset/
Altera Corporation
July 2005
2–13
Stratix Device Handbook, Volume 1
MultiTrack Interconnect
asynchronous load, and clear signals. An asynchronous clear signal takes
precedence if both signals are asserted simultaneously. Each LAB
supports up to two clears and one preset signal.
In addition to the clear and preset ports, Stratix devices provide a chipwide reset pin (DEV_CLRn) that resets all registers in the device. An
option set before compilation in the Quartus II software controls this pin.
This chip-wide reset overrides all other control signals.
MultiTrack
Interconnect
In the Stratix architecture, connections between LEs, TriMatrix memory,
DSP blocks, and device I/O pins are provided by the MultiTrack
interconnect structure with DirectDriveTM technology. The MultiTrack
interconnect consists of continuous, performance-optimized routing lines
of different lengths and speeds used for inter- and intra-design block
connectivity. The Quartus II Compiler automatically places critical design
paths on faster interconnects to improve design performance.
DirectDrive technology is a deterministic routing technology that ensures
identical routing resource usage for any function regardless of placement
within the device. The MultiTrack interconnect and DirectDrive
technology simplify the integration stage of block-based designing by
eliminating the re-optimization cycles that typically follow design
changes and additions.
The MultiTrack interconnect consists of row and column interconnects
that span fixed distances. A routing structure with fixed length resources
for all devices allows predictable and repeatable performance when
migrating through different device densities. Dedicated row
interconnects route signals to and from LABs, DSP blocks, and TriMatrix
memory within the same row. These row resources include:
■
■
■
■
Direct link interconnects between LABs and adjacent blocks.
R4 interconnects traversing four blocks to the right or left.
R8 interconnects traversing eight blocks to the right or left.
R24 row interconnects for high-speed access across the length of the
device.
The direct link interconnect allows an LAB, DSP block, or TriMatrix
memory block to drive into the local interconnect of its left and right
neighbors and then back into itself. Only one side of a M-RAM block
interfaces with direct link and row interconnects. This provides fast
communication between adjacent LABs and/or blocks without using row
interconnect resources.
The R4 interconnects span four LABs, three LABs and one M512 RAM
block, two LABs and one M4K RAM block, or two LABs and one DSP
block to the right or left of a source LAB. These resources are used for fast
2–14
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
row connections in a four-LAB region. Every LAB has its own set of R4
interconnects to drive either left or right. Figure 2–9 shows R4
interconnect connections from an LAB. R4 interconnects can drive and be
driven by DSP blocks and RAM blocks and horizontal IOEs. For LAB
interfacing, a primary LAB or LAB neighbor can drive a given R4
interconnect. For R4 interconnects that drive to the right, the primary
LAB and right neighbor can drive on to the interconnect. For R4
interconnects that drive to the left, the primary LAB and its left neighbor
can drive on to the interconnect. R4 interconnects can drive other R4
interconnects to extend the range of LABs they can drive. R4
interconnects can also drive C4 and C16 interconnects for connections
from one row to another. Additionally, R4 interconnects can drive R24
interconnects.
Figure 2–9. R4 Interconnect Connections
Adjacent LAB can
Drive onto Another
LAB's R4 Interconnect
C4, C8, and C16
Column Interconnects (1)
R4 Interconnect
Driving Right
R4 Interconnect
Driving Left
LAB
Neighbor
Primary
LAB (2)
LAB
Neighbor
Notes to Figure 2–9:
(1)
(2)
C4 interconnects can drive R4 interconnects.
This pattern is repeated for every LAB in the LAB row.
The R8 interconnects span eight LABs, M512 or M4K RAM blocks, or DSP
blocks to the right or left from a source LAB. These resources are used for
fast row connections in an eight-LAB region. Every LAB has its own set
of R8 interconnects to drive either left or right. R8 interconnect
connections between LABs in a row are similar to the R4 connections
shown in Figure 2–9, with the exception that they connect to eight LABs
to the right or left, not four. Like R4 interconnects, R8 interconnects can
drive and be driven by all types of architecture blocks. R8 interconnects
Altera Corporation
July 2005
2–15
Stratix Device Handbook, Volume 1
MultiTrack Interconnect
can drive other R8 interconnects to extend their range as well as C8
interconnects for row-to-row connections. One R8 interconnect is faster
than two R4 interconnects connected together.
R24 row interconnects span 24 LABs and provide the fastest resource for
long row connections between LABs, TriMatrix memory, DSP blocks, and
IOEs. The R24 row interconnects can cross M-RAM blocks. R24 row
interconnects drive to other row or column interconnects at every fourth
LAB and do not drive directly to LAB local interconnects. R24 row
interconnects drive LAB local interconnects via R4 and C4 interconnects.
R24 interconnects can drive R24, R4, C16, and C4 interconnects.
The column interconnect operates similarly to the row interconnect and
vertically routes signals to and from LABs, TriMatrix memory, DSP
blocks, and IOEs. Each column of LABs is served by a dedicated column
interconnect, which vertically routes signals to and from LABs, TriMatrix
memory and DSP blocks, and horizontal IOEs. These column resources
include:
■
■
■
■
■
LUT chain interconnects within an LAB
Register chain interconnects within an LAB
C4 interconnects traversing a distance of four blocks in up and down
direction
C8 interconnects traversing a distance of eight blocks in up and
down direction
C16 column interconnects for high-speed vertical routing through
the device
Stratix devices include an enhanced interconnect structure within LABs
for routing LE output to LE input connections faster using LUT chain
connections and register chain connections. The LUT chain connection
allows the combinatorial output of an LE to directly drive the fast input
of the LE right below it, bypassing the local interconnect. These resources
can be used as a high-speed connection for wide fan-in functions from
LE 1 to LE 10 in the same LAB. The register chain connection allows the
register output of one LE to connect directly to the register input of the
next LE in the LAB for fast shift registers. The Quartus II Compiler
automatically takes advantage of these resources to improve utilization
and performance. Figure 2–10 shows the LUT chain and register chain
interconnects.
2–16
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–10. LUT Chain & Register Chain Interconnects
Local Interconnect
Routing Among LEs
in the LAB
LUT Chain
Routing to
Adjacent LE
LE 1
Register Chain
Routing to Adjacent
LE's Register Input
LE 2
Local
Interconnect
LE 3
LE 4
LE 5
LE 6
LE 7
LE 8
LE 9
LE 10
The C4 interconnects span four LABs, M512, or M4K blocks up or down
from a source LAB. Every LAB has its own set of C4 interconnects to drive
either up or down. Figure 2–11 shows the C4 interconnect connections
from an LAB in a column. The C4 interconnects can drive and be driven
by all types of architecture blocks, including DSP blocks, TriMatrix
memory blocks, and vertical IOEs. For LAB interconnection, a primary
LAB or its LAB neighbor can drive a given C4 interconnect.
C4 interconnects can drive each other to extend their range as well as
drive row interconnects for column-to-column connections.
Altera Corporation
July 2005
2–17
Stratix Device Handbook, Volume 1
MultiTrack Interconnect
Figure 2–11. C4 Interconnect Connections Note (1)
C4 Interconnect
Drives Local and R4
Interconnects
up to Four Rows
C4 Interconnect
Driving Up
LAB
Row
Interconnect
Adjacent LAB can
drive onto neighboring
LAB's C4 interconnect
Local
Interconnect
C4 Interconnect
Driving Down
Note to Figure 2–11:
(1)
Each C4 interconnect can drive either up or down four rows.
2–18
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
C8 interconnects span eight LABs, M512, or M4K blocks up or down from
a source LAB. Every LAB has its own set of C8 interconnects to drive
either up or down. C8 interconnect connections between the LABs in a
column are similar to the C4 connections shown in Figure 2–11 with the
exception that they connect to eight LABs above and below. The C8
interconnects can drive and be driven by all types of architecture blocks
similar to C4 interconnects. C8 interconnects can drive each other to
extend their range as well as R8 interconnects for column-to-column
connections. C8 interconnects are faster than two C4 interconnects.
C16 column interconnects span a length of 16 LABs and provide the
fastest resource for long column connections between LABs, TriMatrix
memory blocks, DSP blocks, and IOEs. C16 interconnects can cross MRAM blocks and also drive to row and column interconnects at every
fourth LAB. C16 interconnects drive LAB local interconnects via C4 and
R4 interconnects and do not drive LAB local interconnects directly.
All embedded blocks communicate with the logic array similar to LABto-LAB interfaces. Each block (i.e., TriMatrix memory and DSP blocks)
connects to row and column interconnects and has local interconnect
regions driven by row and column interconnects. These blocks also have
direct link interconnects for fast connections to and from a neighboring
LAB. All blocks are fed by the row LAB clocks, labclk[7..0].
Altera Corporation
July 2005
2–19
Stratix Device Handbook, Volume 1
MultiTrack Interconnect
Table 2–2 shows the Stratix device’s routing scheme.
Table 2–2. Stratix Device Routing Scheme
Direct Link
Interconnect
v
R4 Interconnect
v
R8 Interconnect
v
v
v
v
R24
Interconnect
v
C4 Interconnect
v
C8 Interconnect
v
v
v
v
v
v
v
v
v
v
v
v
v
v
v
M4K RAM Block
v
v
v
v
v
v
v
v
M-RAM Block
v
v
Row IOE
v
2–20
Stratix Device Handbook, Volume 1
v
v
v
v
v
v
v
v
Column IOE
v
v
v
DSP Blocks
v
v
v
v
v
v
M512 RAM
Block
LE
v
v
v
C16
Interconnect
v
v
Row IOE
v
Column IOE
Local
Interconnect
DSP Blocks
v
M-RAM Block
v
Register Chain
M4K RAM Block
LUT Chain
M512 RAM Block
LE
C16 Interconnect
C8 Interconnect
C4 Interconnect
R24 Interconnect
R8 Interconnect
R4 Interconnect
Direct Link Interconnect
Local Interconnect
LUT Chain
Source
Register Chain
Destination
v
v
v
v
v
v
v
v
v
v
v
Altera Corporation
July 2005
Stratix Architecture
TriMatrix
Memory
TriMatrix memory consists of three types of RAM blocks: M512, M4K,
and M-RAM blocks. Although these memory blocks are different, they
can all implement various types of memory with or without parity,
including true dual-port, simple dual-port, and single-port RAM, ROM,
and FIFO buffers. Table 2–3 shows the size and features of the different
RAM blocks.
Table 2–3. TriMatrix Memory Features (Part 1 of 2)
Memory Feature
Maximum
performance
M512 RAM Block M4K RAM Block
(32 × 18 Bits)
(128 × 36 Bits)
(1)
True dual-port
memory
(1)
(1)
v
v
Simple dual-port
memory
v
v
v
Single-port memory
v
v
v
Shift register
v
v
ROM
v
v
(2)
FIFO buffer
v
v
v
v
v
Parity bits
v
v
v
Mixed clock mode
v
v
v
Memory initialization
v
v
Simple dual-port
memory mixed width
support
v
v
v
v
v
Byte enable
True dual-port
memory mixed width
support
Altera Corporation
July 2005
M-RAM Block
(4K × 144 Bits)
Power-up conditions
Outputs cleared
Outputs cleared
Outputs
unknown
Register clears
Input and output
registers
Input and output
registers
Output registers
Mixed-port readduring-write
Unknown
output/old data
Unknown
output/old data
Unknown output
2–21
Stratix Device Handbook, Volume 1
TriMatrix Memory
Table 2–3. TriMatrix Memory Features (Part 2 of 2)
Memory Feature
Configurations
M512 RAM Block M4K RAM Block
(32 × 18 Bits)
(128 × 36 Bits)
512 × 1
256 × 2
128 × 4
64 × 8
64 × 9
32 × 16
32 × 18
4K × 1
2K × 2
1K × 4
512 × 8
512 × 9
256 × 16
256 × 18
128 × 32
128 × 36
M-RAM Block
(4K × 144 Bits)
64K × 8
64K × 9
32K × 16
32K × 18
16K × 32
16K × 36
8K × 64
8K × 72
4K × 128
4K × 144
Notes to Table 2–3:
(1)
(2)
See Table 4–36 for maximum performance information.
The M-RAM block does not support memory initializations. However, the
M-RAM block can emulate a ROM function using a dual-port RAM bock. The
Stratix device must write to the dual-port memory once and then disable the
write-enable ports afterwards.
1
Violating the setup or hold time on the address registers could
corrupt the memory contents. This applies to both read and
write operations.
Memory Modes
TriMatrix memory blocks include input registers that synchronize writes
and output registers to pipeline designs and improve system
performance. M4K and M-RAM memory blocks offer a true dual-port
mode to support any combination of two-port operations: two reads, two
writes, or one read and one write at two different clock frequencies.
Figure 2–12 shows true dual-port memory.
Figure 2–12. True Dual-Port Memory Configuration
A
dataA[ ]
addressA[ ]
wrenA
clockA
clockenA
qA[ ]
aclrA
2–22
Stratix Device Handbook, Volume 1
B
dataB[ ]
addressB[ ]
wrenB
clockB
clockenB
qB[ ]
aclrB
Altera Corporation
July 2005
Stratix Architecture
In addition to true dual-port memory, the memory blocks support simple
dual-port and single-port RAM. Simple dual-port memory supports a
simultaneous read and write and can either read old data before the write
occurs or just read the don’t care bits. Single-port memory supports nonsimultaneous reads and writes, but the q[] port will output the data once
it has been written to the memory (if the outputs are not registered) or
after the next rising edge of the clock (if the outputs are registered). For
more information, see Chapter 2, TriMatrix Embedded Memory Blocks in
Stratix & Stratix GX Devices of the Stratix Device Handbook, Volume 2.
Figure 2–13 shows these different RAM memory port configurations for
TriMatrix memory.
Figure 2–13. Simple Dual-Port & Single-Port Memory Configurations
Simple Dual-Port Memory
data[ ]
wraddress[ ]
wren
inclock
inclocken
inaclr
rdaddress[ ]
rden
q[ ]
outclock
outclocken
outaclr
Single-Port Memory (1)
data[ ]
address[ ]
wren
inclock
inclocken
inaclr
q[ ]
outclock
outclocken
outaclr
Note to Figure 2–13:
(1)
Two single-port memory blocks can be implemented in a single M4K block as long
as each of the two independent block sizes is equal to or less than half of the M4K
block size.
The memory blocks also enable mixed-width data ports for reading and
writing to the RAM ports in dual-port RAM configuration. For example,
the memory block can be written in ×1 mode at port A and read out in ×16
mode from port B.
Altera Corporation
July 2005
2–23
Stratix Device Handbook, Volume 1
TriMatrix Memory
TriMatrix memory architecture can implement pipelined RAM by
registering both the input and output signals to the RAM block. All
TriMatrix memory block inputs are registered providing synchronous
write cycles. In synchronous operation, the memory block generates its
own self-timed strobe write enable (WREN) signal derived from the global
or regional clock. In contrast, a circuit using asynchronous RAM must
generate the RAM WREN signal while ensuring its data and address
signals meet setup and hold time specifications relative to the WREN
signal. The output registers can be bypassed. Flow-through reading is
possible in the simple dual-port mode of M512 and M4K RAM blocks by
clocking the read enable and read address registers on the negative clock
edge and bypassing the output registers.
Two single-port memory blocks can be implemented in a single M4K
block as long as each of the two independent block sizes is equal to or less
than half of the M4K block size.
The Quartus II software automatically implements larger memory by
combining multiple TriMatrix memory blocks. For example, two
256 × 16-bit RAM blocks can be combined to form a 256 × 32-bit RAM
block. Memory performance does not degrade for memory blocks using
the maximum number of words available in one memory block. Logical
memory blocks using less than the maximum number of words use
physical blocks in parallel, eliminating any external control logic that
would increase delays. To create a larger high-speed memory block, the
Quartus II software automatically combines memory blocks with LE
control logic.
Clear Signals
When applied to input registers, the asynchronous clear signal for the
TriMatrix embedded memory immediately clears the input registers.
However, the output of the memory block does not show the effects until
the next clock edge. When applied to output registers, the asynchronous
clear signal clears the output registers and the effects are seen
immediately.
Parity Bit Support
The memory blocks support a parity bit for each byte. The parity bit,
along with internal LE logic, can implement parity checking for error
detection to ensure data integrity. You can also use parity-size data words
to store user-specified control bits. In the M4K and M-RAM blocks, byte
enables are also available for data input masking during write operations.
2–24
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Shift Register Support
You can configure embedded memory blocks to implement shift registers
for DSP applications such as pseudo-random number generators, multichannel filtering, auto-correlation, and cross-correlation functions. These
and other DSP applications require local data storage, traditionally
implemented with standard flip-flops, which can quickly consume many
logic cells and routing resources for large shift registers. A more efficient
alternative is to use embedded memory as a shift register block, which
saves logic cell and routing resources and provides a more efficient
implementation with the dedicated circuitry.
The size of a w × m × n shift register is determined by the input data
width (w), the length of the taps (m), and the number of taps (n). The size
of a w × m × n shift register must be less than or equal to the maximum
number of memory bits in the respective block: 576 bits for the M512
RAM block and 4,608 bits for the M4K RAM block. The total number of
shift register outputs (number of taps n × width w) must be less than the
maximum data width of the RAM block (18 for M512 blocks, 36 for M4K
blocks). To create larger shift registers, the memory blocks are cascaded
together.
Data is written into each address location at the falling edge of the clock
and read from the address at the rising edge of the clock. The shift register
mode logic automatically controls the positive and negative edge
clocking to shift the data in one clock cycle. Figure 2–14 shows the
TriMatrix memory block in the shift register mode.
Altera Corporation
July 2005
2–25
Stratix Device Handbook, Volume 1
TriMatrix Memory
Figure 2–14. Shift Register Memory Configuration
w × m × n Shift Register
m-Bit Shift Register
w
w
m-Bit Shift Register
w
w
n Number
of Taps
m-Bit Shift Register
w
w
m-Bit Shift Register
w
w
Memory Block Size
TriMatrix memory provides three different memory sizes for efficient
application support. The large number of M512 blocks are ideal for
designs with many shallow first-in first-out (FIFO) buffers. M4K blocks
provide additional resources for channelized functions that do not
require large amounts of storage. The M-RAM blocks provide a large
single block of RAM ideal for data packet storage. The different-sized
blocks allow Stratix devices to efficiently support variable-sized memory
in designs.
The Quartus II software automatically partitions the user-defined
memory into the embedded memory blocks using the most efficient size
combinations. You can also manually assign the memory to a specific
block size or a mixture of block sizes.
2–26
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
M512 RAM Block
The M512 RAM block is a simple dual-port memory block and is useful
for implementing small FIFO buffers, DSP, and clock domain transfer
applications. Each block contains 576 RAM bits (including parity bits).
M512 RAM blocks can be configured in the following modes:
■
■
■
■
■
Simple dual-port RAM
Single-port RAM
FIFO
ROM
Shift register
When configured as RAM or ROM, you can use an initialization file to
pre-load the memory contents.
The memory address depths and output widths can be configured as
512 × 1, 256 × 2, 128 × 4, 64 × 8 (64 × 9 bits with parity), and 32 × 16
(32 × 18 bits with parity). Mixed-width configurations are also possible,
allowing different read and write widths. Table 2–4 summarizes the
possible M512 RAM block configurations.
Table 2–4. M512 RAM Block Configurations (Simple Dual-Port RAM)
Write Port
Read Port
512 × 1
256 × 2
128 × 4
64 × 8
32 × 16
512 × 1
v
v
v
v
v
256 × 2
v
v
v
v
v
128 × 4
v
v
v
64 × 8
v
v
32 × 16
v
v
64 × 9
32 × 18
64 × 9
32 × 18
v
v
v
v
v
v
When the M512 RAM block is configured as a shift register block, a shift
register of size up to 576 bits is possible.
The M512 RAM block can also be configured to support serializer and
deserializer applications. By using the mixed-width support in
combination with DDR I/O standards, the block can function as a
SERDES to support low-speed serial I/O standards using global or
regional clocks. See “I/O Structure” on page 2–104 for details on
dedicated SERDES in Stratix devices.
Altera Corporation
July 2005
2–27
Stratix Device Handbook, Volume 1
TriMatrix Memory
M512 RAM blocks can have different clocks on its inputs and outputs.
The wren, datain, and write address registers are all clocked together
from one of the two clocks feeding the block. The read address, rden, and
output registers can be clocked by either of the two clocks driving the
block. This allows the RAM block to operate in read/write or
input/output clock modes. Only the output register can be bypassed. The
eight labclk signals or local interconnect can drive the inclock,
outclock, wren, rden, inclr, and outclr signals. Because of the
advanced interconnect between the LAB and M512 RAM blocks, LEs can
also control the wren and rden signals and the RAM clock, clock enable,
and asynchronous clear signals. Figure 2–15 shows the M512 RAM block
control signal generation logic.
The RAM blocks within Stratix devices have local interconnects to allow
LEs and interconnects to drive into RAM blocks. The M512 RAM block
local interconnect is driven by the R4, R8, C4, C8, and direct link
interconnects from adjacent LABs. The M512 RAM blocks can
communicate with LABs on either the left or right side through these row
interconnects or with LAB columns on the left or right side with the
column interconnects. Up to 10 direct link input connections to the M512
RAM block are possible from the left adjacent LABs and another
10 possible from the right adjacent LAB. M512 RAM outputs can also
connect to left and right LABs through 10 direct link interconnects. The
M512 RAM block has equal opportunity for access and performance to
and from LABs on either its left or right side. Figure 2–16 shows the M512
RAM block to logic array interface.
2–28
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–15. M512 RAM Block Control Signals
Dedicated
Row LAB
Clocks
8
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
Altera Corporation
July 2005
outclocken
inclocken
inclock
outclock
outclr
wren
rden
inclr
2–29
Stratix Device Handbook, Volume 1
TriMatrix Memory
Figure 2–16. M512 RAM Block LAB Row Interface
C4 and C8
Interconnects
R4 and R8
Interconnects
10
Direct link
interconnect
to adjacent LAB
Direct link
interconnect
to adjacent LAB
dataout
M512 RAM
Block
Direct link
interconnect
from adjacent LAB
Direct link
interconnect
from adjacent LAB
Control
Signals
datain
address
Clocks
2
8
Small RAM Block Local
Interconnect Region
LAB Row Clocks
M4K RAM Blocks
The M4K RAM block includes support for true dual-port RAM. The M4K
RAM block is used to implement buffers for a wide variety of applications
such as storing processor code, implementing lookup schemes, and
implementing larger memory applications. Each block contains
4,608 RAM bits (including parity bits). M4K RAM blocks can be
configured in the following modes:
■
■
■
■
■
■
True dual-port RAM
Simple dual-port RAM
Single-port RAM
FIFO
ROM
Shift register
When configured as RAM or ROM, you can use an initialization file to
pre-load the memory contents.
2–30
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
The memory address depths and output widths can be configured as
4,096 × 1, 2,048 × 2, 1,024 × 4, 512 × 8 (or 512 × 9 bits), 256 × 16 (or
256 × 18 bits), and 128 × 32 (or 128 × 36 bits). The 128 × 32- or 36-bit
configuration is not available in the true dual-port mode. Mixed-width
configurations are also possible, allowing different read and write
widths. Tables 2–5 and 2–6 summarize the possible M4K RAM block
configurations.
Table 2–5. M4K RAM Block Configurations (Simple Dual-Port)
Write Port
Read Port
4K × 1
2K × 2
1K × 4 512 × 8 256 × 16
128 × 32 512 × 9 256 × 18
4K × 1
v
v
v
v
v
v
2K × 2
v
v
v
v
v
v
1K × 4
v
v
v
v
v
v
512 × 8
v
v
v
v
v
v
256 × 16
v
v
v
v
v
v
128 × 32
v
v
v
v
v
v
128 × 36
512 × 9
v
v
v
256 × 18
v
v
v
128 × 36
v
v
v
Table 2–6. M4K RAM Block Configurations (True Dual-Port)
Port B
Port A
4K × 1
2K × 2
1K × 4
512 × 8
256 × 16
512 × 9
256 × 18
4K × 1
v
v
v
v
v
2K × 2
v
v
v
v
v
1K × 4
v
v
v
v
v
512 × 8
v
v
v
v
v
256 × 16
v
v
v
v
v
512 × 9
v
v
256 × 18
v
v
When the M4K RAM block is configured as a shift register block, you can
create a shift register up to 4,608 bits (w × m × n).
Altera Corporation
July 2005
2–31
Stratix Device Handbook, Volume 1
TriMatrix Memory
M4K RAM blocks support byte writes when the write port has a data
width of 16, 18, 32, or 36 bits. The byte enables allow the input data to be
masked so the device can write to specific bytes. The unwritten bytes
retain the previous written value. Table 2–7 summarizes the byte
selection.
Table 2–7. Byte Enable for M4K Blocks Notes (1), (2)
byteena[3..0]
datain ×18
datain ×36
[0] = 1
[8..0]
[8..0]
[1] = 1
[17..9]
[17..9]
[2] = 1
–
[26..18]
[3] = 1
–
[35..27]
Notes to Table 2–7:
(1)
(2)
Any combination of byte enables is possible.
Byte enables can be used in the same manner with 8-bit words, i.e., in × 16 and
× 32 modes.
The M4K RAM blocks allow for different clocks on their inputs and
outputs. Either of the two clocks feeding the block can clock M4K RAM
block registers (renwe, address, byte enable, datain, and output
registers). Only the output register can be bypassed. The eight labclk
signals or local interconnects can drive the control signals for the A and B
ports of the M4K RAM block. LEs can also control the clock_a,
clock_b, renwe_a, renwe_b, clr_a, clr_b, clocken_a, and
clocken_b signals, as shown in Figure 2–17.
The R4, R8, C4, C8, and direct link interconnects from adjacent LABs
drive the M4K RAM block local interconnect. The M4K RAM blocks can
communicate with LABs on either the left or right side through these row
resources or with LAB columns on either the right or left with the column
resources. Up to 10 direct link input connections to the M4K RAM Block
are possible from the left adjacent LABs and another 10 possible from the
right adjacent LAB. M4K RAM block outputs can also connect to left and
right LABs through 10 direct link interconnects each. Figure 2–18 shows
the M4K RAM block to logic array interface.
2–32
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–17. M4K RAM Block Control Signals
Dedicated
Row LAB
Clocks
8
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
alcr_a
clocken_a
clock_b
renwe_b
Local
Interconnect
Local
Interconnect
clock_a
renwe_a
alcr_b
clocken_b
Figure 2–18. M4K RAM Block LAB Row Interface
C4 and C8
Interconnects
Direct link
interconnect
to adjacent LAB
R4 and R8
Interconnects
10
Direct link
interconnect
to adjacent LAB
dataout
Direct link
interconnect
from adjacent LAB
M4K RAM
Block
Direct link
interconnect
from adjacent LAB
Byte enable
Control
Signals
Clocks
address
datain
8
M4K RAM Block Local
Interconnect Region
Altera Corporation
July 2005
LAB Row Clocks
2–33
Stratix Device Handbook, Volume 1
TriMatrix Memory
M-RAM Block
The largest TriMatrix memory block, the M-RAM block, is useful for
applications where a large volume of data must be stored on-chip. Each
block contains 589,824 RAM bits (including parity bits). The M-RAM
block can be configured in the following modes:
■
■
■
■
True dual-port RAM
Simple dual-port RAM
Single-port RAM
FIFO RAM
You cannot use an initialization file to initialize the contents of a M-RAM
block. All M-RAM block contents power up to an undefined value. Only
synchronous operation is supported in the M-RAM block, so all inputs
are registered. Output registers can be bypassed. The memory address
and output width can be configured as 64K × 8 (or 64K × 9 bits), 32K × 16
(or 32K × 18 bits), 16K × 32 (or 16K × 36 bits), 8K × 64 (or 8K × 72 bits), and
4K × 128 (or 4K × 144 bits). The 4K × 128 configuration is unavailable in
true dual-port mode because there are a total of 144 data output drivers
in the block. Mixed-width configurations are also possible, allowing
different read and write widths. Tables 2–8 and 2–9 summarize the
possible M-RAM block configurations:
Table 2–8. M-RAM Block Configurations (Simple Dual-Port)
Write Port
Read Port
64K × 9
32K × 18
16K × 36
8K × 72
64K × 9
v
v
v
v
32K × 18
v
v
v
v
16K × 36
v
v
v
v
8K × 72
v
v
v
v
4K × 144
2–34
Stratix Device Handbook, Volume 1
4K × 144
v
Altera Corporation
July 2005
Stratix Architecture
Table 2–9. M-RAM Block Configurations (True Dual-Port)
Port B
Port A
64K × 9
32K × 18
16K × 36
8K × 72
64K × 9
v
v
v
v
32K × 18
v
v
v
v
16K × 36
v
v
v
v
8K × 72
v
v
v
v
The read and write operation of the memory is controlled by the WREN
signal, which sets the ports into either read or write modes. There is no
separate read enable (RE) signal.
Writing into RAM is controlled by both the WREN and byte enable
(byteena) signals for each port. The default value for the byteena
signal is high, in which case writing is controlled only by the WREN signal.
The byte enables are available for the ×18, ×36, and ×72 modes. In the
×144 simple dual-port mode, the two sets of byteena signals
(byteena_a and byteena_b) are combined to form the necessary
16 byte enables. Tables 2–10 and 2–11 summarize the byte selection.
Table 2–10. Byte Enable for M-RAM Blocks Notes (1), (2)
Altera Corporation
July 2005
byteena[3..0]
datain ×18
datain ×36
datain ×72
[0] = 1
[8..0]
[8..0]
[8..0]
[1] = 1
[17..9]
[17..9]
[17..9]
[2] = 1
–
[26..18]
[26..18]
[3] = 1
–
[35..27]
[35..27]
[4] = 1
–
–
[44..36]
[5] = 1
–
–
[53..45]
[6] = 1
–
–
[62..54]
[7] = 1
–
–
[71..63]
2–35
Stratix Device Handbook, Volume 1
TriMatrix Memory
Table 2–11. M-RAM Combined Byte Selection for ×144 Mode Notes (1), (2)
byteena[15..0]
datain ×144
[0] = 1
[8..0]
[1] = 1
[17..9]
[2] = 1
[26..18]
[3] = 1
[35..27]
[4] = 1
[44..36]
[5] = 1
[53..45]
[6] = 1
[62..54]
[7] = 1
[71..63]
[8] = 1
[80..72]
[9] = 1
[89..81]
[10] = 1
[98..90]
[11] = 1
[107..99]
[12] = 1
[116..108]
[13] = 1
[125..117]
[14] = 1
[134..126]
[15] = 1
[143..135]
Notes to Tables 2–10 and 2–11:
(1)
(2)
Any combination of byte enables is possible.
Byte enables can be used in the same manner with 8-bit words, i.e., in × 16, × 32,
× 64, and × 128 modes.
Similar to all RAM blocks, M-RAM blocks can have different clocks on
their inputs and outputs. All input registers—renwe, datain, address,
and byte enable registers—are clocked together from either of the two
clocks feeding the block. The output register can be bypassed. The eight
labclk signals or local interconnect can drive the control signals for the
A and B ports of the M-RAM block. LEs can also control the clock_a,
clock_b, renwe_a, renwe_b, clr_a, clr_b, clocken_a, and
clocken_b signals as shown in Figure 2–19.
2–36
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–19. M-RAM Block Control Signals
Dedicated
Row LAB
Clocks
8
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
clocken_b
clocken_a
clock_a
clock_b
renwe_b
aclr_b
aclr_a
renwe_a
One of the M-RAM block’s horizontal sides drive the address and control
signal (clock, renwe, byteena, etc.) inputs. Typically, the horizontal side
closest to the device perimeter contains the interfaces. The one exception
is when two M-RAM blocks are paired next to each other. In this case, the
side of the M-RAM block opposite the common side of the two blocks
contains the input interface. The top and bottom sides of any M-RAM
block contain data input and output interfaces to the logic array. The top
side has 72 data inputs and 72 data outputs for port B, and the bottom side
has another 72 data inputs and 72 data outputs for port A. Figure 2–20
shows an example floorplan for the EP1S60 device and the location of the
M-RAM interfaces.
Altera Corporation
July 2005
2–37
Stratix Device Handbook, Volume 1
TriMatrix Memory
Figure 2–20. EP1S60 Device with M-RAM Interface Locations Note (1)
Independent M-RAM blocks
interface to top, bottom, and side facing
device perimeter for easy access
to horizontal I/O pins.
M-RAM pairs interface to
top, bottom, and side opposite
of block-to-block border.
DSP
Blocks
M-RAM
Block
M-RAM
Block
M-RAM
Block
M-RAM
Block
M-RAM
Block
M-RAM
Block
M512
Blocks
M4K
Blocks
LABs
DSP
Blocks
Note to Figure 2–20:
(1)
Device shown is an EP1S60 device. The number and position of M-RAM blocks varies in other devices.
The M-RAM block local interconnect is driven by the R4, R8, C4, C8, and
direct link interconnects from adjacent LABs. For independent M-RAM
blocks, up to 10 direct link address and control signal input connections
to the M-RAM block are possible from the left adjacent LABs for M-RAM
2–38
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
blocks facing to the left, and another 10 possible from the right adjacent
LABs for M-RAM blocks facing to the right. For column interfacing, every
M-RAM column unit connects to the right and left column lines, allowing
each M-RAM column unit to communicate directly with three columns of
LABs. Figures 2–21 through 2–23 show the interface between the M-RAM
block and the logic array.
Altera Corporation
July 2005
2–39
Stratix Device Handbook, Volume 1
TriMatrix Memory
Figure 2–21. Left-Facing M-RAM to Interconnect Interface Notes (1), (2)
M512 RAM Block Columns
Row Unit Interface
Allows LAB Rows to
Drive Address and
Control Signals to
M-RAM Block
LABs in Column
M-RAM Boundary
Column Interface Block
Drives to and from
C4 and C8 Interconnects
B1
B2
B3
B4
B5
B6
A4
A5
A6
Port B
R11
R10
R9
R8
R7
M-RAM Block
R6
R5
R4
R3
R2
R1
Port A
A1
A2
A3
Column Interface Block
Allows LAB Columns to
Drive datain and dataout to
and from M-RAM Block
LABs in Row
M-RAM Boundary
LAB Interface
Blocks
Notes to Figure 2–21:
(1)
(2)
Only R24 and C16 interconnects cross the M-RAM block boundaries.
The right-facing M-RAM block has interface blocks on the right side, but none on the left. B1 to B6 and A1 to A6
orientation is clipped across the vertical axis for right-facing M-RAM blocks.
2–40
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–22. M-RAM Row Unit Interface to Interconnect
C4 and C8 Interconnects
R4 and R8 Interconnects
M-RAM Block
LAB
10
Direct Link
Interconnects
Up to 24
addressa
addressb
renwe_a
renwe_b
byteenaA[ ]
byteenaB[ ]
clocken_a
clocken_b
clock_a
clock_b
aclr_a
aclr_b
Row Interface Block
M-RAM Block to
LAB Row Interface
Block Interconnect Region
Altera Corporation
July 2005
2–41
Stratix Device Handbook, Volume 1
TriMatrix Memory
Figure 2–23. M-RAM Column Unit Interface to Interconnect
C4 and C8 Interconnects
LAB
LAB
LAB
M-RAM Block to
LAB Row Interface
Block Interconnect
Region
Column Interface
Block
12
12
datain
dataout
M-RAM Block
2–42
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Table 2–12 shows the input and output data signal connections for the
column units (B1 to B6 and A1 to A6). It also shows the address and
control signal input connections to the row units (R1 to R11).
Table 2–12. M-RAM Row & Column Interface Unit Signals
Altera Corporation
July 2005
Unit Interface Block
Input SIgnals
R1
addressa[7..0]
R2
addressa[15..8]
R3
byte_enable_a[7..0]
renwe_a
R4
-
R5
-
R6
clock_a
clocken_a
clock_b
clocken_b
R7
-
R8
-
R9
byte_enable_b[7..0]
renwe_b
R10
addressb[15..8]
Output Signals
R11
addressb[7..0]
B1
datain_b[71..60]
dataout_b[71..60]
B2
datain_b[59..48]
dataout_b[59..48]
B3
datain_b[47..36]
dataout_b[47..36]
B4
datain_b[35..24]
dataout_b[35..24]
B5
datain_b[23..12]
dataout_b[23..12]
B6
datain_b[11..0]
dataout_b[11..0]
A1
datain_a[71..60]
dataout_a[71..60]
A2
datain_a[59..48]
dataout_a[59..48]
A3
datain_a[47..36]
dataout_a[47..36]
A4
datain_a[35..24]
dataout_a[35..24]
A5
datain_a[23..12]
dataout_a[23..12]
A6
datain_a[11..0]
dataout_a[11..0]
2–43
Stratix Device Handbook, Volume 1
TriMatrix Memory
Independent Clock Mode
The memory blocks implement independent clock mode for true dualport memory. In this mode, a separate clock is available for each port
(ports A and B). Clock A controls all registers on the port A side, while
clock B controls all registers on the port B side. Each port, A and B, also
supports independent clock enables and asynchronous clear signals for
port A and B registers. Figure 2–24 shows a TriMatrix memory block in
independent clock mode.
2–44
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
(1)
(2)
Altera Corporation
July 2005
clockA
clkenA
wrenA
addressA[ ]
byteenaA[ ]
dataA[ ]
8
ENA
D
ENA
D
ENA
D
ENA
D
8 LAB Row Clocks
Q
Q
Q
Q
Write
Pulse
Generator
Q
Data Out
Write/Read
Enable
Address A
qA[ ]
B
Data In
qB[ ]
Q
D
ENA
Data Out
Write/Read
Enable
Address B
Byte Enable B
Memory Block
256 ´ 16 (2)
512 ´ 8
1,024 ´ 4
2,048 ´ 2
4,096 ´ 1
Byte Enable A
ENA
D
A
Data In
Write
Pulse
Generator
Q
Q
Q
Q
D
ENA
D
ENA
D
ENA
D
ENA
8
clockB
clkenB
wrenB
addressB[ ]
byteenaB[ ]
dataB[ ]
Stratix Architecture
Figure 2–24. Independent Clock Mode Notes (1), (2)
Notes to Figure 2–24
All registers shown have asynchronous clear ports.
Violating the setup or hold time on the address registers could corrupt the memory
contents. This applies to both read and write operations.
2–45
Stratix Device Handbook, Volume 1
TriMatrix Memory
Input/Output Clock Mode
Input/output clock mode can be implemented for both the true and
simple dual-port memory modes. On each of the two ports, A or B, one
clock controls all registers for inputs into the memory block: data input,
wren, and address. The other clock controls the block’s data output
registers. Each memory block port, A or B, also supports independent
clock enables and asynchronous clear signals for input and output
registers. Figures 2–25 and 2–26 show the memory block in input/output
clock mode.
2–46
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
(1)
(2)
Altera Corporation
July 2005
clockA
clkenA
wrenA
addressA[ ]
byteenaA[ ]
dataA[ ]
8
ENA
D
ENA
D
ENA
D
ENA
D
8 LAB Row Clocks
Q
Q
Q
Q
Write
Pulse
Generator
Q
Data Out
Write/Read
Enable
Address A
ENA
D
A
qA[ ]
Data In
B
qB[ ]
Q
D
ENA
Data Out
Write/Read
Enable
Address B
Byte Enable B
Memory Block
256 × 16 (2)
512 × 8
1,024 × 4
2,048 × 2
4,096 × 1
Byte Enable A
Data In
Write
Pulse
Generator
Q
Q
Q
Q
ENA
D
ENA
D
ENA
D
ENA
D
8
clockB
clkenB
wrenB
addressB[ ]
byteenaB[ ]
dataB[ ]
Stratix Architecture
Figure 2–25. Input/Output Clock Mode in True Dual-Port Mode Notes (1), (2)
Notes to Figure 2–25:
All registers shown have asynchronous clear ports.
Violating the setup or hold time on the address registers could corrupt the memory
contents. This applies to both read and write operations.
2–47
Stratix Device Handbook, Volume 1
TriMatrix Memory
Figure 2–26. Input/Output Clock Mode in Simple Dual-Port Mode Notes (1), (2)
8 LAB Row
Clocks
Memory Block
256 ´ 16
Data In
512 ´ 8
1,024 ´ 4
2,048 ´ 2
4,096 ´ 1
8
data[ ]
D
Q
ENA
address[ ]
D
Q
ENA
Read Address
Data Out
byteena[ ]
D
Q
ENA
Byte Enable
wraddress[ ]
D
Q
ENA
Write Address
D
Q
ENA
Read Enable
D
Q
ENA
To MultiTrack
Interconnect
rden
wren
outclken
inclken
wrclock
D
Q
ENA
Write
Pulse
Generator
Write Enable
rdclock
Notes to Figure 2–26:
(1)
(2)
All registers shown except the rden register have asynchronous clear ports.
Violating the setup or hold time on the address registers could corrupt the memory contents. This applies to both
read and write operations.
2–48
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Read/Write Clock Mode
The memory blocks implement read/write clock mode for simple dualport memory. You can use up to two clocks in this mode. The write clock
controls the block’s data inputs, wraddress, and wren. The read clock
controls the data output, rdaddress, and rden. The memory blocks
support independent clock enables for each clock and asynchronous clear
signals for the read- and write-side registers. Figure 2–27 shows a
memory block in read/write clock mode.
Altera Corporation
July 2005
2–49
Stratix Device Handbook, Volume 1
TriMatrix Memory
Figure 2–27. Read/Write Clock Mode in Simple Dual-Port Mode Notes (1), (2)
8 LAB Row
Clocks
Memory Block
256 × 16
512 × 8
1,024 × 4
Data In
2,048 × 2
4,096 × 1
8
data[ ]
D
Q
ENA
Data Out
address[ ]
D
Q
ENA
Read Address
wraddress[ ]
D
Q
ENA
Write Address
byteena[ ]
D
Q
ENA
Byte Enable
D
Q
ENA
Read Enable
D
Q
ENA
To MultiTrack
Interconnect
rden
wren
outclken
inclken
wrclock
D
Q
ENA
Write
Pulse
Generator
Write Enable
rdclock
Notes to Figure 2–27:
(1)
(2)
All registers shown except the rden register have asynchronous clear ports.
Violating the setup or hold time on the address registers could corrupt the memory contents. This applies to both
read and write operations.
2–50
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Single-Port Mode
The memory blocks also support single-port mode, used when
simultaneous reads and writes are not required. See Figure 2–28. A single
block in a memory block can support up to two single-port mode RAM
blocks in the M4K RAM blocks if each RAM block is less than or equal to
2K bits in size.
Figure 2–28. Single-Port Mode Note (1)
8 LAB Row
Clocks
RAM/ROM
256 × 16
512 × 8
1,024 × 4
Data In
2,048 × 2
4,096 × 1
8
data[ ]
D
Q
ENA
Data Out
address[ ]
D
Q
ENA
Address
D
Q
ENA
To MultiTrack
Interconnect
wren
Write Enable
outclken
inclken
inclock
D
Q
ENA
Write
Pulse
Generator
outclock
Note to Figure 2–28:
(1)
Violating the setup or hold time on the address registers could corrupt the memory contents. This applies to both
read and write operations.
Altera Corporation
July 2005
2–51
Stratix Device Handbook, Volume 1
Digital Signal Processing Block
Digital Signal
Processing
Block
The most commonly used DSP functions are finite impulse response (FIR)
filters, complex FIR filters, infinite impulse response (IIR) filters, fast
Fourier transform (FFT) functions, direct cosine transform (DCT)
functions, and correlators. All of these blocks have the same fundamental
building block: the multiplier. Additionally, some applications need
specialized operations such as multiply-add and multiply-accumulate
operations. Stratix devices provide DSP blocks to meet the arithmetic
requirements of these functions.
Each Stratix device has two columns of DSP blocks to efficiently
implement DSP functions faster than LE-based implementations. Larger
Stratix devices have more DSP blocks per column (see Table 2–13). Each
DSP block can be configured to support up to:
■
■
■
Eight 9 × 9-bit multipliers
Four 18 × 18-bit multipliers
One 36 × 36-bit multiplier
As indicated, the Stratix DSP block can support one 36 × 36-bit multiplier
in a single DSP block. This is true for any matched sign multiplications
(either unsigned by unsigned or signed by signed), but the capabilities for
dynamic and mixed sign multiplications are handled differently. The
following list provides the largest functions that can fit into a single DSP
block.
■
■
■
■
■
■
■
■
■
■
36 × 36-bit unsigned by unsigned multiplication
36 × 36-bit signed by signed multiplication
35 × 36-bit unsigned by signed multiplication
36 × 35-bit signed by unsigned multiplication
36 × 35-bit signed by dynamic sign multiplication
35 × 36-bit dynamic sign by signed multiplication
35 × 36-bit unsigned by dynamic sign multiplication
36 × 35-bit dynamic sign by unsigned multiplication
35 × 35-bit dynamic sign multiplication when the sign controls for
each operand are different
36 × 36-bit dynamic sign multiplication when the same sign control
is used for both operands
1
This list only shows functions that can fit into a single DSP block.
Multiple DSP blocks can support larger multiplication
functions.
Figure 2–29 shows one of the columns with surrounding LAB rows.
2–52
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–29. DSP Blocks Arranged in Columns
DSP Block
Column
8 LAB
Rows
Altera Corporation
July 2005
DSP Block
2–53
Stratix Device Handbook, Volume 1
Digital Signal Processing Block
Table 2–13 shows the number of DSP blocks in each Stratix device.
Table 2–13. DSP Blocks in Stratix Devices Notes (1), (2)
DSP Blocks
Total 9 × 9
Multipliers
Total 18 × 18
Multipliers
Total 36 × 36
Multipliers
EP1S10
6
48
24
6
EP1S20
10
80
40
10
EP1S25
10
80
40
10
Device
EP1S30
12
96
48
12
EP1S40
14
112
56
14
EP1S60
18
144
72
18
EP1S80
22
176
88
22
Notes to Table 2–13:
(1)
(2)
Each device has either the number of 9 × 9-, 18 × 18-, or 36 × 36-bit multipliers
shown. The total number of multipliers for each device is not the sum of all the
multipliers.
The number of supported multiply functions shown is based on signed/signed
or unsigned/unsigned implementations.
DSP block multipliers can optionally feed an adder/subtractor or
accumulator within the block depending on the configuration. This
makes routing to LEs easier, saves LE routing resources, and increases
performance, because all connections and blocks are within the DSP
block. Additionally, the DSP block input registers can efficiently
implement shift registers for FIR filter applications.
Figure 2–30 shows the top-level diagram of the DSP block configured for
18 × 18-bit multiplier mode. Figure 2–31 shows the 9 × 9-bit multiplier
configuration of the DSP block.
2–54
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–30. DSP Block Diagram for 18 × 18-Bit Configuration
Optional Serial Shift Register
Inputs from Previous
DSP Block
Multiplier Stage
D
Optional Stage Configurable
as Accumulator or Dynamic
Adder/Subtractor
Q
ENA
CLRN
D
D
ENA
CLRN
Q
Output Selection
Multiplexer
Q
ENA
CLRN
Adder/
Subtractor/
Accumulator
1
D
Q
ENA
CLRN
D
D
ENA
CLRN
Q
Q
ENA
CLRN
Summation
D
Q
ENA
CLRN
D
D
ENA
CLRN
Q
Q
Summation Stage
for Adding Four
Multipliers Together
Optional Output
Register Stage
ENA
CLRN
Adder/
Subtractor/
Accumulator
2
D
Optional Serial
Shift Register
Outputs to
Next DSP Block
in the Column
Q
ENA
CLRN
D
D
ENA
CLRN
Q
ENA
CLRN
Altera Corporation
July 2005
Q
Optional Pipeline
Register Stage
Optional Input Register
Stage with Parallel Input or
Shift Register Configuration
to MultiTrack
Interconnect
2–55
Stratix Device Handbook, Volume 1
Digital Signal Processing Block
Figure 2–31. DSP Block Diagram for 9 × 9-Bit Configuration
D
Q
ENA
CLRN
D
Q
ENA
D
Q
ENA
CLRN
Adder/
Subtractor/
1a
CLRN
D
Q
ENA
CLRN
D
Q
ENA
D
Q
ENA
CLRN
CLRN
Summation
D
Q
ENA
CLRN
D
Q
ENA
D
Q
ENA
CLRN
Adder/
Subtractor/
1b
CLRN
D
Q
ENA
CLRN
D
Q
ENA
D
Q
ENA
CLRN
Output
Selection
Multiplexer
CLRN
D
Q
ENA
CLRN
D
Q
ENA
D
Q
ENA
CLRN
D
Q
ENA
CLRN
Adder/
Subtractor/
2a
CLRN
D
Q
ENA
CLRN
D
Q
ENA
D
Q
ENA
CLRN
CLRN
Summation
D
Q
ENA
CLRN
D
Q
ENA
D
Q
ENA
CLRN
Adder/
Subtractor/
2b
CLRN
D
Q
ENA
CLRN
D
Q
ENA
D
Q
ENA
CLRN
CLRN
To MultiTrack
Interconnect
2–56
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
The DSP block consists of the following elements:
■
■
Multiplier block
Adder/output block
Multiplier Block
The DSP block multiplier block consists of the input registers, a
multiplier, and pipeline register for pipelining multiply-accumulate and
multiply-add/subtract functions as shown in Figure 2–32.
Figure 2–32. Multiplier Sub-Block within Stratix DSP Block
sign_a (1)
sign_b (1)
aclr[3..0]
clock[3..0]
ena[3..0]
shiftin A
shiftin B
D
Data A
Q
ENA
CLRN
D
ENA
Q
CLRN
D
Data B
Q
ENA
Result
to Adder
blocks
Optional
Multiply-Accumulate
and Multiply-Add
Pipeline
CLRN
shiftout B
shiftout A
Note to Figure 2–32:
(1)
These signals can be unregistered or registered once to match data path pipelines if required.
Altera Corporation
July 2005
2–57
Stratix Device Handbook, Volume 1
Digital Signal Processing Block
Input Registers
A bank of optional input registers is located at the input of each multiplier
and multiplicand inputs to the multiplier. When these registers are
configured for parallel data inputs, they are driven by regular routing
resources. You can use a clock signal, asynchronous clear signal, and a
clock enable signal to independently control each set of A and B inputs for
each multiplier in the DSP block. You select these control signals from a
set of four different clock[3..0], aclr[3..0], and ena[3..0]
signals that drive the entire DSP block.
You can also configure the input registers for a shift register application.
In this case, the input registers feed the multiplier and drive two
dedicated shift output lines: shiftoutA and shiftoutB. The shift
outputs of one multiplier block directly feed the adjacent multiplier block
in the same DSP block (or the next DSP block) as shown in Figure 2–33, to
form a shift register chain. This chain can terminate in any block, that is,
you can create any length of shift register chain up to 224 registers. You
can use the input shift registers for FIR filter applications. One set of shift
inputs can provide data for a filter, and the other are coefficients that are
optionally loaded in serial or parallel. When implementing 9 × 9- and
18 × 18-bit multipliers, you do not need to implement external shift
registers in LAB LEs. You implement all the filter circuitry within the DSP
block and its routing resources, saving LE and general routing resources
for general logic. External registers are needed for shift register inputs
when using 36 × 36-bit multipliers.
2–58
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–33. Multiplier Sub-Blocks Using Input Shift Register Connections
Note (1)
Data A
D
Q
ENA
A[n] × B[n]
CLRN
D
Data B
Q
D
ENA
Q
CLRN
ENA
CLRN
Data B
Data A
D
Q
ENA
A[n Ð 1] × B[n Ð 1]
CLRN
D
Q
D
ENA
Q
CLRN
ENA
CLRN
Data B
Data A
D
Q
ENA
A[n Ð 2] × B[n Ð 2]
CLRN
D
Q
D
ENA
Q
CLRN
ENA
CLRN
Note to Figure 2–33:
(1)
Altera Corporation
July 2005
Either Data A or Data B input can be set to a parallel input for constant coefficient
multiplication.
2–59
Stratix Device Handbook, Volume 1
Digital Signal Processing Block
Table 2–14 shows the summary of input register modes for the DSP block.
Table 2–14. Input Register Modes
Register Input Mode
9×9
18 × 18
36 × 36
Parallel input
v
v
v
Shift register input
v
v
Multiplier
The multiplier supports 9 × 9-, 18 × 18-, or 36 × 36-bit multiplication. Each
DSP block supports eight possible 9 × 9-bit or smaller multipliers. There
are four multiplier blocks available for multipliers larger than 9 × 9 bits
but smaller than 18 × 18 bits. There is one multiplier block available for
multipliers larger than 18 × 18 bits but smaller than or equal to 36 × 36
bits. The ability to have several small multipliers is useful in applications
such as video processing. Large multipliers greater than 18 × 18 bits are
useful for applications such as the mantissa multiplication of a singleprecision floating-point number.
The multiplier operands can be signed or unsigned numbers, where the
result is signed if either input is signed as shown in Table 2–15. The
sign_a and sign_b signals provide dynamic control of each operand’s
representation: a logic 1 indicates the operand is a signed number, a logic
0 indicates the operand is an unsigned number. These sign signals affect
all multipliers and adders within a single DSP block and you can register
them to match the data path pipeline. The multipliers are full precision
(that is, 18 bits for the 18-bit multiply, 36-bits for the 36-bit multiply, and
so on) regardless of whether sign_a or sign_b set the operands as
signed or unsigned numbers.
Table 2–15. Multiplier Signed Representation
2–60
Stratix Device Handbook, Volume 1
Data A
Data B
Result
Unsigned
Unsigned
Unsigned
Unsigned
Signed
Signed
Signed
Unsigned
Signed
Signed
Signed
Signed
Altera Corporation
July 2005
Stratix Architecture
Pipeline/Post Multiply Register
The output of 9 × 9- or 18 × 18-bit multipliers can optionally feed a register
to pipeline multiply-accumulate and multiply-add/subtract functions.
For 36 × 36-bit multipliers, this register will pipeline the multiplier
function.
Adder/Output Blocks
The result of the multiplier sub-blocks are sent to the adder/output block
which consist of an adder/subtractor/accumulator unit, summation unit,
output select multiplexer, and output registers. The results are used to
configure the adder/output block as a pure output, accumulator, a simple
two-multiplier adder, four-multiplier adder, or final stage of the 36-bit
multiplier. You can configure the adder/output block to use output
registers in any mode, and must use output registers for the accumulator.
The system cannot use adder/output blocks independently of the
multiplier. Figure 2–34 shows the adder and output stages.
Altera Corporation
July 2005
2–61
Stratix Device Handbook, Volume 1
Digital Signal Processing Block
Figure 2–34. Adder/Output Blocks Note (1)
Accumulator Feedback
accum_sload0 (2)
Result A
addnsub1 (2)
overflow0
Adder/
Subtractor/
Accumulator1
Output Selection
Multiplexer
Result B
signa (2)
Summation
Output
Register Block
signb (2)
Result C
addnsub3 (2)
Adder/
Subtractor/
Accumulator2
overflow1
Result D
accum_sload1 (2)
Accumulator Feedback
Notes to Figure 2–34:
(1)
(2)
Adder/output block shown in Figure 2–34 is in 18 × 18-bit mode. In 9 × 9-bit mode, there are four adder/subtractor
blocks and two summation blocks.
These signals are either not registered, registered once, or registered twice to match the data path pipeline.
2–62
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Adder/Subtractor/Accumulator
The adder/subtractor/accumulator is the first level of the adder/output
block and can be used as an accumulator or as an adder/subtractor.
Adder/Subtractor
Each adder/subtractor/accumulator block can perform addition or
subtraction using the addnsub independent control signal for each firstlevel adder in 18 × 18-bit mode. There are two addnsub[1..0] signals
available in a DSP block for any configuration. For 9 × 9-bit mode, one
addnsub[1..0] signal controls the top two one-level adders and
another addnsub[1..0] signal controls the bottom two one-level
adders. A high addnsub signal indicates addition, and a low signal
indicates subtraction. The addnsub control signal can be unregistered or
registered once or twice when feeding the adder blocks to match data
path pipelines.
The signa and signb signals serve the same function as the multiplier
block signa and signb signals. The only difference is that these signals
can be registered up to two times. These signals are tied to the same
signa and signb signals from the multiplier and must be connected to
the same clocks and control signals.
Accumulator
When configured for accumulation, the adder/output block output feeds
back to the accumulator as shown in Figure 2–34. The
accum_sload[1..0] signal synchronously loads the multiplier result
to the accumulator output. This signal can be unregistered or registered
once or twice. Additionally, the overflow signal indicates the
accumulator has overflowed or underflowed in accumulation mode. This
signal is always registered and must be externally latched in LEs if the
design requires a latched overflow signal.
Summation
The output of the adder/subtractor/accumulator block feeds to an
optional summation block. This block sums the outputs of the DSP block
multipliers. In 9 × 9-bit mode, there are two summation blocks providing
the sums of two sets of four 9 × 9-bit multipliers. In 18 × 18-bit mode, there
is one summation providing the sum of one set of four 18 × 18-bit
multipliers.
Altera Corporation
July 2005
2–63
Stratix Device Handbook, Volume 1
Digital Signal Processing Block
Output Selection Multiplexer
The outputs from the various elements of the adder/output block are
routed through an output selection multiplexer. Based on the DSP block
operational mode and user settings, the multiplexer selects whether the
output from the multiplier, the adder/subtractor/accumulator, or
summation block feeds to the output.
Output Registers
Optional output registers for the DSP block outputs are controlled by four
sets of control signals: clock[3..0], aclr[3..0], and ena[3..0].
Output registers can be used in any mode.
Modes of Operation
The adder, subtractor, and accumulate functions of a DSP block have four
modes of operation:
■
■
■
■
Simple multiplier
Multiply-accumulator
Two-multipliers adder
Four-multipliers adder
1
Each DSP block can only support one mode. Mixed modes in the
same DSP block is not supported.
Simple Multiplier Mode
In simple multiplier mode, the DSP block drives the multiplier sub-block
result directly to the output with or without an output register. Up to four
18 × 18-bit multipliers or eight 9 × 9-bit multipliers can drive their results
directly out of one DSP block. See Figure 2–35.
2–64
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–35. Simple Multiplier Mode
signa (1)
signb (1)
aclr
clock
ena
shiftin A
shiftin B
D
Data A
Q
Data Out
ENA
CLRN
D
ENA
Q
D
ENA
Q
CLRN
CLRN
D
Data B
Q
ENA
CLRN
shiftout B
shiftout A
Note to Figure 2–35:
(1)
These signals are not registered or registered once to match the data path pipeline.
DSP blocks can also implement one 36 × 36-bit multiplier in multiplier
mode. DSP blocks use four 18 × 18-bit multipliers combined with
dedicated adder and internal shift circuitry to achieve 36-bit
multiplication. The input shift register feature is not available for the
36 × 36-bit multiplier. In 36 × 36-bit mode, the device can use the register
that is normally a multiplier-result-output register as a pipeline stage for
the 36 × 36-bit multiplier. Figure 2–36 shows the 36 × 36-bit multiply
mode.
Altera Corporation
July 2005
2–65
Stratix Device Handbook, Volume 1
Digital Signal Processing Block
Figure 2–36. 36 × 36 Multiply Mode
signa (1)
signb (1)
aclr
clock
ena
A[17..0]
D
Q
ENA
CLRN
D
Q
ENA
CLRN
B[17..0]
D
Q
ENA
CLRN
A[35..18]
D
Q
CLRN
D
Q
ENA
36 × 36
Multiplier
Adder
CLRN
B[35..18]
D
Data Out
D
Q
ENA
ENA
CLRN
Q
signa (2)
ENA
signb (2)
CLRN
A[35..18]
D
Q
ENA
CLRN
D
Q
ENA
CLRN
B[17..0]
D
Q
ENA
CLRN
A[17..0]
D
Q
ENA
CLRN
D
Q
ENA
CLRN
B[35..18]
D
Q
ENA
CLRN
Notes to Figure 2–36:
(1)
(2)
These signals are not registered or registered once to match the pipeline.
These signals are not registered, registered once, or registered twice for latency to match the pipeline.
2–66
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Multiply-Accumulator Mode
In multiply-accumulator mode (see Figure 2–37), the DSP block drives
multiplied results to the adder/subtractor/accumulator block configured
as an accumulator. You can implement one or two multiply-accumulators
up to 18 × 18 bits in one DSP block. The first and third multiplier subblocks are unused in this mode, because only one multiplier can feed one
of two accumulators. The multiply-accumulator output can be up to 52
bits—a maximum of a 36-bit result with 16 bits of accumulation. The
accum_sload and overflow signals are only available in this mode.
The addnsub signal can set the accumulator for decimation and the
overflow signal indicates underflow condition.
Figure 2–37. Multiply-Accumulate Mode
signa (1)
signb (1)
aclr
clock
ena
Shiftin A
Shiftin B
D
Data A
Q
ENA
CLRN
D
Q
ENA
Accumulator
D
Q
ENA
Data Out
CLRN
CLRN
D
Data B
Q
overflow
ENA
CLRN
Shiftout B Shiftout A
addnsub (2)
signa (2)
signb (2)
accum_sload (2)
Notes to Figure 2–37:
(1)
(2)
These signals are not registered or registered once to match the data path pipeline.
These signals are not registered, registered once, or registered twice for latency to match the data path pipeline.
Two-Multipliers Adder Mode
The two-multipliers adder mode uses the adder/subtractor/accumulator
block to add or subtract the outputs of the multiplier block, which is
useful for applications such as FFT functions and complex FIR filters. A
Altera Corporation
July 2005
2–67
Stratix Device Handbook, Volume 1
Digital Signal Processing Block
single DSP block can implement two sums or differences from two
18 × 18-bit multipliers each or four sums or differences from two 9 × 9-bit
multipliers each.
You can use the two-multipliers adder mode for complex multiplications,
which are written as:
(a + jb) × (c + jd) = [(a × c) – (b × d)] + j × [(a × d) + (b × c)]
The two-multipliers adder mode allows a single DSP block to calculate
the real part [(a × c) – (b × d)] using one subtractor and the imaginary part
[(a × d) + (b × c)] using one adder, for data widths up to 18 bits. Two
complex multiplications are possible for data widths up to 9 bits using
four adder/subtractor/accumulator blocks. Figure 2–38 shows an 18-bit
two-multipliers adder.
Figure 2–38. Two-Multipliers Adder Mode Implementing Complex Multiply
18
DSP Block
18
A
36
18
18
C
18
37
(A × C) − (B × D)
(Real Part)
Subtractor
18
B
36
18
18
D
18
A
36
18
D
37
Adder
18
B
(A × D) + (B × C)
(Imaginary Part)
36
18
C
Four-Multipliers Adder Mode
In the four-multipliers adder mode, the DSP block adds the results of two
first -stage adder/subtractor blocks. One sum of four 18 × 18-bit
multipliers or two different sums of two sets of four 9 × 9-bit multipliers
can be implemented in a single DSP block. The product width for each
multiplier must be the same size. The four-multipliers adder mode is
useful for FIR filter applications. Figure 2–39 shows the four multipliers
adder mode.
2–68
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–39. Four-Multipliers Adder Mode
signa (1)
signb (1)
aclr
clock
ena
shiftin A
shiftin B
D
Data A
Q
ENA
CLRN
D
ENA
Q
Adder/Subtractor
CLRN
D
Data B
Q
ENA
CLRN
D
Data A
Q
D
ENA
ENA
CLRN
D
ENA
Q
CLRN
D
Data B
Q
addnsub1 (2)
signa (2)
signb (2)
Q
Data Out
Summation
CLRN
addnsub3 (2)
ENA
CLRN
D
Data A
Q
ENA
CLRN
D
ENA
Q
Adder/Subtractor
CLRN
D
Data B
Q
ENA
CLRN
D
Data A
Q
ENA
CLRN
D
ENA
Q
CLRN
D
Data B
Q
ENA
CLRN
shiftout B
shiftout A
Notes to Figure 2–39:
(1)
(2)
These signals are not registered or registered once to match the data path pipeline.
These signals are not registered, registered once, or registered twice for latency to match the data path pipeline.
Altera Corporation
July 2005
2–69
Stratix Device Handbook, Volume 1
Digital Signal Processing Block
For FIR filters, the DSP block combines the four-multipliers adder mode
with the shift register inputs. One set of shift inputs contains the filter
data, while the other holds the coefficients loaded in serial or parallel. The
input shift register eliminates the need for shift registers external to the
DSP block (i.e., implemented in LEs). This architecture simplifies filter
design since the DSP block implements all of the filter circuitry.
One DSP block can implement an entire 18-bit FIR filter with up to four
taps. For FIR filters larger than four taps, DSP blocks can be cascaded with
additional adder stages implemented in LEs.
Table 2–16 shows the different number of multipliers possible in each
DSP block mode according to size. These modes allow the DSP blocks to
implement numerous applications for DSP including FFTs, complex FIR,
FIR, and 2D FIR filters, equalizers, IIR, correlators, matrix multiplication
and many other functions.
Table 2–16. Multiplier Size & Configurations per DSP block
DSP Block Mode
9×9
18 × 18
36 × 36 (1)
Multiplier
Eight multipliers with
eight product outputs
Four multipliers with four
product outputs
One multiplier with one
product output
Multiply-accumulator
Two multiply and
accumulate (52 bits)
Two multiply and
accumulate (52 bits)
–
Two-multipliers adder
Four sums of two
multiplier products each
Two sums of two
multiplier products each
–
Four-multipliers adder
Two sums of four
multiplier products each
One sum of four multiplier
products each
–
Note to Table 2–16:
(1)
The number of supported multiply functions shown is based on signed/signed or unsigned/unsigned
implementations.
DSP Block Interface
Stratix device DSP block outputs can cascade down within the same DSP
block column. Dedicated connections between DSP blocks provide fast
connections between the shift register inputs to cascade the shift register
chains. You can cascade DSP blocks for 9 × 9- or 18 × 18-bit FIR filters
larger than four taps, with additional adder stages implemented in LEs.
If the DSP block is configured as 36 × 36 bits, the adder, subtractor, or
accumulator stages are implemented in LEs. Each DSP block can route the
shift register chain out of the block to cascade two full columns of DSP
blocks.
2–70
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
The DSP block is divided into eight block units that interface with eight
LAB rows on the left and right. Each block unit can be considered half of
an 18 × 18-bit multiplier sub-block with 18 inputs and 18 outputs. A local
interconnect region is associated with each DSP block. Like an LAB, this
interconnect region can be fed with 10 direct link interconnects from the
LAB to the left or right of the DSP block in the same row. All row and
column routing resources can access the DSP block’s local interconnect
region. The outputs also work similarly to LAB outputs as well. Nine
outputs from the DSP block can drive to the left LAB through direct link
interconnects and nine can drive to the right LAB though direct link
interconnects. All 18 outputs can drive to all types of row and column
routing. Outputs can drive right- or left-column routing. Figures 2–40
and 2–41 show the DSP block interfaces to LAB rows.
Figure 2–40. DSP Block Interconnect Interface
DSP Block
MultiTrack
Interconnect
OA[17..0]
MultiTrack
Interconnect
A1[17..0]
OB[17..0]
B1[17..0]
OC[17..0]
A2[17..0]
OD[17..0]
B2[17..0]
OE[17..0]
A3[17..0]
OF[17..0]
B3[17..0]
OG[17..0]
A4[17..0]
OH[17..0]
B4[17..0]
Altera Corporation
July 2005
2–71
Stratix Device Handbook, Volume 1
Digital Signal Processing Block
Figure 2–41. DSP Block Interface to Interconnect
C4 and C8
Interconnects
Direct Link Interconnect
from Adjacent LAB
R4 and R8 Interconnects
Nine Direct Link Outputs
to Adjacent LABs
Direct Link Interconnect
from Adjacent LAB
18
DSP Block
Row Structure
LAB
10
LAB
9
9
10
3
Control
18
18
[17..0]
[17..0]
Row Interface
Block
DSP Block to
LAB Row Interface
Block Interconnect Region
18 Inputs per Row
18 Outputs per Row
A bus of 18 control signals feeds the entire DSP block. These signals
include clock[0..3] clocks, aclr[0..3] asynchronous clears,
ena[1..4] clock enables, signa, signb signed/unsigned control
signals, addnsub1 and addnsub3 addition and subtraction control
signals, and accum_sload[0..1] accumulator synchronous loads. The
2–72
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
clock signals are routed from LAB row clocks and are generated from
specific LAB rows at the DSP block interface. The LAB row source for
control signals, data inputs, and outputs is shown in Table 2–17.
Table 2–17. DSP Block Signal Sources & Destinations
LAB Row at
Interface
PLLs & Clock
Networks
Control Signals
Generated
Data Inputs
Data Outputs
1
signa
A1[17..0]
OA[17..0]
2
aclr0
accum_sload0
B1[17..0]
OB[17..0]
3
addnsub1
clock0
ena0
A2[17..0]
OC[17..0]
4
aclr1
clock1
ena1
B2[17..0]
OD[17..0]
5
aclr2
clock2
ena2
A3[17..0]
OE[17..0]
6
sign_b
clock3
ena3
B3[17..0]
OF[17..0]
7
clear3
accum_sload1
A4[17..0]
OG[17..0]
8
addnsub3
B4[17..0]
OH[17..0]
Stratix devices provide a hierarchical clock structure and multiple PLLs
with advanced features. The large number of clocking resources in
combination with the clock synthesis precision provided by enhanced
and fast PLLs provides a complete clock management solution.
Global & Hierarchical Clocking
Stratix devices provide 16 dedicated global clock networks, 16 regional
clock networks (four per device quadrant), and 8 dedicated fast regional
clock networks (for EP1S10, EP1S20, and EP1S25 devices), and
16 dedicated fast regional clock networks (for EP1S30 EP1S40, and
EP1S60, and EP1S80 devices). These clocks are organized into a
hierarchical clock structure that allows for up to 22 clocks per device
region with low skew and delay. This hierarchical clocking scheme
provides up to 48 unique clock domains within Stratix devices.
Altera Corporation
July 2005
2–73
Stratix Device Handbook, Volume 1
PLLs & Clock Networks
There are 16 dedicated clock pins (CLK[15..0]) to drive either the global
or regional clock networks. Four clock pins drive each side of the device,
as shown in Figure 2–42. Enhanced and fast PLL outputs can also drive
the global and regional clock networks.
Global Clock Network
These clocks drive throughout the entire device, feeding all device
quadrants. The global clock networks can be used as clock sources for all
resources within the device—IOEs, LEs, DSP blocks, and all memory
blocks. These resources can also be used for control signals, such as clock
enables and synchronous or asynchronous clears fed from the external
pin. The global clock networks can also be driven by internal logic for
internally generated global clocks and asynchronous clears, clock
enables, or other control signals with large fanout. Figure 2–42 shows the
16 dedicated CLK pins driving global clock networks.
2–74
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–42. Global Clocking Note (1)
CLK[15..12]
Global Clock [15..0]
CLK[3..0]
Global Clock [15..0]
CLK[11..8]
CLK[7..4]
Note to Figure 2–42:
(1)
The corner fast PLLs can also be driven through the global or regional clock
networks. The global or regional clock input to the fast PLL can be driven by an
output from another PLL, a pin-driven global or regional clock, or internallygenerated global signals.
Regional Clock Network
There are four regional clock networks within each quadrant of the Stratix
device that are driven by the same dedicated CLK[15..0] input pins or
from PLL outputs. From a top view of the silicon, RCLK[0..3] are in the
top left quadrant, RCLK[8..11] are in the top-right quadrant,
RCLK[4..7] are in the bottom-left quadrant, and RCLK[12..15] are in
the bottom-right quadrant. The regional clock networks only pertain to
the quadrant they drive into. The regional clock networks provide the
lowest clock delay and skew for logic contained within a single quadrant.
RCLK cannot be driven by internal logic. The CLK clock pins
symmetrically drive the RCLK networks within a particular quadrant, as
shown in Figure 2–43. See Figures 2–50 and 2–51 for RCLK connections
from PLLs and CLK pins.
Altera Corporation
July 2005
2–75
Stratix Device Handbook, Volume 1
PLLs & Clock Networks
Figure 2–43. Regional Clocks
RCLK[2..3]
RCLK[11..10]
CLK[15..12]
RCLK[9..8]
RCLK[1..0]
CLK[3..0]
CLK[11..8]
RCLK[14..15]
RCLK[4..5]
CLK[7..4]
Regional Clocks Only Drive a Device
Quadrant from Specified CLK Pins or
PLLs within that Quadrant
RCLK[6..7]
RCLK[12..13]
Fast Regional Clock Network
In EP1S25, EP1S20, and EP1S10 devices, there are two fast regional clock
networks, FCLK[1..0], within each quadrant, fed by input pins that can
connect to fast regional clock networks (see Figure 2–44). In EP1S30 and
larger devices, there are two fast regional clock networks within each
half-quadrant (see Figure 2–45). Dual-purpose FCLK pins drive the fast
clock networks. All devices have eight FCLK pins to drive fast regional
clock networks. Any I/O pin can drive a clock or control signal onto any
fast regional clock network with the addition of a delay. This signal is
driven via the I/O interconnect. The fast regional clock networks can also
be driven from internal logic elements.
2–76
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–44. EP1S25, EP1S20 & EP1S10 Device Fast Clock Pin Connections to
Fast Regional Clocks
FCLK[1..0]
FCLK[7..6]
2
(1), (2)
2
(1), (2)
2
2
FCLK[1..0]
FCLK[1..0]
FCLK[1..0]
FCLK[1..0]
2
(1), (2)
2
(1), (2)
2
FCLK[3..2]
2
FCLK[5..4]
Notes to Figure 2–44:
(1)
(2)
Altera Corporation
July 2005
This is a set of two multiplexers.
In addition to the FCLK pin inputs, there is also an input from the I/O interconnect.
2–77
Stratix Device Handbook, Volume 1
PLLs & Clock Networks
Figure 2–45. EP1S30 Device Fast Regional Clock Pin Connections to Fast
Regional Clocks
FCLK1
FCLK0
(1), (2)
FCLK7
FCLK6
(1), (2)
(1), (2)
(1), (2)
fclk[1..0]
(1), (2)
(1), (2)
(1), (2)
FCLK3
FCLK2
(1), (2)
FCLK5
FCLK4
Notes to Figure 2–45:
(1)
(2)
This is a set of two multiplexers.
In addition to the FCLK pin inputs, there is also an input from the I/O interconnect.
Combined Resources
Within each region, there are 22 distinct dedicated clocking resources
consisting of 16 global clock lines, four regional clock lines, and two fast
regional clock lines. Multiplexers are used with these clocks to form eight
bit busses to drive LAB row clocks, column IOE clocks, or row IOE clocks.
Another multiplexer is used at the LAB level to select two of the eight row
clocks to feed the LE registers within the LAB. See Figure 2–46.
2–78
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–46. Regional Clock Bus
Clocks Available
to a Quadrant
or Half-Quadrant
Vertical I/O Cell
IO_CLK[7..0]
Global Clock Network [15..0]
Regional Clock Network [3..0]
Clock [21..0]
Lab Row Clock [7..0]
Fast Regional Clock Network [1..0]
Horizontal I/O
Cell IO_CLK[7..0]
IOE clocks have horizontal and vertical block regions that are clocked by
eight I/O clock signals chosen from the 22 quadrant or half-quadrant
clock resources. Figures 2–47 and 2–48 show the quadrant and halfquadrant relationship to the I/O clock regions, respectively. The vertical
regions (column pins) have less clock delay than the horizontal regions
(row pins).
Altera Corporation
July 2005
2–79
Stratix Device Handbook, Volume 1
PLLs & Clock Networks
Figure 2–47. EP1S10, EP1S20 & EP1S25 Device I/O Clock Groups
IO_CLKA[7..0]
IO_CLKB[7..0]
8
8
I/O Clock Regions
8
22 Clocks in
the Quadrant
22 Clocks in
the Quadrant
IO_CLKH[7..0]
IO_CLKC[7..0]
8
8
IO_CLKG[7..0]
IO_CLKD[7..0]
22 Clocks in
the Quadrant
22 Clocks in
the Quadrant
8
8
8
IO_CLKF[7..0]
2–80
Stratix Device Handbook, Volume 1
IO_CLKE[7..0]
Altera Corporation
July 2005
Stratix Architecture
Figure 2–48. EP1S30, EP1S40, EP1S60, EP1S80 Device I/O Clock Groups
IO_CLKA[7:0]
IO_CLKB[7:0]
8
IO_CLKC[7:0]
8
IO_CLKD[7:0]
8
8
I/O Clock Regions
8
8
IO_CLKE[7:0]
IO_CLKP[7:0]
22 Clocks in the
Half-Quadrant
22 Clocks in the
Half-Quadrant
22 Clocks in the
Half-Quadrant
22 Clocks in the
Half-Quadrant
8
8
IO_CLKF[7:0]
IO_CLKO[7:0]
8
8
IO_CLKN[7:0]
IO_CLKG[7:0]
22 Clocks in the
Half-Quadrant
22 Clocks in the
Half-Quadrant
22 Clocks in the
Half-Quadrant
22 Clocks in the
Half-Quadrant
8
8
IO_CLKH[7:0]
IO_CLKM[7:0]
8
8
IO_CLKL[7:0]
8
IO_CLKK[7:0]
8
IO_CLKJ[7:0]
IO_CLKI[7:0]
You can use the Quartus II software to control whether a clock input pin
is either global, regional, or fast regional. The Quartus II software
automatically selects the clocking resources if not specified.
Enhanced & Fast PLLs
Stratix devices provide robust clock management and synthesis using up
to four enhanced PLLs and eight fast PLLs. These PLLs increase
performance and provide advanced clock interfacing and clockfrequency synthesis. With features such as clock switchover, spread
spectrum clocking, programmable bandwidth, phase and delay control,
and PLL reconfiguration, the Stratix device’s enhanced PLLs provide you
with complete control of your clocks and system timing. The fast PLLs
Altera Corporation
July 2005
2–81
Stratix Device Handbook, Volume 1
PLLs & Clock Networks
provide general purpose clocking with multiplication and phase shifting
as well as high-speed outputs for high-speed differential I/O support.
Enhanced and fast PLLs work together with the Stratix high-speed I/O
and advanced clock architecture to provide significant improvements in
system performance and bandwidth.
The Quartus II software enables the PLLs and their features without
requiring any external devices. Table 2–18 shows the PLLs available for
each Stratix device.
Table 2–18. Stratix Device PLL Availability
Fast PLLs
Enhanced PLLs
Device
1
2
3
4
EP1S10
v
v
v
EP1S20
v
v
EP1S25
v
v
EP1S30
v
v
5(1)
6(1)
v
v
v
v
v
v
v
v
v
v
7
8
9
10
v
v
v
v (3)
v (3)
v (3)
v (3)
v
v
v (3)
v (3)
v (3)
v
v
11(2)
12(2)
EP1S40
v
v
v
v
v (3)
EP1S60
v
v
v
v
v
v
v
v
v
v
v
v
EP1S80
v
v
v
v
v
v
v
v
v
v
v
v
v(3) v(3)
Notes to Table 2–18:
(1)
(2)
(3)
PLLs 5 and 6 each have eight single-ended outputs or four differential outputs.
PLLs 11 and 12 each have one single-ended output.
EP1S30 and EP1S40 devices do not support these PLLs in the 780-pin FineLine BGA® package.
2–82
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Table 2–19 shows the enhanced PLL and fast PLL features in Stratix
devices.
Table 2–19. Stratix PLL Features
Feature
Enhanced PLL
Fast PLL
Clock multiplication and division
m/(n × post-scale counter) (1)
m/(post-scale counter) (2)
Phase shift
Down to 156.25-ps increments (3), (4)
Down to 125-ps increments (3), (4)
Delay shift
250-ps increments for ±3 ns
Clock switchover
v
PLL reconfiguration
v
Programmable bandwidth
v
Spread spectrum clocking
v
Programmable duty cycle
v
v
Number of internal clock outputs
6
3 (5)
Number of external clock outputs
Four differential/eight singled-ended
or one single-ended (6)
(7)
Number of feedback clock inputs
2 (8)
Notes to Table 2–19:
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
For enhanced PLLs, m, n, range from 1 to 512 and post-scale counters g, l, e range from 1 to 1024 with 50% duty
cycle. With a non-50% duty cycle the post-scale counters g, l, e range from 1 to 512.
For fast PLLs, m and post-scale counters range from 1 to 32.
The smallest phase shift is determined by the voltage controlled oscillator (VCO) period divided by 8.
For degree increments, Stratix devices can shift all output frequencies in increments of at least 45° . Smaller degree
increments are possible depending on the frequency and divide parameters.
PLLs 7, 8, 9, and 10 have two output ports per PLL. PLLs 1, 2, 3, and 4 have three output ports per PLL.
Every Stratix device has two enhanced PLLs (PLLs 5 and 6) with either eight single-ended outputs or four
differential outputs each. Two additional enhanced PLLs (PLLs 11 and 12) in EP1S80, EP1S60, and EP1S40 devices
each have one single-ended output. Devices in the 780 pin FineLine BGA packages do not support PLLs 11 and 12.
Fast PLLs can drive to any I/O pin as an external clock. For high-speed differential I/O pins, the device uses a data
channel to generate txclkout.
Every Stratix device has two enhanced PLLs with one single-ended or differential external feedback input per PLL.
Altera Corporation
July 2005
2–83
Stratix Device Handbook, Volume 1
PLLs & Clock Networks
Figure 2–49 shows a top-level diagram of the Stratix device and PLL
floorplan.
Figure 2–49. PLL Locations
CLK[15..12]
5
11
FPLL7CLK
7
10
FPLL10CLK
CLK[3..0]
1
2
4
3
CLK[8..11]
8
9
FPLL9CLK
PLLs
FPLL8CLK
6
12
CLK[7..4]
2–84
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–50 shows the global and regional clocking from the PLL outputs
and the CLK pins.
Figure 2–50. Global & Regional Clock Connections from Side Pins & Fast PLL Outputs Note (1), (2)
RCLK1
RCLK0
FPLL7CLK
G1
G0
G3
G2
G8
G9
G10
G11
RCLK9
RCLK8
l0
l0
PLL 7 l1
CLK0
CLK1
l1 PLL 10
g0
g0
l0
l0
PLL 1 l1
l1 PLL 4
g0
CLK2
CLK3
CLK10
CLK11
g0
l02
PLL 2 l1
g0
2l0
l1 PLL 3
g0
l0
FPLL8CLK
FPLL10CLK
CLK8
CLK9
l0
PLL 8 l1
g0
l1 PLL 9
g0
RCLK4
RCLK5
Regional
Clocks
FPLL9CLK
RCLK14
Global
Clocks
RCLK15
Regional
Clocks
Notes to Figure 2–50:
(1)
(2)
PLLs 1 to 4 and 7 to 10 are fast PLLs. PLLs 5, 6, 11, and 12 are enhanced PLLs.
The global or regional clocks in a fast PLL’s quadrant can drive the fast PLL input. A pin or other PLL must drive
the global or regional source. The source cannot be driven by internally generated logic before driving the fast PLL.
Figure 2–51 shows the global and regional clocking from enhanced PLL
outputs and top CLK pins.
Altera Corporation
July 2005
2–85
Stratix Device Handbook, Volume 1
PLLs & Clock Networks
Figure 2–51. Global & Regional Clock Connections from Top Clock Pins & Enhanced PLL Outputs Note (1)
PLL5_OUT[3..0] CLK14 (1)
PLL5_FB
CLK15 (2)
CLK12 (1)
CLK13 (2)
E[0..3]
PLL 5
PLL 11
L0 L1 G0 G1 G2 G3
G0 G1 G2 G3 L0 L1
PLL11_OUT
RCLK10
RCLK11
Regional
Clocks
RCLK2
RCLK3
G12
G13
G14
G15
Global
Clocks
Regional
Clocks
G4
G5
G6
G7
RCLK6
RCLK7
RCLK12
RCLK13
PLL12_OUT
L0 L1 G0 G1 G2 G3
G0 G1 G2 G3 L0 L1
PLL 6
PLL6_OUT[3..0]
PLL 12
PLL6_FB
CLK4 (1)
CLK6 (1)
CLK7 (2)
CLK5 (2)
Notes to Figure 2–51:
(1)
(2)
(3)
(4)
PLLs 1 to 4 and 7 to 10 are fast PLLs. PLLs 5, 6, 11, and 12 are enhanced PLLs.
CLK4, CLK6, CLK12, and CLK14 feed the corresponding PLL’s inclk0 port.
CLK5, CLK7, CLK13, and CLK15 feed the corresponding PLL’s inclk1 port.
The EP1S40 device in the 780-pin FineLine BGA package does not support PLLs 11 and 12.
2–86
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Enhanced PLLs
Stratix devices contain up to four enhanced PLLs with advanced clock
management features. Figure 2–52 shows a diagram of the enhanced PLL.
Figure 2–52. Stratix Enhanced PLL
Programmable
Time Delay on
Each PLL Port
Post-Scale
Counters
VCO Phase Selection
Selectable at Each
PLL Output Port
From Adjacent PLL
/l0
Δt
/l1
Δt
Regional
Clocks
Clock
Switch-Over
Circuitry
Spread
Spectrum
Phase Frequency
Detector
INCLK0
4
Δt
/n
PFD
Charge
Pump
8
Loop
Filter
VCO
INCLK1
(1)
FBIN
Δt
/m
/g0
Δt
/g1
Δt
/g2
Δt
/g3
Δt
Global
Clocks
I/O buffers (2)
To I/O buffers or general
routing
Lock Detect
& Filter
VCO Phase Selection
Affecting All Outputs
/e0
Δt
/e1
Δt
/e2
Δt
/e3
Δt
4
I/O Buffers (3)
Notes to Figure 2–52:
(1)
(2)
(3)
(4)
External feedback is available in PLLs 5 and 6.
This single-ended external output is available from the g0 counter for PLLs 11 and 12.
These four counters and external outputs are available in PLLs 5 and 6.
This connection is only available on EP1S40 and larger Stratix devices. For example, PLLs 5 and 11 are adjacent and
PLLs 6 and 12 are adjacent. The EP1S40 device in the 780-pin FineLine BGA package does not support PLLs 11
and 12.
Altera Corporation
July 2005
2–87
Stratix Device Handbook, Volume 1
PLLs & Clock Networks
Clock Multiplication & Division
Each Stratix device enhanced PLL provides clock synthesis for PLL
output ports using m/(n × post-scale counter) scaling factors. The input
clock is divided by a pre-scale divider, n, and is then multiplied by the m
feedback factor. The control loop drives the VCO to match fIN × (m/n).
Each output port has a unique post-scale counter that divides down the
high-frequency VCO. For multiple PLL outputs with different
frequencies, the VCO is set to the least common multiple of the output
frequencies that meets its frequency specifications. Then, the post-scale
dividers scale down the output frequency for each output port. For
example, if output frequencies required from one PLL are 33 and 66 MHz,
set the VCO to 330 MHz (the least common multiple in the VCO’s range).
There is one pre-scale counter, n, and one multiply counter, m, per PLL,
with a range of 1 to 512 on each. There are two post-scale counters (l) for
regional clock output ports, four counters (g) for global clock output
ports, and up to four counters (e) for external clock outputs, all ranging
from 1 to 1024 with a 50% duty cycle setting. The post-scale counters
range from 1 to 512 with any non-50% duty cycle setting. The Quartus II
software automatically chooses the appropriate scaling factors according
to the input frequency, multiplication, and division values entered.
Clock Switchover
To effectively develop high-reliability network systems, clocking schemes
must support multiple clocks to provide redundancy. For this reason,
Stratix device enhanced PLLs support a flexible clock switchover
capability. Figure 2–53 shows a block diagram of the switchover
circuit.The switchover circuit is configurable, so you can define how to
implement it. Clock-sense circuitry automatically switches from the
primary to secondary clock for PLL reference when the primary clock
signal is not present.
2–88
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–53. Clock Switchover Circuitry
CLK0_BAD
CLK1_BAD
Active Clock
SMCLKSW
Clock
Sense
Switch-Over
State Machine
CLKLOSS
CLKSWITCH
Δt
INCLK0
MUXOUT
INCLK1
n Counter
PFD
FBCLK
Enhanced PLL
There are two possible ways to use the clock switchover feature.
■
■
Altera Corporation
July 2005
Use automatic switchover circuitry for switching between inputs of
the same frequency. For example, in applications that require a
redundant clock with the same frequency as the primary clock, the
switchover state machine generates a signal that controls the
multiplexer select input on the bottom of Figure 2–53. In this case, the
secondary clock becomes the reference clock for the PLL.
Use the clkswitch input for user- or system-controlled switch
conditions. This is possible for same-frequency switchover or to
switch between inputs of different frequencies. For example, if
inclk0 is 66 MHz and inclk1 is 100 MHz, you must control the
switchover because the automatic clock-sense circuitry cannot
monitor primary and secondary clock frequencies with a frequency
difference of more than ±20%. This feature is useful when clock
sources can originate from multiple cards on the backplane,
requiring a system-controlled switchover between frequencies of
operation. You can use clkswitch together with the lock signal to
trigger the switch from a clock that is running but becomes unstable
and cannot be locked onto.
2–89
Stratix Device Handbook, Volume 1
PLLs & Clock Networks
During switchover, the PLL VCO continues to run and will either slow
down or speed up, generating frequency drift on the PLL outputs. The
clock switchover transitions without any glitches. After the switch, there
is a finite resynchronization period to lock onto new clock as the VCO
ramps up. The exact amount of time it takes for the PLL to relock relates
to the PLL configuration and may be adjusted by using the
programmable bandwidth feature of the PLL. The specification for the
maximum time to relock is 100 µs.
f
For more information on clock switchover, see AN 313, Implementing
Clock Switchover in Stratix & Stratix GX Devices.
PLL Reconfiguration
The PLL reconfiguration feature enables system logic to change Stratix
device enhanced PLL counters and delay elements without reloading a
Programmer Object File (.pof). This provides considerable flexibility for
frequency synthesis, allowing real-time PLL frequency and output clock
delay variation. You can sweep the PLL output frequencies and clock
delay in prototype environments. The PLL reconfiguration feature can
also dynamically or intelligently control system clock speeds or tCO
delays in end systems.
Clock delay elements at each PLL output port implement variable delay.
Figure 2–54 shows a diagram of the overall dynamic PLL control feature
for the counters and the clock delay elements. The configuration time is
less than 20 μs for the enhanced PLL using a input shift clock rate of
22 MHz. The charge pump, loop filter components, and phase shifting
using VCO phase taps cannot be dynamically adjusted.
2–90
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–54. Dynamically Programmable Counters & Delays in Stratix Device Enhanced PLLs
Counters and Clock
Delay Settings are
Programmable
fREF
Δt
÷n
All Output Counters and
Clock Delay Settings can
be Programmed Dynamically
PFD
Charge
Pump
Loop
Filter
VCO
÷g
Δt
÷l
Δt
÷e
Δt
scandata
scanclk
÷m
Δt
scanaclr
PLL reconfiguration data is shifted into serial registers from the logic
array or external devices. The PLL input shift data uses a reference input
shift clock. Once the last bit of the serial chain is clocked in, the register
chain is synchronously loaded into the PLL configuration bits. The shift
circuitry also provides an asynchronous clear for the serial registers.
f
For more information on PLL reconfiguration, see AN 282: Implementing
PLL Reconfiguration in Stratix & Stratix GX Devices.
Programmable Bandwidth
You have advanced control of the PLL bandwidth using the
programmable control of the PLL loop characteristics, including loop
filter and charge pump. The PLL’s bandwidth is a measure of its ability to
track the input clock and jitter. A high-bandwidth PLL can quickly lock
onto a reference clock and react to any changes in the clock. It also will
allow a wide band of input jitter spectrum to pass to the output. A lowbandwidth PLL will take longer to lock, but it will attenuate all highfrequency jitter components. The Quartus II software can adjust PLL
characteristics to achieve the desired bandwidth. The programmable
Altera Corporation
July 2005
2–91
Stratix Device Handbook, Volume 1
PLLs & Clock Networks
bandwidth is tuned by varying the charge pump current, loop filter
resistor value, high frequency capacitor value, and m counter value. You
can manually adjust these values if desired. Bandwidth is programmable
from 200 kHz to 1.5 MHz.
External Clock Outputs
Enhanced PLLs 5 and 6 each support up to eight single-ended clock
outputs (or four differential pairs). Differential SSTL and HSTL outputs
are implemented using 2 single-ended output buffers which are
programmed to have opposite polarity. In Quartus II software, simply
assign the appropriate differential I/O standard and the software will
implement the inversion. See Figure 2–55.
2–92
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–55. External Clock Outputs for PLLs 5 & 6
From IOE (1), (2)
pll_out0p (3), (4)
(3)
e0 Counter
From IOE (1)
From IOE (1)
pll_out0n (3), (4)
pll_out1p (3), (4)
e1 Counter
4
From IOE (1)
From IOE (1)
pll_out1n (3), (4)
pll_out2p (3), (4)
e2 Counter
pll_out2n (3), (4)
From IOE (1)
From IOE (1)
pll_out3p (3), (4)
e3 Counter
From IOE (1)
pll_out3n (3), (4)
Notes to Figure 2–55:
(1)
(2)
(3)
(4)
The design can use each external clock output pin as a general-purpose output pin from the logic array. These pins
are multiplexed with IOE outputs.
Two single-ended outputs are possible per output counter⎯either two outputs of the same frequency and phase or
one shifted 180° .
EP1S10, EP1S20, and EP1S25 devices in 672-pin BGA and 484- and 672-pin FineLine BGA packages only have two
pairs of external clocks (i.e., pll_out0p, pll_out0n, pll_out1p, and pll_out1n).
Differential SSTL and HSTL outputs are implemented using two single-ended output buffers, which are
programmed to have opposite polarity.
Altera Corporation
July 2005
2–93
Stratix Device Handbook, Volume 1
PLLs & Clock Networks
Any of the four external output counters can drive the single-ended or
differential clock outputs for PLLs 5 and 6. This means one counter or
frequency can drive all output pins available from PLL 5 or PLL 6. Each
pair of output pins (four pins total) has dedicated VCC and GND pins to
reduce the output clock’s overall jitter by providing improved isolation
from switching I/O pins.
For PLLs 5 and 6, each pin of a single-ended output pair can either be in
phase or 180° out of phase. The clock output pin pairs support the same
I/O standards as standard output pins (in the top and bottom banks) as
well as LVDS, LVPECL, 3.3-V PCML, HyperTransport technology,
differential HSTL, and differential SSTL. Table 2–20 shows which I/O
standards the enhanced PLL clock pins support. When in single-ended or
differential mode, the two outputs operate off the same power supply.
Both outputs use the same standards in single-ended mode to maintain
performance. You can also use the external clock output pins as user
output pins if external enhanced PLL clocking is not needed.
Table 2–20. I/O Standards Supported for Enhanced PLL Pins (Part 1 of 2)
Input
Output
I/O Standard
INCLK
FBIN
PLLENABLE
EXTCLK
LVTTL
v
v
v
v
LVCMOS
v
v
v
v
2.5 V
v
v
v
1.8 V
v
v
v
1.5 V
v
v
v
3.3-V PCI
v
v
v
3.3-V PCI-X 1.0
v
v
v
LVPECL
v
v
v
3.3-V PCML
v
v
v
LVDS
v
v
v
HyperTransport technology
v
v
v
Differential HSTL
v
v
v
Differential SSTL
3.3-V GTL
v
v
v
3.3-V GTL+
v
v
v
1.5-V HSTL Class I
v
v
v
2–94
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Table 2–20. I/O Standards Supported for Enhanced PLL Pins (Part 2 of 2)
Input
Output
I/O Standard
INCLK
FBIN
PLLENABLE
EXTCLK
1.5-V HSTL Class II
v
v
v
1.8-V HSTL Class I
v
v
v
1.8-V HSTL Class II
v
v
v
SSTL-18 Class I
v
v
v
SSTL-18 Class II
v
v
v
SSTL-2 Class I
v
v
v
SSTL-2 Class II
v
v
v
SSTL-3 Class I
v
v
v
SSTL-3 Class II
v
v
v
AGP (1× and 2× )
v
v
v
CTT
v
v
v
Enhanced PLLs 11 and 12 support one single-ended output each (see
Figure 2–56). These outputs do not have their own VCC and GND signals.
Therefore, to minimize jitter, do not place switching I/O pins next to this
output pin.
Figure 2–56. External Clock Outputs for Enhanced PLLs 11 & 12
g0
Counter
CLK13n, I/O, PLL11_OUT
or CLK6n, I/O, PLL12_OUT (1)
From Internal
Logic or IOE
Note to Figure 2–56:
(1)
For PLL 11, this pin is CLK13n; for PLL 12 this pin is CLK7n.
Stratix devices can drive any enhanced PLL driven through the global
clock or regional clock network to any general I/O pin as an external
output clock. The jitter on the output clock is not guaranteed for these
cases.
Altera Corporation
July 2005
2–95
Stratix Device Handbook, Volume 1
PLLs & Clock Networks
Clock Feedback
The following four feedback modes in Stratix device enhanced PLLs
allow multiplication and/or phase and delay shifting:
■
Zero delay buffer: The external clock output pin is phase-aligned
with the clock input pin for zero delay. Altera recommends using the
same I/O standard on the input clock and the output clocks for
optimum performance.
■
External feedback: The external feedback input pin, FBIN, is phasealigned with the clock input, CLK, pin. Aligning these clocks allows
you to remove clock delay and skew between devices. This mode is
only possible for PLLs 5 and 6. PLLs 5 and 6 each support feedback
for one of the dedicated external outputs, either one single-ended or
one differential pair. In this mode, one e counter feeds back to the PLL
FBIN input, becoming part of the feedback loop. Altera recommends
using the same I/O standard on the input clock, the FBIN pin, and
the output clocks for optimum performance.
■
Normal mode: If an internal clock is used in this mode, it is phasealigned to the input clock pin. The external clock output pin will
have a phase delay relative to the clock input pin if connected in this
mode. You define which internal clock output from the PLL should
be phase-aligned to the internal clock pin.
■
No compensation: In this mode, the PLL will not compensate for any
clock networks or external clock outputs.
Phase & Delay Shifting
Stratix device enhanced PLLs provide advanced programmable phase
and clock delay shifting. These parameters are set in the Quartus II
software.
Phase Delay
The Quartus II software automatically sets the phase taps and counter
settings according to the phase shift entry. You enter a desired phase shift
and the Quartus II software automatically sets the closest setting
achievable. This type of phase shift is not reconfigurable during system
operation. For phase shifting, enter a phase shift (in degrees or time units)
for each PLL clock output port or for all outputs together in one shift. You
can select phase-shifting values in time units with a resolution of 156.25
to 416.66 ps. This resolution is a function of frequency input and the
multiplication and division factors (that is, it is a function of the VCO
period), with the finest step being equal to an eighth (×0.125) of the VCO
period. Each clock output counter can choose a different phase of the
2–96
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
VCO period from up to eight taps for individual fine step selection. Also,
each clock output counter can use a unique initial count setting to achieve
individual coarse shift selection in steps of one VCO period. The
combination of coarse and fine shifts allows phase shifting for the entire
input clock period.
The equation to determine the precision of the phase shifting in degrees
is: 45° ÷ post-scale counter value. Therefore, the maximum step size is
45° , and smaller steps are possible depending on the multiplication and
division ratio necessary on the output counter port.
This type of phase shift provides the highest precision since it is the least
sensitive to process, supply, and temperature variation.
Clock Delay
In addition to the phase shift feature, the ability to fine tune the Δt clock
delay provides advanced time delay shift control on each of the four PLL
outputs. There are time delays for each post-scale counter (e, g, or l) from
the PLL, the n counter, and m counter. Each of these can shift in 250-ps
increments for a range of 3.0 ns. The m delay shifts all outputs earlier in
time, while n delay shifts all outputs later in time. Individual delays on
post-scale counters (e, g, and l) provide positive delay for each output.
Table 2–21 shows the combined delay for each output for normal or zero
delay buffer mode where Δte, Δtg, or Δtl is unique for each PLL output.
The tOUTPUT for a single output can range from –3 ns to +6 ns. The total
delay shift difference between any two PLL outputs, however, must be
less than ±3 ns. For example, shifts on two outputs of –1 and +2 ns is
allowed, but not –1 and +2.5 ns because these shifts would result in a
difference of 3.5 ns. If the design uses external feedback, the Δte delay will
remove delay from outputs, represented by a negative sign (see
Table 2–21). This effect occurs because the Δte delay is then part of the
feedback loop.
Table 2–21. Output Clock Delay for Enhanced PLLs
Normal or Zero Delay Buffer Mode
ΔteOUTPUT = Δtn −Δtm + Δte
ΔtgOUTPUT = Δtn −Δtm + Δtg
ΔtlOUTPUT = Δtn −Δtm + Δtl
External Feedback Mode
ΔteOUTPUT = Δtn −Δtm −Δte (1)
ΔtgOUTPUT = Δtn −Δtm + Δtg
ΔtlOUTPUT = Δtn −Δtm + Δtl
Note to Table 2–21:
(1)
Altera Corporation
July 2005
Δte removes delay from outputs in external feedback mode.
2–97
Stratix Device Handbook, Volume 1
PLLs & Clock Networks
The variation due to process, voltage, and temperature is about ±15% on
the delay settings. PLL reconfiguration can control the clock delay shift
elements, but not the VCO phase shift multiplexers, during system
operation.
Spread-Spectrum Clocking
Stratix device enhanced PLLs use spread-spectrum technology to reduce
electromagnetic interference generation from a system by distributing the
energy over a broader frequency range. The enhanced PLL typically
provides 0.5% down spread modulation using a triangular profile. The
modulation frequency is programmable. Enabling spread-spectrum for a
PLL affects all of its outputs.
Lock Detect
The lock output indicates that there is a stable clock output signal in
phase with the reference clock. Without any additional circuitry, the lock
signal may toggle as the PLL begins tracking the reference clock. You may
need to gate the lock signal for use as a system control. The lock signal
from the locked port can drive the logic array or an output pin.
Whenever the PLL loses lock (for example, inclk jitter, clock switchover,
PLL reconfiguration, power supply noise, and so on), the PLL must be
reset with the areset signal to guarantee correct phase relationship
between the PLL output clocks. If the phase relationship between the
input clock versus output clock, and between different output clocks
from the PLL is not important in the design, then the PLL need not be
reset.
f
See the Stratix FPGA Errata Sheet for more information on implementing
the gated lock signal in a design.
Programmable Duty Cycle
The programmable duty cycle allows enhanced PLLs to generate clock
outputs with a variable duty cycle. This feature is supported on each
enhanced PLL post-scale counter (g0..g3, l0..l3, e0..e3). The duty cycle
setting is achieved by a low and high time count setting for the post-scale
dividers. The Quartus II software uses the frequency input and the
required multiply or divide rate to determine the duty cycle choices.
Advanced Clear & Enable Control
There are several control signals for clearing and enabling PLLs and their
outputs. You can use these signals to control PLL resynchronization and
gate PLL output clocks for low-power applications.
2–98
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
The pllenable pin is a dedicated pin that enables/disables PLLs. When
the pllenable pin is low, the clock output ports are driven by GND and
all the PLLs go out of lock. When the pllenable pin goes high again, the
PLLs relock and resynchronize to the input clocks. You can choose which
PLLs are controlled by the pllenable signal by connecting the
pllenable input port of the altpll megafunction to the common
pllenable input pin.
The areset signals are reset/resynchronization inputs for each PLL. The
areset signal should be asserted every time the PLL loses lock to
guarantee correct phase relationship between the PLL output clocks.
Users should include the areset signal in designs if any of the following
conditions are true:
■
■
PLL Reconfiguration or Clock switchover enables in the design.
Phase relationships between output clocks need to be maintained
after a loss of lock condition
The device input pins or logic elements (LEs) can drive these input
signals. When driven high, the PLL counters will reset, clearing the PLL
output and placing the PLL out of lock. The VCO will set back to its
nominal setting (~700 MHz). When driven low again, the PLL will
resynchronize to its input as it relocks. If the target VCO frequency is
below this nominal frequency, then the output frequency will start at a
higher value than desired as the PLL locks. If the system cannot tolerate
this, the clkena signal can disable the output clocks until the PLL locks.
The pfdena signals control the phase frequency detector (PFD) output
with a programmable gate. If you disable the PFD, the VCO operates at
its last set value of control voltage and frequency with some long-term
drift to a lower frequency. The system continues running when the PLL
goes out of lock or the input clock is disabled. By maintaining the last
locked frequency, the system has time to store its current settings before
shutting down. You can either use your own control signal or a clkloss
status signal to trigger pfdena.
The clkena signals control the enhanced PLL regional and global
outputs. Each regional and global output port has its own clkena signal.
The clkena signals synchronously disable or enable the clock at the PLL
output port by gating the outputs of the g and l counters. The clkena
signals are registered on the falling edge of the counter output clock to
enable or disable the clock without glitches. Figure 2–57 shows the
waveform example for a PLL clock port enable. The PLL can remain
locked independent of the clkena signals since the loop-related counters
are not affected. This feature is useful for applications that require a low
power or sleep mode. Upon re-enabling, the PLL does not need a
Altera Corporation
July 2005
2–99
Stratix Device Handbook, Volume 1
PLLs & Clock Networks
resynchronization or relock period. The clkena signal can also disable
clock outputs if the system is not tolerant to frequency overshoot during
resynchronization.
The extclkena signals work in the same way as the clkena signals, but
they control the external clock output counters (e0, e1, e2, and e3). Upon
re-enabling, the PLL does not need a resynchronization or relock period
unless the PLL is using external feedback mode. In order to lock in
external feedback mode, the external output must drive the board trace
back to the FBIN pin.
Figure 2–57. extclkena Signals
COUNTER
OUTPUT
CLKENA
CLKOUT
Fast PLLs
Stratix devices contain up to eight fast PLLs with high-speed serial
interfacing ability, along with general-purpose features. Figure 2–58
shows a diagram of the fast PLL.
2–100
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–58. Stratix Device Fast PLL
Post-Scale
Counters
diffioclk1 (2)
÷l0
VCO Phase Selection
Selectable at each PLL
Output Port
Global or
regional clock (1)
Clock
Input
Phase
Frequency
Detector
Global or
regional clock
txload_en (3)
rxload_en (3)
÷l1
Global or
regional clock
diffioclk2 (2)
PFD
Charge
Pump
8
Loop
Filter
VCO
÷g0
Global or
regional clock
÷m
Notes to Figure 2–58:
(1)
(2)
(3)
The global or regional clock input can be driven by an output from another PLL or any dedicated CLK or FCLK pin.
It cannot be driven by internally-generated global signals.
In high-speed differential I/O support mode, this high-speed PLL clock feeds the SERDES. Stratix devices only
support one rate of data transfer per fast PLL in high-speed differential I/O support mode.
This signal is a high-speed differential I/O support SERDES control signal.
Clock Multiplication & Division
Stratix device fast PLLs provide clock synthesis for PLL output ports
using m/(post scaler) scaling factors. The input clock is multiplied by the
m feedback factor. Each output port has a unique post scale counter to
divide down the high-frequency VCO. There is one multiply divider, m,
per fast PLL with a range of 1 to 32. There are two post scale L dividers
for regional and/or LVDS interface clocks, and g0 counter for global clock
output port; all range from 1 to 32.
In the case of a high-speed differential interface, set the output counter to
1 to allow the high-speed VCO frequency to drive the SERDES. When
used for clocking the SERDES, the m counter can range from 1 to 30. The
VCO frequency is equal to fIN×m, where VCO frequency must be between
300 and 1000 MHz.
Altera Corporation
July 2005
2–101
Stratix Device Handbook, Volume 1
PLLs & Clock Networks
External Clock Inputs
Each fast PLL supports single-ended or differential inputs for source
synchronous transmitters or for general-purpose use. Sourcesynchronous receivers support differential clock inputs. The fast PLL
inputs are fed by CLK[0..3], CLK[8..11], and FPLL[7..10]CLK
pins, as shown in Figure 2–50 on page 2–85.
Table 2–22 shows the I/O standards supported by fast PLL input pins.
Table 2–22. Fast PLL Port I/O Standards (Part 1 of 2)
Input
I/O Standard
INCLK
PLLENABLE
LVTTL
v
v
LVCMOS
v
v
2.5 V
v
1.8 V
v
1.5 V
v
3.3-V PCI
3.3-V PCI-X 1.0
LVPECL
v
3.3-V PCML
v
LVDS
v
HyperTransport technology
v
Differential HSTL
v
Differential SSTL
3.3-V GTL
3.3-V GTL+
v
1.5-V HSTL Class I
v
1.5-V HSTL Class II
1.8-V HSTL Class I
v
1.8-V HSTL Class II
SSTL-18 Class I
v
SSTL-18 Class II
SSTL-2 Class I
2–102
Stratix Device Handbook, Volume 1
v
Altera Corporation
July 2005
Stratix Architecture
Table 2–22. Fast PLL Port I/O Standards (Part 2 of 2)
Input
I/O Standard
INCLK
SSTL-2 Class II
v
SSTL-3 Class I
v
SSTL-3 Class II
v
PLLENABLE
AGP (1× and 2× )
v
CTT
Table 2–23 shows the performance on each of the fast PLL clock inputs
when using LVDS, LVPECL, 3.3-V PCML, or HyperTransport technology.
Table 2–23. LVDS Performance on Fast PLL Input
Fast PLL Clock Input
CLK0, CLK2, CLK9, CLK11,
FPLL7CLK, FPLL8CLK, FPLL9CLK,
FPLL10CLK
CLK1, CLK3, CLK8, CLK10
Maximum Input Frequency (MHz)
717(1)
645
Note to Table 2–23:
(1)
See the chapter DC & Switching Characteristics of the Stratix Device Handbook,
Volume 1 for more information.
External Clock Outputs
Each fast PLL supports differential or single-ended outputs for sourcesynchronous transmitters or for general-purpose external clocks. There
are no dedicated external clock output pins. Any I/O pin can be driven
by the fast PLL global or regional outputs as an external output pin. The
I/O standards supported by any particular bank determines what
standards are possible for an external clock output driven by the fast PLL
in that bank.
Phase Shifting
Stratix device fast PLLs have advanced clock shift capability that enables
programmable phase shifts. You can enter a phase shift (in degrees or
time units) for each PLL clock output port or for all outputs together in
one shift. You can perform phase shifting in time units with a resolution
range of 125 to 416.66 ps. This resolution is a function of the VCO period,
with the finest step being equal to an eighth (×0.125) of the VCO period.
Altera Corporation
July 2005
2–103
Stratix Device Handbook, Volume 1
I/O Structure
Control Signals
The fast PLL has the same lock output, pllenable input, and areset
input control signals as the enhanced PLL.
If the input clock stops and causes the PLL to lose lock, then the PLL must
be reset for correct phase shift operation.
For more information on high-speed differential I/O support, see “HighSpeed Differential I/O Support” on page 2–130.
I/O Structure
IOEs provide many features, including:
■
■
■
■
■
■
■
■
■
■
■
■
■
■
■
Dedicated differential and single-ended I/O buffers
3.3-V, 64-bit, 66-MHz PCI compliance
3.3-V, 64-bit, 133-MHz PCI-X 1.0 compliance
Joint Test Action Group (JTAG) boundary-scan test (BST) support
Differential on-chip termination for LVDS I/O standard
Programmable pull-up during configuration
Output drive strength control
Slew-rate control
Tri-state buffers
Bus-hold circuitry
Programmable pull-up resistors
Programmable input and output delays
Open-drain outputs
DQ and DQS I/O pins
Double-data rate (DDR) Registers
The IOE in Stratix devices contains a bidirectional I/O buffer, six
registers, and a latch for a complete embedded bidirectional single data
rate or DDR transfer. Figure 2–59 shows the Stratix IOE structure. The
IOE contains two input registers (plus a latch), two output registers, and
two output enable registers. The design can use both input registers and
the latch to capture DDR input and both output registers to drive DDR
outputs. Additionally, the design can use the output enable (OE) register
for fast clock-to-output enable timing. The negative edge-clocked OE
register is used for DDR SDRAM interfacing. The Quartus II software
automatically duplicates a single OE register that controls multiple
output or bidirectional pins.
2–104
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–59. Stratix IOE Structure
Logic Array
OE Register
D
OE
Q
OE Register
D
Q
Output Register
Output A
D
Q
CLK
Output Register
Output B
D
Q
Input Register
D
Q
Input A
Input B
Input Register
D
Q
Input Latch
D
Q
ENA
The IOEs are located in I/O blocks around the periphery of the Stratix
device. There are up to four IOEs per row I/O block and six IOEs per
column I/O block. The row I/O blocks drive row, column, or direct link
interconnects. The column I/O blocks drive column interconnects.
Figure 2–60 shows how a row I/O block connects to the logic array.
Figure 2–61 shows how a column I/O block connects to the logic array.
Altera Corporation
July 2005
2–105
Stratix Device Handbook, Volume 1
I/O Structure
Figure 2–60. Row I/O Block Connection to the Interconnect
R4, R8 & R24
Interconnects
C4, C8 & C16
Interconnects
I/O Interconnect
I/O Block Local
Interconnect
16 Control Signals
from I/O Interconnect (1)
16
28 Data & Control
Signals from
Logic Array (2)
28
LAB
Horizontal
I/O Block
io_dataouta[3..0]
io_dataoutb[3..0]
Direct Link
Interconnect
to Adjacent LAB
Direct Link
Interconnect
to Adjacent LAB
io_clk[7:0]
LAB Local
Interconnect
Horizontal I/O
Block Contains
up to Four IOEs
Notes to Figure 2–60:
(1)
(2)
The 16 control signals are composed of four output enables io_boe[3..0], four clock enables io_bce[3..0],
four clocks io_clk[3..0], and four clear signals io_bclr[3..0].
The 28 data and control signals consist of eight data out lines: four lines each for DDR applications
io_dataouta[3..0] and io_dataoutb[3..0], four output enables io_coe[3..0], four input clock enables
io_cce_in[3..0], four output clock enables io_cce_out[3..0], four clocks io_cclk[3..0], and four clear
signals io_cclr[3..0].
2–106
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–61. Column I/O Block Connection to the Interconnect
42 Data &
Control Signals
from Logic Array (2)
16 Control
Signals from I/O
Interconnect (1)
Vertical I/O
Block Contains
up to Six IOEs
Vertical I/O Block
16
42
io_clk[7..0]
IO_datain[3:0]
I/O Block
Local Interconnect
I/O Interconnect
R4, R8 & R24
Interconnects
LAB
LAB Local
Interconnect
LAB
LAB
C4, C8 & C16
Interconnects
Notes to Figure 2–61:
(1)
(2)
The 16 control signals are composed of four output enables io_boe[3..0], four clock enables io_bce[3..0],
four clocks io_bclk[3..0], and four clear signals io_bclr[3..0].
The 42 data and control signals consist of 12 data out lines; six lines each for DDR applications
io_dataouta[5..0] and io_dataoutb[5..0], six output enables io_coe[5..0], six input clock enables
io_cce_in[5..0], six output clock enables io_cce_out[5..0], six clocks io_cclk[5..0], and six clear
signals io_cclr[5..0].
Altera Corporation
July 2005
2–107
Stratix Device Handbook, Volume 1
I/O Structure
Stratix devices have an I/O interconnect similar to the R4 and C4
interconnect to drive high-fanout signals to and from the I/O blocks.
There are 16 signals that drive into the I/O blocks composed of four
output enables io_boe[3..0], four clock enables io_bce[3..0], four
clocks io_bclk[3..0], and four clear signals io_bclr[3..0]. The
pin’s datain signals can drive the IO interconnect, which in turn drives
the logic array or other I/O blocks. In addition, the control and data
signals can be driven from the logic array, providing a slower but more
flexible routing resource. The row or column IOE clocks, io_clk[7..0],
provide a dedicated routing resource for low-skew, high-speed clocks.
I/O clocks are generated from regional, global, or fast regional clocks (see
“PLLs & Clock Networks” on page 2–73). Figure 2–62 illustrates the
signal paths through the I/O block.
Figure 2–62. Signal Path through the I/O Block
Row or Column
io_clk[7..0]
io_boe[3..0]
From I/O
Interconnect
To Other
IOEs
io_bce[3..0]
io_bclk[3..0]
io_bclr[3..0]
To Logic
Array
io_datain0
io_datain1
oe
ce_in
ce_out
io_coe
io_cce_in
Control
Signal
Selection
aclr/apreset
IOE
sclr/spreset
io_cce_out
From Logic
Array
clk_in
io_cclr
clk_out
io_cclk
io_dataout0
io_dataout1
2–108
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Each IOE contains its own control signal selection for the following
control signals: oe, ce_in, ce_out, aclr/preset, sclr/preset,
clk_in, and clk_out. Figure 2–63 illustrates the control signal
selection.
Figure 2–63. Control Signal Selection per IOE
io_bclk[3..0]
io_bce[3..0]
io_bclr[3..0]
io_boe[3..0]
Dedicated I/O
Clock [7..0]
I/O Interconnect
[15..0]
Local
Interconnect
io_coe
Local
Interconnect
io_cclr
Local
Interconnect
io_cce_out
Local
Interconnect
io_cce_in
Local
Interconnect
io_cclk
ce_out
clk_out
clk_in
ce_in
sclr/preset
aclr/preset
oe
In normal bidirectional operation, the input register can be used for input
data requiring fast setup times. The input register can have its own clock
input and clock enable separate from the OE and output registers. The
output register can be used for data requiring fast clock-to-output
performance. The OE register can be used for fast clock-to-output enable
timing. The OE and output register share the same clock source and the
same clock enable source from local interconnect in the associated LAB,
dedicated I/O clocks, and the column and row interconnects. Figure 2–64
shows the IOE in bidirectional configuration.
Altera Corporation
July 2005
2–109
Stratix Device Handbook, Volume 1
I/O Structure
Figure 2–64. Stratix IOE in Bidirectional I/O Configuration Note (1)
Column or Row
Interconnect
ioe_clk[7..0]
I/O Interconnect
[15..0]
OE
OE Register
D
Output
tZX Delay
Q
clkout
Output
Enable Clock
Enable Delay
ce_out
ENA
CLRN/PRN
OE Register
tCO Delay
VCCIO
Output Clock
Enable Delay
Optional
PCI Clamp
VCCIO
Programmable
Pull-Up
Resistor
aclr/prn
Chip-Wide Reset
Logic Array
to Output
Register Delay
Output Register
D
sclr/preset
Q
ENA
CLRN/PRN
Output
Pin Delay
Drive Strength Control
Open-Drain Output
Slew Control
Input Pin to
Logic Array Delay
Input Register
D
clkin
ce_in
Input Clock
Enable Delay
Input Pin to
Input Register Delay
Bus-Hold
Circuit
Q
ENA
CLRN/PRN
Note to Figure 2–64:
(1)
All input signals to the IOE can be inverted at the IOE.
The Stratix device IOE includes programmable delays that can be
activated to ensure zero hold times, input IOE register-to-logic array
register transfers, or logic array-to-output IOE register transfers.
A path in which a pin directly drives a register may require the delay to
ensure zero hold time, whereas a path in which a pin drives a register
through combinatorial logic may not require the delay. Programmable
delays exist for decreasing input-pin-to-logic-array and IOE input
register delays. The Quartus II Compiler can program these delays to
automatically minimize setup time while providing a zero hold time.
Programmable delays can increase the register-to-pin delays for output
2–110
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
and/or output enable registers. A programmable delay exists to increase
the tZX delay to the output pin, which is required for ZBT interfaces.
Table 2–24 shows the programmable delays for Stratix devices.
Table 2–24. Stratix Programmable Delay Chain
Programmable Delays
Quartus II Logic Option
Input pin to logic array delay
Decrease input delay to internal cells
Input pin to input register delay
Decrease input delay to input register
Output pin delay
Increase delay to output pin
Output enable register tCO delay
Increase delay to output enable pin
Output tZX delay
Increase tZX delay to output pin
Output clock enable delay
Increase output clock enable delay
Input clock enable delay
Increase input clock enable delay
Logic array to output register delay
Decrease input delay to output register
Output enable clock enable delay
Increase output enable clock enable delay
The IOE registers in Stratix devices share the same source for clear or
preset. You can program preset or clear for each individual IOE. You can
also program the registers to power up high or low after configuration is
complete. If programmed to power up low, an asynchronous clear can
control the registers. If programmed to power up high, an asynchronous
preset can control the registers. This feature prevents the inadvertent
activation of another device’s active-low input upon power-up. If one
register in an IOE uses a preset or clear signal then all registers in the IOE
must use that same signal if they require preset or clear. Additionally a
synchronous reset signal is available for the IOE registers.
Double-Data Rate I/O Pins
Stratix devices have six registers in the IOE, which support DDR
interfacing by clocking data on both positive and negative clock edges.
The IOEs in Stratix devices support DDR inputs, DDR outputs, and
bidirectional DDR modes.
When using the IOE for DDR inputs, the two input registers clock double
rate input data on alternating edges. An input latch is also used within the
IOE for DDR input acquisition. The latch holds the data that is present
during the clock high times. This allows both bits of data to be
synchronous with the same clock edge (either rising or falling).
Figure 2–65 shows an IOE configured for DDR input. Figure 2–66 shows
the DDR input timing diagram.
Altera Corporation
July 2005
2–111
Stratix Device Handbook, Volume 1
I/O Structure
Figure 2–65. Stratix IOE in DDR Input I/O Configuration Note (1)
Column or Row
Interconnect
VCCIO
ioe_clk[7..0] (1)
I/O Interconnect
[15..0] (1)
To DQS Local
Bus (3)
DQS Local
Bus (1), (2)
Optional
PCI Clamp
VCCIO
Programmable
Pull-Up
Resistor
Input Pin to
Input Register Delay
sclr
Input Register
D
Q
clkin
Output Clock
Enable Delay
ENA
CLRN/PRN
Bus-Hold
Circuit
aclr/prn
Chip-Wide Reset
Latch
Input Register
D
Q
ENA
CLRN/PRN
D
Q
ENA
CLRN/PRN
Notes to Figure 2–65:
(1)
(2)
(3)
All input signals to the IOE can be inverted at the IOE.
This signal connection is only allowed on dedicated DQ function pins.
This signal is for dedicated DQS function pins only.
2–112
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–66. Input Timing Diagram in DDR Mode
Data at
input pin
A0
B1
A1
B2
A2
B3
A3
B4
CLK
A'
A1
A2
A3
B'
B1
B2
B3
Input To
Logic Array
When using the IOE for DDR outputs, the two output registers are
configured to clock two data paths from LEs on rising clock edges. These
output registers are multiplexed by the clock to drive the output pin at a
×2 rate. One output register clocks the first bit out on the clock high time,
while the other output register clocks the second bit out on the clock low
time. Figure 2–67 shows the IOE configured for DDR output. Figure 2–68
shows the DDR output timing diagram.
Altera Corporation
July 2005
2–113
Stratix Device Handbook, Volume 1
I/O Structure
Figure 2–67. Stratix IOE in DDR Output I/O Configuration Notes (1), (2)
Column or Row
Interconnect
IOE_CLK[7..0]
I/O Interconnect
[15..0]
OE Register
D
Q
Output
tZX Delay
clkout
ENA
CLRN/PRN
OE Register
tCO Delay
Output
Enable Clock
Enable Delay
Output Clock
Enable Delay
aclr/prn
VCCIO
Optional
PCI Clamp
Chip-Wide Reset
OE Register
D
VCCIO
Q
sclr
ENA
CLRN/PRN
Logic Array
to Output
Register Delay
Programmable
Pull-Up
Resistor
Output Register
D
Q
Output
Pin Delay
ENA
CLRN/PRN
Logic Array
to Output
Register Delay
Used for
DDR SDRAM
Output Register
D
clk
Drive Strength Control
Open-Drain Output
Slew Control
Q
ENA
CLRN/PRN
Bus-Hold
Circuit
Notes to Figure 2–67:
(1)
(2)
All input signals to the IOE can be inverted at the IOE.
The tristate is by default active high. It can, however, be designed to be active low.
2–114
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–68. Output Timing Diagram in DDR Mode
CLK
A
A1
A2
A3
A4
B
B1
B2
B3
B4
From Internal
Registers
B1
DDR output
A1
B2
A2
B3
A3
The Stratix IOE operates in bidirectional DDR mode by combining the
DDR input and DDR output configurations. Stratix device I/O pins
transfer data on a DDR bidirectional bus to support DDR SDRAM. The
negative-edge-clocked OE register holds the OE signal inactive until the
falling edge of the clock. This is done to meet DDR SDRAM timing
requirements.
External RAM Interfacing
Stratix devices support DDR SDRAM at up to 200 MHz (400-Mbps data
rate) through dedicated phase-shift circuitry, QDR and QDRII SRAM
interfaces up to 167 MHz, and ZBT SRAM interfaces up to 200 MHz.
Stratix devices also provide preliminary support for reduced latency
DRAM II (RLDRAM II) at rates up to 200 MHz through the dedicated
phase-shift circuitry.
1
f
Altera Corporation
July 2005
In addition to the required signals for external memory
interfacing, Stratix devices offer the optional clock enable signal.
By default the Quartus II software sets the clock enable signal
high, which tells the output register to update with new values.
The output registers hold their own values if the design sets the
clock enable signal low. See Figure 2–64.
To find out more about the DDR SDRAM specification, see the JEDEC
web site (www.jedec.org). For information on memory controller
megafunctions for Stratix devices, see the Altera web site
(www.altera.com). See AN 342: Interfacing DDR SDRAM with Stratix &
Stratix GX Devices for more information on DDR SDRAM interface in
Stratix. Also see AN 349: QDR SRAM Controller Reference Design for
Stratix & Stratix GX Devices and AN 329: ZBT SRAM Controller Reference
Design for Stratix & Stratix GX Devices.
2–115
Stratix Device Handbook, Volume 1
I/O Structure
Tables 2–25 and 2–26 show the performance specification for DDR
SDRAM, RLDRAM II, QDR SRAM, QDRII SRAM, and ZBT SRAM
interfaces in EP1S10 through EP1S40 devices and in EP1S60 and EP1S80
devices. The DDR SDRAM and QDR SRAM numbers in Table 2–25 have
been verified with hardware characterization with third-party DDR
SDRAM and QDR SRAM devices over temperature and voltage
extremes.
Table 2–25. External RAM Support in EP1S10 through EP1S40 Devices
Maximum Clock Rate (MHz)
DDR Memory Type
I/O
Standard
-5 Speed
Grade
-6 Speed Grade
Flip-Chip Flip-Chip
-7 Speed Grade
-8 Speed Grade
WireBond
FlipChip
WireBond
FlipChip
WireBond
DDR SDRAM (1), (2)
SSTL-2
200
167
133
133
100
100
100
DDR SDRAM - side
banks (2), (3), (4)
SSTL-2
150
133
110
133
100
100
100
RLDRAM II (4)
1.8-V HSTL
200
(5)
(5)
(5)
(5)
(5)
(5)
QDR SRAM (6)
1.5-V HSTL
167
167
133
133
100
100
100
QDRII SRAM (6)
1.5-V HSTL
200
167
133
133
100
100
100
ZBT SRAM (7)
LVTTL
200
200
200
167
167
133
133
Notes to Table 2–25:
(1)
(2)
(3)
(4)
(5)
(6)
(7)
These maximum clock rates apply if the Stratix device uses DQS phase-shift circuitry to interface with DDR
SDRAM. DQS phase-shift circuitry is only available in the top and bottom I/O banks (I/O banks 3, 4, 7, and 8).
For more information on DDR SDRAM, see AN 342: Interfacing DDR SDRAM with Stratix & Stratix GX Devices.
DDR SDRAM is supported on the Stratix device side I/O banks (I/O banks 1, 2, 5, and 6) without dedicated DQS
phase-shift circuitry. The read DQS signal is ignored in this mode.
These performance specifications are preliminary.
This device does not support RLDRAM II.
For more information on QDR or QDRII SRAM, see AN 349: QDR SRAM Controller Reference Design for Stratix &
Stratix GX Devices.
For more information on ZBT SRAM, see AN 329: ZBT SRAM Controller Reference Design for Stratix & Stratix GX
Devices.
2–116
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Table 2–26. External RAM Support in EP1S60 & EP1S80 Devices
Maximum Clock Rate (MHz)
DDR Memory Type
I/O Standard
DDR SDRAM (1), (2)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
SSTL-2
167
167
133
DDR SDRAM - side banks (2), (3) SSTL-2
150
133
133
QDR SRAM (4)
1.5-V HSTL
133
133
133
QDRII SRAM (4)
1.5-V HSTL
167
167
133
ZBT SRAM (5)
LVTTL
200
200
167
Notes to Table 2–26:
(1)
(2)
(3)
(4)
(5)
These maximum clock rates apply if the Stratix device uses DQS phase-shift circuitry to interface with DDR
SDRAM. DQS phase-shift circuitry is only available in the top and bottom I/O banks (I/O banks 3, 4, 7, and 8).
For more information on DDR SDRAM, see AN 342: Interfacing DDR SDRAM with Stratix & Stratix GX Devices.
DDR SDRAM is supported on the Stratix device side I/O banks (I/O banks 1, 2, 5, and 6) without dedicated DQS
phase-shift circuitry. The read DQS signal is ignored in this mode. Numbers are preliminary.
For more information on QDR or QDRII SRAM, see AN 349: QDR SRAM Controller Reference Design for Stratix &
Stratix GX Devices.
For more information on ZBT SRAM, see AN 329: ZBT SRAM Controller Reference Design for Stratix & Stratix GX
Devices.
In addition to six I/O registers and one input latch in the IOE for
interfacing to these high-speed memory interfaces, Stratix devices also
have dedicated circuitry for interfacing with DDR SDRAM. In every
Stratix device, the I/O banks at the top (I/O banks 3 and 4) and bottom
(I/O banks 7 and 8) of the device support DDR SDRAM up to 200 MHz.
These pins support DQS signals with DQ bus modes of ×8, ×16, or ×32.
Table 2–27 shows the number of DQ and DQS buses that are supported
per device.
Table 2–27. DQS & DQ Bus Mode Support
Number of ×8
Groups
Number of ×16
Groups
Number of ×32
Groups
672-pin BGA
672-pin FineLine BGA
12 (2)
0
0
484-pin FineLine BGA
780-pin FineLine BGA
16 (3)
0
4
484-pin FineLine BGA
18(4)
7 (5)
4
672-pin BGA
672-pin FineLine BGA
16(3)
7 (5)
4
780-pin FineLine BGA
20
7 (5)
4
Device
EP1S10
EP1S20
(Part 1 of 2) Note (1)
Package
Altera Corporation
July 2005
2–117
Stratix Device Handbook, Volume 1
I/O Structure
Table 2–27. DQS & DQ Bus Mode Support
(Part 2 of 2) Note (1)
Number of ×8
Groups
Number of ×16
Groups
Number of ×32
Groups
16 (3)
8
4
780-pin FineLine BGA
1,020-pin FineLine BGA
20
8
4
EP1S30
956-pin BGA
780-pin FineLine BGA
1,020-pin FineLine BGA
20
8
4
EP1S40
956-pin BGA
1,020-pin FineLine BGA
1,508-pin FineLine BGA
20
8
4
EP1S60
956-pin BGA
1,020-pin FineLine BGA
1,508-pin FineLine BGA
20
8
4
EP1S80
956-pin BGA
1,508-pin FineLine BGA
1,923-pin FineLine BGA
20
8
4
Device
EP1S25
Package
672-pin BGA
672-pin FineLine BGA
Notes to Table 2–27:
(1)
(2)
(3)
(4)
(5)
See the Selectable I/O Standards in Stratix & Stratix GX Devices chapter in the Stratix Device Handbook, Volume 2
for VREF guidelines.
These packages have six groups in I/O banks 3 and 4 and six groups in I/O banks 7 and 8.
These packages have eight groups in I/O banks 3 and 4 and eight groups in I/O banks 7 and 8.
This package has nine groups in I/O banks 3 and 4 and nine groups in I/O banks 7 and 8.
These packages have three groups in I/O banks 3 and 4 and four groups in I/O banks 7 and 8.
A compensated delay element on each DQS pin automatically aligns
input DQS synchronization signals with the data window of their
corresponding DQ data signals. The DQS signals drive a local DQS bus in
the top and bottom I/O banks. This DQS bus is an additional resource to
the I/O clocks and is used to clock DQ input registers with the DQS
signal.
Two separate single phase-shifting reference circuits are located on the
top and bottom of the Stratix device. Each circuit is driven by a system
reference clock through the CLK pins that is the same frequency as the
DQS signal. Clock pins CLK[15..12]p feed the phase-shift circuitry on
the top of the device and clock pins CLK[7..4]p feed the phase-shift
circuitry on the bottom of the device. The phase-shifting reference circuit
on the top of the device controls the compensated delay elements for all
10 DQS pins located at the top of the device. The phase-shifting reference
circuit on the bottom of the device controls the compensated delay
elements for all 10 DQS pins located on the bottom of the device. All
10 delay elements (DQS signals) on either the top or bottom of the device
2–118
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
shift by the same degree amount. For example, all 10 DQS pins on the top
of the device can be shifted by 90° and all 10 DQS pins on the bottom of
the device can be shifted by 72°. The reference circuits require a maximum
of 256 system reference clock cycles to set the correct phase on the DQS
delay elements. Figure 2–69 illustrates the phase-shift reference circuit
control of each DQS delay shift on the top of the device. This same circuit
is duplicated on the bottom of the device.
Figure 2–69. Simplified Diagram of the DQS Phase-Shift Circuitry
Input
Reference
Clock
Phase
Comparator
Up/Down
Counter
Delay Chains
6
Control Signals
to DQS Pins
See the External Memory Interfaces chapter in the Stratix Device Handbook,
Volume 2 for more information on external memory interfaces.
Programmable Drive Strength
The output buffer for each Stratix device I/O pin has a programmable
drive strength control for certain I/O standards. The LVTTL and
LVCMOS standard has several levels of drive strength that the user can
control. SSTL-3 Class I and II, SSTL-2 Class I and II, HSTL Class I and II,
and 3.3-V GTL+ support a minimum setting, the lowest drive strength
that guarantees the IOH/IOL of the standard. Using minimum settings
provides signal slew rate control to reduce system noise and signal
overshoot.
Altera Corporation
July 2005
2–119
Stratix Device Handbook, Volume 1
I/O Structure
Table 2–28 shows the possible settings for the I/O standards with drive
strength control.
Table 2–28. Programmable Drive Strength
I/O Standard
IOH / IOL Current Strength Setting (mA)
3.3-V LVTTL
24 (1), 16, 12, 8, 4
3.3-V LVCMOS
24 (2), 12 (1), 8, 4, 2
2.5-V LVTTL/LVCMOS
16 (1), 12, 8, 2
1.8-V LVTTL/LVCMOS
12 (1), 8, 2
1.5-V LVCMOS
8 (1), 4, 2
GTL/GTL+
1.5-V HSTL Class I and II
1.8-V HSTL Class I and II
SSTL-3 Class I and II
SSTL-2 Class I and II
SSTL-18 Class I and II
Support max and min strength
Notes to Table 2–28:
(1)
(2)
This is the Quartus II software default current setting.
I/O banks 1, 2, 5, and 6 do not support this setting.
Quartus II software version 4.2 and later will report current strength as
“PCI Compliant” for 3.3-V PCI, 3.3-V PCI-X 1.0, and Compact PCI I/O
standards.
Stratix devices support series on-chip termination (OCT) using
programmable drive strength. For more information, contact your Altera
Support Representative.
Open-Drain Output
Stratix devices provide an optional open-drain (equivalent to an opencollector) output for each I/O pin. This open-drain output enables the
device to provide system-level control signals (e.g., interrupt and writeenable signals) that can be asserted by any of several devices.
Slew-Rate Control
The output buffer for each Stratix device I/O pin has a programmable
output slew-rate control that can be configured for low-noise or highspeed performance. A faster slew rate provides high-speed transitions for
high-performance systems. However, these fast transitions may
introduce noise transients into the system. A slow slew rate reduces
system noise, but adds a nominal delay to rising and falling edges. Each
2–120
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
I/O pin has an individual slew-rate control, allowing you to specify the
slew rate on a pin-by-pin basis. The slew-rate control affects both the
rising and falling edges.
Bus Hold
Each Stratix device I/O pin provides an optional bus-hold feature. The
bus-hold circuitry can weakly hold the signal on an I/O pin at its lastdriven state. Since the bus-hold feature holds the last-driven state of the
pin until the next input signal is present, an external pull-up or pull-down
resistor is not needed to hold a signal level when the bus is tri-stated.
Table 2–29 shows bus hold support for different pin types.
Table 2–29. Bus Hold Support
Pin Type
I/O pins
Bus Hold
v
CLK[15..0]
CLK[0,1,2,3,8,9,10,11]
FCLK
v
FPLL[7..10]CLK
The bus-hold circuitry also pulls undriven pins away from the input
threshold voltage where noise can cause unintended high-frequency
switching. You can select this feature individually for each I/O pin. The
bus-hold output drives no higher than VCCIO to prevent overdriving
signals. If the bus-hold feature is enabled, the programmable pull-up
option cannot be used. Disable the bus-hold feature when using opendrain outputs with the GTL+ I/O standard or when the I/O pin has been
configured for differential signals.
The bus-hold circuitry uses a resistor with a nominal resistance (RBH) of
approximately 7 kΩ to weakly pull the signal level to the last-driven state.
See the DC & Switching Characteristics chapter of the Stratix Device
Handbook, Volume 1 for the specific sustaining current driven through this
resistor and overdrive current used to identify the next-driven input
level. This information is provided for each VCCIO voltage level.
The bus-hold circuitry is active only after configuration. When going into
user mode, the bus-hold circuit captures the value on the pin present at
the end of configuration.
Altera Corporation
July 2005
2–121
Stratix Device Handbook, Volume 1
I/O Structure
Programmable Pull-Up Resistor
Each Stratix device I/O pin provides an optional programmable pull-up
resistor during user mode. If this feature is enabled for an I/O pin, the
pull-up resistor (typically 25 kΩ) weakly holds the output to the VCCIO
level of the output pin’s bank. Table 2–30 shows which pin types support
the weak pull-up resistor feature.
Table 2–30. Programmable Weak Pull-Up Resistor Support
Pin Type
I/O pins
Programmable Weak Pull-Up Resistor
v
CLK[15..0]
FCLK
v
FPLL[7..10]CLK
Configuration pins
JTAG pins
v (1)
Note to Table 2–30:
(1)
TDO pins do not support programmable weak pull-up resistors.
Advanced I/O Standard Support
Stratix device IOEs support the following I/O standards:
■
■
■
■
■
■
■
■
■
■
■
■
■
■
■
■
LVTTL
LVCMOS
1.5 V
1.8 V
2.5 V
3.3-V PCI
3.3-V PCI-X 1.0
3.3-V AGP (1× and 2×)
LVDS
LVPECL
3.3-V PCML
HyperTransport
Differential HSTL (on input/output clocks only)
Differential SSTL (on output column clock pins only)
GTL/GTL+
1.5-V HSTL Class I and II
2–122
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
■
■
■
■
■
1.8-V HSTL Class I and II
SSTL-3 Class I and II
SSTL-2 Class I and II
SSTL-18 Class I and II
CTT
Table 2–31 describes the I/O standards supported by Stratix devices.
Table 2–31. Stratix Supported I/O Standards
Type
Input Reference
Voltage (VREF)
(V)
Output Supply
Voltage (VCCIO)
(V)
Board
Termination
Voltage (VTT)
(V)
LVTTL
Single-ended
N/A
3.3
N/A
LVCMOS
Single-ended
N/A
3.3
N/A
2.5 V
Single-ended
N/A
2.5
N/A
1.8 V
Single-ended
N/A
1.8
N/A
1.5 V
Single-ended
N/A
1.5
N/A
3.3-V PCI
Single-ended
N/A
3.3
N/A
3.3-V PCI-X 1.0
I/O Standard
Single-ended
N/A
3.3
N/A
LVDS
Differential
N/A
3.3
N/A
LVPECL
Differential
N/A
3.3
N/A
3.3-V PCML
Differential
N/A
3.3
N/A
HyperTransport
Differential
N/A
2.5
N/A
Differential HSTL (1)
Differential
0.75
1.5
0.75
Differential SSTL (2)
Differential
1.25
2.5
1.25
GTL
Voltage-referenced
0.8
N/A
1.20
GTL+
Voltage-referenced
1.0
N/A
1.5
1.5-V HSTL Class I and II
Voltage-referenced
0.75
1.5
0.75
1.8-V HSTL Class I and II
Voltage-referenced
0.9
1.8
0.9
SSTL-18 Class I and II
Voltage-referenced
0.90
1.8
0.90
SSTL-2 Class I and II
Voltage-referenced
1.25
2.5
1.25
SSTL-3 Class I and II
Voltage-referenced
1.5
3.3
1.5
AGP (1× and 2° )
Voltage-referenced
1.32
3.3
N/A
CTT
Voltage-referenced
1.5
3.3
1.5
Notes to Table 2–31:
(1)
(2)
This I/O standard is only available on input and output clock pins.
This I/O standard is only available on output column clock pins.
Altera Corporation
July 2005
2–123
Stratix Device Handbook, Volume 1
I/O Structure
f
For more information on I/O standards supported by Stratix devices, see
the Selectable I/O Standards in Stratix & Stratix GX Devices chapter of the
Stratix Device Handbook, Volume 2.
Stratix devices contain eight I/O banks in addition to the four enhanced
PLL external clock out banks, as shown in Figure 2–70. The four I/O
banks on the right and left of the device contain circuitry to support highspeed differential I/O for LVDS, LVPECL, 3.3-V PCML, and
HyperTransport inputs and outputs. These banks support all I/O
standards listed in Table 2–31 except PCI I/O pins or PCI-X 1.0, GTL,
SSTL-18 Class II, and HSTL Class II outputs. The top and bottom I/O
banks support all single-ended I/O standards. Additionally, Stratix
devices support four enhanced PLL external clock output banks,
allowing clock output capabilities such as differential support for SSTL
and HSTL. Table 2–32 shows I/O standard support for each I/O bank.
2–124
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–70. Stratix I/O Banks Notes (1), (2), (3)
DQS5T
9
DQS4T
PLL11
(5)
DQS1T
DQS0T
10
Bank 4
LVDS, LVPECL, 3.3-V PCML,
and HyperTransport I/O Block
and Regular I/O Pins (4)
(5)
I/O Banks 1, 2, 5, and 6 Support All
Single-Ended I/O Standards Except
Differential HSTL Output Clocks,
Differential SSTL-2 Output Clocks,
HSTL Class II, GTL, SSTL-18 Class II,
PCI, PCI-X 1.0, and AGP 1×/2×
PLL2
Bank 1
DQS2T
I/O Banks 3, 4, 9 & 10 Support
All Single-Ended I/O Standards
PLL1
Bank 8
PLL3
DQS8B
DQS7B
DQS6B
DQS5B
(5)
LVDS, LVPECL, 3.3-V PCML,
and HyperTransport I/O Block
and Regular I/O Pins (4)
11
VREF5B8 VREF4B8 VREF3B8 VREF2B8 VREF1B8
DQS9B
PLL4
I/O Banks 7, 8, 11 & 12 Support
All Single-Ended I/O Standards
(5)
LVDS, LVPECL, 3.3-V PCML,
and HyperTransport I/O Block
and Regular I/O Pins (4)
PLL8
DQS3T
VREF1B4 VREF2B4 VREF3B4 VREF4B4 VREF5B4 PLL10
LVDS, LVPECL, 3.3-V PCML,
and HyperTransport I/O Block
and Regular I/O Pins (4)
Bank 2
VREF1B2 VREF2B2 VREF3B2 VREF4B2
Bank 3
VREF1B1 VREF2B1 VREF3B1 VREF4B1
PLL5
12
PLL6
Bank 5
DQS6T
VREF4B5 VREF3B5 VREF2B5 VREF1B5
DQS7T
Bank 6
DQS8T
VREF1B3 VREF2B3 VREF3B3 VREF4B3 VREF5B3
VREF4B6 VREF3B6 VREF2B6 VREF1B6
DQS9T
PLL7
Bank 7
PLL12
VREF5B7 VREF4B7 VREF3B7 VREF2B7 VREF1B7
DQS4B
DQS3B
DQS2B
DQS1B
PLL9
DQS0B
Notes to Figure 2–70:
(1)
(2)
(3)
(4)
(5)
Figure 2–70 is a top view of the silicon die. This will correspond to a top-down view for non-flip-chip packages, but
will be a reverse view for flip-chip packages.
Figure 2–70 is a graphic representation only. See the device pin-outs on the web (www.altera.com) and the
Quartus II software for exact locations.
Banks 9 through 12 are enhanced PLL external clock output banks.
If the high-speed differential I/O pins are not used for high-speed differential signaling, they can support all of the
I/O standards except HSTL Class I and II, GTL, SSTL-18 Class II, PCI, PCI-X 1.0, and AGP 1× /2× .
For guidelines for placing single-ended I/O pads next to differential I/O pads, see the Selectable I/O Standards in
Stratix and Stratix GX Devices chapter in the Stratix Device Handbook, Volume 2.
Altera Corporation
July 2005
2–125
Stratix Device Handbook, Volume 1
I/O Structure
Table 2–32 shows I/O standard support for each I/O bank.
Table 2–32. I/O Support by Bank (Part 1 of 2)
Top & Bottom Banks
(3, 4, 7 & 8)
Left & Right Banks
(1, 2, 5 & 6)
Enhanced PLL External
Clock Output Banks
(9, 10, 11 & 12)
LVTTL
v
v
v
LVCMOS
v
v
v
2.5 V
v
v
v
1.8 V
v
v
v
1.5 V
v
v
v
3.3-V PCI
v
3.3-V PCI-X 1.0
v
I/O Standard
v
v
LVPECL
v
v
3.3-V PCML
v
v
LVDS
v
v
HyperTransport technology
v
v
Differential HSTL (clock
inputs)
v
v
Differential HSTL (clock
outputs)
v
Differential SSTL (clock
outputs)
v
3.3-V GTL
v
3.3-V GTL+
v
v
v
v
1.5-V HSTL Class I
v
v
v
1.5-V HSTL Class II
v
1.8-V HSTL Class I
v
1.8-V HSTL Class II
v
SSTL-18 Class I
v
SSTL-18 Class II
v
SSTL-2 Class I
v
v
v
v
v
v
v
v
v
SSTL-2 Class II
v
v
v
SSTL-3 Class I
v
v
v
2–126
Stratix Device Handbook, Volume 1
v
Altera Corporation
July 2005
Stratix Architecture
Table 2–32. I/O Support by Bank (Part 2 of 2)
Top & Bottom Banks
(3, 4, 7 & 8)
Left & Right Banks
(1, 2, 5 & 6)
Enhanced PLL External
Clock Output Banks
(9, 10, 11 & 12)
SSTL-3 Class II
v
v
v
AGP (1× and 2× )
v
CTT
v
I/O Standard
v
v
v
Each I/O bank has its own VCCIO pins. A single device can support 1.5-,
1.8-, 2.5-, and 3.3-V interfaces; each bank can support a different standard
independently. Each bank also has dedicated VREF pins to support any
one of the voltage-referenced standards (such as SSTL-3) independently.
Each I/O bank can support multiple standards with the same VCCIO for
input and output pins. Each bank can support one voltage-referenced
I/O standard. For example, when VCCIO is 3.3 V, a bank can support
LVTTL, LVCMOS, 3.3-V PCI, and SSTL-3 for inputs and outputs.
Differential On-Chip Termination
Stratix devices provide differential on-chip termination (LVDS I/O
standard) to reduce reflections and maintain signal integrity. Differential
on-chip termination simplifies board design by minimizing the number
of external termination resistors required. Termination can be placed
inside the package, eliminating small stubs that can still lead to
reflections. The internal termination is designed using transistors in the
linear region of operation.
Stratix devices support internal differential termination with a nominal
resistance value of 137.5 Ω for LVDS input receiver buffers. LVPECL
signals require an external termination resistor. Figure 2–71 shows the
device with differential termination.
Altera Corporation
July 2005
2–127
Stratix Device Handbook, Volume 1
I/O Structure
Figure 2–71. LVDS Input Differential On-Chip Termination
Transmitting
Device
Receiving Device with
Differential Termination
Z0
+
+
RD
Ð
Ð
Z0
I/O banks on the left and right side of the device support LVDS receiver
(far-end) differential termination.
Table 2–33 shows the Stratix device differential termination support.
Table 2–33. Differential Termination Supported by I/O Banks
Differential Termination Support
I/O Standard Support
Differential termination (1), (2)
Top & Bottom
Banks (3, 4, 7 & 8)
Left & Right Banks
(1, 2, 5 & 6)
v
LVDS
Notes to Table 2–33:
(1)
(2)
Clock pin CLK0, CLK2, CLK9, CLK11, and pins FPLL[7..10]CLK do not support differential termination.
Differential termination is only supported for LVDS because of a 3.3-V VC C I O .
Table 2–34 shows the termination support for different pin types.
Table 2–34. Differential Termination Support Across Pin Types
Pin Type
RD
Top and bottom I/O banks (3, 4, 7, and 8)
DIFFIO_RX[]
v
CLK[0,2,9,11],CLK[4-7],CLK[12-15]
CLK[1,3,8,10]
v
FCLK
FPLL[7..10]CLK
The differential on-chip resistance at the receiver input buffer is
118 Ω ±20 %.
2–128
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
However, there is additional resistance present between the device ball
and the input of the receiver buffer, as shown in Figure 2–72. This
resistance is because of package trace resistance (which can be calculated
as the resistance from the package ball to the pad) and the parasitic layout
metal routing resistance (which is shown between the pad and the
intersection of the on-chip termination and input buffer).
Figure 2–72. Differential Resistance of LVDS Differential Pin Pair (RD)
Pad
Package Ball
0.3 Ω
9.3 Ω
0.3 Ω
9.3 Ω
LVDS
Input Buffer
RD
Differential On-Chip
Termination Resistor
Table 2–35 defines the specification for internal termination resistance for
commercial devices.
Table 2–35. Differential On-Chip Termination
Resistance
Symbol
RD (2)
Description
Internal differential termination for LVDS
Conditions
Unit
Min
Typ
Max
Commercial (1), (3)
110
135
165
W
Industrial (2), (3)
100
135
170
W
Notes to Table 2–35:
(1)
(2)
(3)
Data measured over minimum conditions (Tj = 0 C, VC C I O +5%) and maximum conditions (Tj = 85 C,
VC C I O = –5%).
Data measured over minimum conditions (Tj = –40 C, VCCIO +5%) and maximum conditions (Tj = 100 C,
VCCIO = –5%).
LVDS data rate is supported for 840 Mbps using internal differential termination.
MultiVolt I/O Interface
The Stratix architecture supports the MultiVolt I/O interface feature,
which allows Stratix devices in all packages to interface with systems of
different supply voltages.
The Stratix VCCINT pins must always be connected to a 1.5-V power
supply. With a 1.5-V VCCINT level, input pins are 1.5-V, 1.8-V, 2.5-V, and
3.3-V tolerant. The VCCIO pins can be connected to either a 1.5-V, 1.8-V,
2.5-V, or 3.3-V power supply, depending on the output requirements.
Altera Corporation
July 2005
2–129
Stratix Device Handbook, Volume 1
High-Speed Differential I/O Support
The output levels are compatible with systems of the same voltage as the
power supply (i.e., when VCCIO pins are connected to a 1.5-V power
supply, the output levels are compatible with 1.5-V systems). When
VCCIO pins are connected to a 3.3-V power supply, the output high is
3.3 V and is compatible with 3.3-V or 5.0-V systems.
Table 2–36 summarizes Stratix MultiVolt I/O support.
Table 2–36. Stratix MultiVolt I/O Support Note (1)
Input Signal (5)
VCCIO (V)
Output Signal (6)
1.5 V
1.8 V
2.5 V
3.3 V
5.0 V
1.5
v
v
v (2)
v (2)
v
1.8
v (2)
v
v (2)
v (2)
v (3)
v
2.5
v
v
v (3)
v (3)
v
3.3
v (2)
v
v (3)
v (3)
v (3)
v (4)
1.5 V
1.8 V
2.5 V
3.3 V
5.0 V
v
v
Notes to Table 2–36:
(1)
(2)
(3)
(4)
(5)
(6)
To drive inputs higher than VCCIO but less than 4.1 V, disable the PCI clamping diode. However, to drive 5.0-V
inputs to the device, enable the PCI clamping diode to prevent VI from rising above 4.0 V.
The input pin current may be slightly higher than the typical value.
Although VCCIO specifies the voltage necessary for the Stratix device to drive out, a receiving device powered at a
different level can still interface with the Stratix device if it has inputs that tolerate the VCCIO value.
Stratix devices can be 5.0-V tolerant with the use of an external resistor and the internal PCI clamp diode.
This is the external signal that is driving the Stratix device.
This represents the system voltage that Stratix supports when a VCCIO pin is connected to a specific voltage level.
For example, when VCCIO is 3.3 V and if the I/O standard is LVTTL/LVCMOS, the output high of the signal
coming out from Stratix is 3.3 V and is compatible with 3.3-V or 5.0-V systems.
High-Speed
Differential I/O
Support
Stratix devices contain dedicated circuitry for supporting differential
standards at speeds up to 840 Mbps. The following differential I/O
standards are supported in the Stratix device: LVDS, LVPECL,
HyperTransport, and 3.3-V PCML.
There are four dedicated high-speed PLLs in the EP1S10 to EP1S25
devices and eight dedicated high-speed PLLs in the EP1S30 to EP1S80
devices to multiply reference clocks and drive high-speed differential
SERDES channels.
f
See the Stratix device pin-outs at www.altera.com for additional high
speed DIFFIO pin information for Stratix devices.
2–130
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Table 2–37 shows the number of channels that each fast PLL can clock in
EP1S10, EP1S20, and EP1S25 devices. Tables 2–38 through Table 2–41
show this information for EP1S30, EP1S40, EP1S60, and EP1S80 devices.
Table 2–37. EP1S10, EP1S20 & EP1S25 Device Differential Channels (Part 1 of 2) Note (1)
Device
EP1S10
Package
Transmitter/
Receiver
484-pin FineLine BGA Transmitter (2)
Receiver
672-pin FineLine BGA Transmitter (2)
672-pin BGA
Receiver
780-pin FineLine BGA Transmitter (2)
Receiver
EP1S20
484-pin FineLine BGA Transmitter (2)
Receiver
672-pin FineLine BGA Transmitter (2)
672-pin BGA
Receiver
780-pin FineLine BGA Transmitter (2)
Receiver
Altera Corporation
July 2005
Total
Channels
20
20
36
36
44
44
24
20
48
50
66
66
Maximum
Speed
(Mbps)
Center Fast PLLs
PLL 1
PLL 2
PLL 3
PLL 4
840 (4)
5
5
5
5
840 (3)
10
10
10
10
840 (4)
5
5
5
5
840 (3)
10
10
10
10
624 (4)
9
9
9
9
624 (3)
18
18
18
18
624 (4)
9
9
9
9
624 (3)
18
18
18
18
840 (4)
11
11
11
11
840 (3)
22
22
22
22
840 (4)
11
11
11
11
840 (3)
22
22
22
22
840 (4)
6
6
6
6
840 (3)
12
12
12
12
840 (4)
5
5
5
5
840 (3)
10
10
10
10
624 (4)
12
12
12
12
624 (3)
24
24
24
24
624 (4)
13
12
12
13
624 (3)
25
25
25
25
840 (4)
17
16
16
17
840 (3)
33
33
33
33
840 (4)
17
16
16
17
840 (3)
33
33
33
33
2–131
Stratix Device Handbook, Volume 1
High-Speed Differential I/O Support
Table 2–37. EP1S10, EP1S20 & EP1S25 Device Differential Channels (Part 2 of 2) Note (1)
Device
EP1S25
Transmitter/
Receiver
Package
672-pin FineLine BGA Transmitter (2)
672-pin BGA
Receiver
780-pin FineLine BGA Transmitter (2)
Receiver
1,020-pin FineLine
BGA
Transmitter (2)
Receiver
Total
Channels
56
58
70
66
78
78
Maximum
Speed
(Mbps)
Center Fast PLLs
PLL 1
PLL 2
PLL 3
PLL 4
624 (4)
14
14
14
14
624 (3)
28
28
28
28
624 (4)
14
15
15
14
624 (3)
29
29
29
29
840 (4)
18
17
17
18
840 (3)
35
35
35
35
840 (4)
17
16
16
17
840 (3)
33
33
33
33
840 (4)
19
20
20
19
840 (3)
39
39
39
39
840 (4)
19
20
20
19
840 (3)
39
39
39
39
Notes to Table 2–37:
(1)
(2)
(3)
(4)
The first row for each transmitter or receiver reports the number of channels driven directly by the PLL. The second
row below it shows the maximum channels a PLL can drive if cross bank channels are used from the adjacent center
PLL. For example, in the 484-pin FineLine BGA EP1S10 device, PLL 1 can drive a maximum of five channels at
840 Mbps or a maximum of 10 channels at 840 Mbps. The Quartus II software may also merge receiver and
transmitter PLLs when a receiver is driving a transmitter. In this case, one fast PLL can drive both the maximum
numbers of receiver and transmitter channels.
The number of channels listed includes the transmitter clock output (tx_outclock) channel. If the design requires
a DDR clock, it can use an extra data channel.
These channels span across two I/O banks per side of the device. When a center PLL clocks channels in the opposite
bank on the same side of the device it is called cross-bank PLL support. Both center PLLs can clock cross-bank
channels simultaneously if, for example, PLL_1 is clocking all receiver channels and PLL_2 is clocking all
transmitter channels. You cannot have two adjacent PLLs simultaneously clocking cross-bank receiver channels or
two adjacent PLLs simultaneously clocking transmitter channels. Cross-bank allows for all receiver channels on
one side of the device to be clocked on one clock while all transmitter channels on the device are clocked on the
other center PLL. Crossbank PLLs are supported at full-speed, 840 Mbps. For wire-bond devices, the full-speed is
624 Mbps.
These values show the channels available for each PLL without crossing another bank.
When you span two I/O banks using cross-bank support, you can route
only two load enable signals total between the PLLs. When you enable
rx_data_align, you use both rxloadena and txloadena of a PLL.
That leaves no loadena for the second PLL.
2–132
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
The only way you can use the rx_data_align is if one of the following
is true:
■
■
The receiver PLL is only clocking receive channels (no resources for
the transmitter)
If all channels can fit in one I/O bank
Table 2–38. EP1S30 Differential Channels Note (1)
Package
780-pin
FineLine
BGA
956-pin
BGA
1,020-pin
FineLine
BGA
Transmitter
Total
/Receiver Channels
Transmitter
(4)
70
Receiver
66
Transmitter
(4)
80
Receiver
80
Transmitter
(4)
Receiver
80 (2) (7)
80 (2) (7)
Maximum
Center Fast PLLs
Corner Fast PLLs (2), (3)
Speed
(Mbps) PLL1 PLL2 PLL3 PLL4 PLL7 PLL8 PLL9 PLL10
840
18
17
17
18
(6)
(6)
(6)
(6)
840 (5)
35
35
35
35
(6)
(6)
(6)
(6)
840
17
16
16
17
(6)
(6)
(6)
(6)
840 (5)
33
33
33
33
(6)
(6)
(6)
(6)
840
19
20
20
19
20
20
20
20
840 (5)
39
39
39
39
20
20
20
20
840
20
20
20
20
19
20
20
19
840 (5)
40
40
40
40
19
20
20
19
840
19
(1)
20
20
19
(1)
20
20
20
20
840 (5),(8)
39
(1)
39
(1)
39
(1)
39
(1)
20
20
20
20
840
20
20
20
20
19 (1)
20
20
19 (1)
840 (5),(8)
40
40
40
40
19 (1)
20
20
19 (1)
Table 2–39. EP1S40 Differential Channels (Part 1 of 2) Note (1)
Package
780-pin
FineLine
BGA
Transmitter/
Total
Receiver Channels
Transmitter
(4)
68
Receiver
66
Altera Corporation
July 2005
Maximum
Speed
(Mbps)
Center Fast PLLs
Corner Fast PLLs (2), (3)
PLL1 PLL2 PLL3 PLL4 PLL7 PLL8 PLL9 PLL10
840
18
16
16
18
(6)
(6)
(6)
(6)
840 (5)
34
34
34
34
(6)
(6)
(6)
(6)
840
17
16
16
17
(6)
(6)
(6)
(6)
840 (5)
33
33
33
33
(6)
(6)
(6)
(6)
2–133
Stratix Device Handbook, Volume 1
High-Speed Differential I/O Support
Table 2–39. EP1S40 Differential Channels (Part 2 of 2) Note (1)
Package
956-pin
BGA
1,020-pin
FineLine
BGA
Transmitter/
Total
Receiver Channels
Transmitter
(4)
80
Receiver
Transmitter
(4)
Receiver
1,508-pin
FineLine
BGA
Transmitter
(4)
Receiver
Maximum
Speed
(Mbps)
Center Fast PLLs
Corner Fast PLLs (2), (3)
PLL1 PLL2 PLL3 PLL4 PLL7 PLL8 PLL9 PLL10
840
18
17
17
18
20
20
20
20
840 (5)
35
35
35
35
20
20
20
20
80
840
20
20
20
20
18
17
17
18
840 (5)
40
40
40
40
18
17
17
18
80 (10)
(7)
840
18
(2)
17
(3)
17
(3)
18
(2)
20
20
20
20
840 (5), (8)
35
(5)
35
(5)
35
(5)
35
(5)
20
20
20
20
840
20
20
20
20
18
(2)
17
(3)
17
(3)
18 (2)
840 (5), (8)
40
40
40
40
18
(2)
17
(3)
17
(3)
18 (2)
840
18
(2)
17
(3)
17
(3)
18
(2)
20
20
20
20
840 (5), (8)
35
(5)
35
(5)
35
(5)
35
(5)
20
20
20
20
840
20
20
20
20
18
(2)
17
(3)
17
(3)
18 (2)
840 (5), (8)
40
40
40
40
18
(2)
17
(3)
17
(3)
18 (2)
80 (10)
(7)
80 (10)
(7)
80 (10)
(7)
Table 2–40. EP1S60 Differential Channels (Part 1 of 2) Note (1)
Package
956-pin
BGA
Transmitter/
Total
Receiver Channels
Transmitter
(4)
80
Receiver
80
2–134
Stratix Device Handbook, Volume 1
Maximum
Center Fast PLLs
Corner Fast PLLs (2), (3)
Speed
PLL1 PLL2 PLL3 PLL4 PLL7 PLL8 PLL9 PLL10
(Mbps)
840
12
10
10
12
20
20
20
20
840 (5), (8)
22
22
22
22
20
20
20
20
840
20
20
20
20
12
10
10
12
840 (5), (8)
40
40
40
40
12
10
10
12
Altera Corporation
July 2005
Stratix Architecture
Table 2–40. EP1S60 Differential Channels (Part 2 of 2) Note (1)
Package
1,020-pin
FineLine
BGA
Transmitter/
Total
Receiver Channels
Transmitter
(4)
Receiver
1,508-pin
FineLine
BGA
Transmitter
(4)
Receiver
80 (12)
(7)
80 (10)
(7)
80 (36)
(7)
80 (36)
(7)
Maximum
Center Fast PLLs
Corner Fast PLLs (2), (3)
Speed
PLL1 PLL2 PLL3 PLL4 PLL7 PLL8 PLL9 PLL10
(Mbps)
840
12
(2)
10
(4)
10
(4)
12
(2)
20
20
20
20
840 (5), (8)
22
(6)
22
(6)
22
(6)
22
(6)
20
20
20
20
840
20
20
20
20
12
(8)
10
(10)
10
(10)
12 (8)
840 (5), (8)
40
40
40
40
12
(8)
10
(10)
10
(10)
12 (8)
840
12
(8)
10
(10)
10
(10)
12
(8)
20
20
20
20
840 (5),(8)
22
(18)
22
(18)
22
(18)
22
(18)
20
20
20
20
840
20
20
20
20
12
(8)
10
(10)
10
(10)
12 (8)
840 (5),(8)
40
40
40
40
12
(8)
10
(10)
10
(10)
12 (8)
Table 2–41. EP1S80 Differential Channels (Part 1 of 2) Note (1)
Package
956-pin
BGA
1,020-pin
FineLine
BGA
Transmitter/
Total
Receiver Channels
Maximum
Center Fast PLLs
Corner Fast PLLs (2), (3)
Speed
(Mbps) PLL1 PLL2 PLL3 PLL4 PLL7 PLL8 PLL9 PLL10
Transmitter
(4)
80 (40)
(7)
840
10
10
10
10
20
20
20
20
840 (5),(8)
20
20
20
20
20
20
20
20
Receiver
80
840
20
20
20
20
10
10
10
10
840 (5),(8)
40
40
40
40
10
10
10
10
Transmitter
(4)
92 (12)
(7)
840
10
(2)
10
(4)
10
(4)
10
(2)
20
20
20
20
840 (5),(8)
20
(6)
20
(6)
20
(6)
20
(6)
20
20
20
20
840
20
20
20
20
10
(2)
10
(3)
10 (3)
10 (2)
840 (5),(8)
40
40
40
40
10
(2)
10
(3)
10 (3)
10 (2)
Receiver
Altera Corporation
July 2005
90 (10)
(7)
2–135
Stratix Device Handbook, Volume 1
High-Speed Differential I/O Support
Table 2–41. EP1S80 Differential Channels (Part 2 of 2) Note (1)
Package
1,508-pin
FineLine
BGA
Transmitter/
Total
Receiver Channels
Transmitter
(4)
Receiver
80 (72)
(7)
80 (56)
(7)
Maximum
Center Fast PLLs
Corner Fast PLLs (2), (3)
Speed
(Mbps) PLL1 PLL2 PLL3 PLL4 PLL7 PLL8 PLL9 PLL10
840
10
(10)
10
(10)
10
(10)
10
(10)
20
(8)
20
(8)
20 (8)
20 (8)
840 (5),(8)
20
(20)
20
(20)
20
(20)
20
(20)
20
(8)
20
(8)
20 (8)
20 (8)
840
20
20
20
20
10
(14)
10
(14)
10
(14)
10
(14)
840 (5),(8)
40
40
40
40
10
(14)
10
(14)
10
(14)
10
(14)
Notes to Tables 2–38 through 2–41:
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
The first row for each transmitter or receiver reports the number of channels driven directly by the PLL. The second
row below it shows the maximum channels a PLL can drive if cross bank channels are used from the adjacent center
PLL. For example, in the 780-pin FineLine BGA EP1S30 device, PLL 1 can drive a maximum of 18 transmitter
channels at 840 Mbps or a maximum of 35 transmitter channels at 840 Mbps. The Quartus II software may also
merge transmitter and receiver PLLs when a receiver is driving a transmitter. In this case, one fast PLL can drive
both the maximum numbers of receiver and transmitter channels.
Some of the channels accessible by the center fast PLL and the channels accessible by the corner fast PLL overlap.
Therefore, the total number of channels is not the addition of the number of channels accessible by PLLs 1, 2, 3, and
4 with the number of channels accessible by PLLs 7, 8, 9, and 10. For more information on which channels overlap,
see the Stratix device pin-outs at www.altera.com.
The corner fast PLLs in this device support a data rate of 840 Mbps for channels labeled “high” speed in the device
pin-outs at www.altera.com.
The numbers of channels listed include the transmitter clock output (tx_outclock) channel. An extra data
channel can be used if a DDR clock is needed.
These channels span across two I/O banks per side of the device. When a center PLL clocks channels in the opposite
bank on the same side of the device it is called cross-bank PLL support. Both center PLLs can clock cross-bank
channels simultaneously if say PLL_1 is clocking all receiver channels and PLL_2 is clocking all transmitter
channels. You cannot have two adjacent PLLs simultaneously clocking cross-bank receiver channels or two adjacent
PLLs simultaneously clocking transmitter channels. Cross-bank allows for all receiver channels on one side of the
device to be clocked on one clock while all transmitter channels on the device are clocked on the other center PLL.
Crossbank PLLs are supported at full-speed, 840 Mbps. For wire-bond devices, the full-speed is 624 Mbps.
PLLs 7, 8, 9, and 10 are not available in this device.
The number in parentheses is the number of slow-speed channels, guaranteed to operate at up to 462 Mbps. These
channels are independent of the high-speed differential channels. For the location of these channels, see the device
pin-outs at www.altera.com.
See the Stratix device pin-outs at www.altera.com. Channels marked “high” speed are 840 MBps and “low” speed
channels are 462 MBps.
The high-speed differential I/O circuitry supports the following high
speed I/O interconnect standards and applications:
■
■
■
■
UTOPIA IV
SPI-4 Phase 2 (POS-PHY Level 4)
SFI-4
10G Ethernet XSBI
2–136
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
■
■
RapidIO
HyperTransport
Dedicated Circuitry
Stratix devices support source-synchronous interfacing with LVDS,
LVPECL, 3.3-V PCML, or HyperTransport signaling at up to 840 Mbps.
Stratix devices can transmit or receive serial channels along with a
low-speed or high-speed clock. The receiving device PLL multiplies the
clock by a integer factor W (W = 1 through 32). For example, a
HyperTransport application where the data rate is 800 Mbps and the
clock rate is 400 MHz would require that W be set to 2. The SERDES factor
J determines the parallel data width to deserialize from receivers or to
serialize for transmitters. The SERDES factor J can be set to 4, 7, 8, or 10
and does not have to equal the PLL clock-multiplication W value. For a J
factor of 1, the Stratix device bypasses the SERDES block. For a J factor of
2, the Stratix device bypasses the SERDES block, and the DDR input and
output registers are used in the IOE. See Figure 2–73.
Figure 2–73. High-Speed Differential I/O Receiver / Transmitter Interface Example
R4, R8, and R24
Interconnect
8
840 Mbps
+
–
Data
+
–
8
840 Mbps
8
Data
Dedicated
Receiver
Interface
8×
105 MHz
Dedicated
Transmitter
Interface
Local
Interconnect
Fast
PLL
rx_load_en
8×
tx_load_en
Regional or
global clock
An external pin or global or regional clock can drive the fast PLLs, which
can output up to three clocks: two multiplied high-speed differential I/O
clocks to drive the SERDES block and/or external pin, and a low-speed
clock to drive the logic array.
Altera Corporation
July 2005
2–137
Stratix Device Handbook, Volume 1
High-Speed Differential I/O Support
The Quartus II MegaWizard® Plug-In Manager only allows the
implementation of up to 20 receiver or 20 transmitter channels for each
fast PLL. These channels operate at up to 840 Mbps. The receiver and
transmitter channels are interleaved such that each I/O bank on the left
and right side of the device has one receiver channel and one transmitter
channel per LAB row. Figure 2–74 shows the fast PLL and channel layout
in EP1S10, EP1S20, and EP1S25 devices. Figure 2–75 shows the fast PLL
and channel layout in the EP1S30 to EP1S80 devices.
Figure 2–74. Fast PLL & Channel Layout in the EP1S10, EP1S20 or EP1S25 Devices Note (1)
Up to 20 Receiver and
Transmitter Channels (2)
Transmitter
Up to 20 Receiver and
Transmitter Channels (2)
Transmitter
Receiver
Receiver
CLKIN
Fast
PLL 1
CLKIN
Fast
PLL 2
(3)
Transmitter
Receiver
Up to 20 Receiver and
Transmitter Channels (2)
Fast
PLL 4
CLKIN
Fast
PLL 3
CLKIN
(3)
Transmitter
Receiver
Up to 20 Receiver and
Transmitter Channels (2)
Notes to Figure 2–74:
(1)
(2)
(3)
Wire-bond packages support up to 624 Mbps.
See Table 2–41 for the number of channels each device supports.
There is a multiplexer here to select the PLL clock source. If a PLL uses this multiplexer to clock channels outside of
its bank quadrant, those clocked channels support up to 840 Mbps for “high” speed channels and 462 Mbps for
“low” speed channels, as labeled in the device pin-outs at www.altera.com.
2–138
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Stratix Architecture
Figure 2–75. Fast PLL & Channel Layout in the EP1S30 to EP1S80 Devices Note (1)
FPLL7CLK
Fast
PLL 7
Fast
PLL 10
Up to 20 Receiver and
20 Transmitter
Channels in 20 Rows (2)
Transmitter
FPLL10CLK
Transmitter
Receiver
Receiver
Up to 20 Receiver and
20 Transmitter
Channels in 20 Rows (2)
Transmitter
Transmitter
Receiver
CLKIN
CLKIN
Receiver
Fast
PLL 1
(3)
(3)
Fast
PLL 2
Fast
PLL 4
CLKIN
Fast
PLL 3
CLKIN
Transmitter
Transmitter
Receiver
Receiver
Up to 20 Receiver and
20 Transmitter
Channels in 20 Rows (2)
Transmitter
Transmitter
Receiver
FPLL8CLK
Receiver
Fast
PLL 8
Up to 20 Receiver and
20 Transmitter
Channels in 20 Rows (2)
Fast
PLL 9
FPLL9CLK
Notes to Figure 2–75:
(1)
(2)
(3)
Wire-bond packages support up to 624 Mbps.
See Table 2–38 through 2–41 for the number of channels each device supports.
There is a multiplexer here to select the PLL clock source. If a PLL uses this multiplexer to clock channels outside of
its bank quadrant, those clocked channels support up to 840 Mbps for “high” speed channels and 462 Mbps for
“low” speed channels as labeled in the device pin-outs at www.altera.com.
Altera Corporation
July 2005
2–139
Stratix Device Handbook, Volume 1
Power Sequencing & Hot Socketing
The transmitter external clock output is transmitted on a data channel.
The txclk pin for each bank is located in between data transmitter pins.
For ×1 clocks (e.g., 622 Mbps, 622 MHz), the high-speed PLL clock
bypasses the SERDES to drive the output pins. For half-rate clocks (e.g.,
622 Mbps, 311 MHz) or any other even-numbered factor such as 1/4, 1/7,
1/8, or 1/10, the SERDES automatically generates the clock in the
Quartus II software.
For systems that require more than four or eight high-speed differential
I/O clock domains, a SERDES bypass implementation is possible using
IOEs.
Byte Alignment
For high-speed source synchronous interfaces such as POS-PHY 4, XSBI,
RapidIO, and HyperTransport technology, the source synchronous clock
rate is not a byte- or SERDES-rate multiple of the data rate. Byte
alignment is necessary for these protocols since the source synchronous
clock does not provide a byte or word boundary since the clock is one half
the data rate, not one eighth. The Stratix device’s high-speed differential
I/O circuitry provides dedicated data realignment circuitry for usercontrolled byte boundary shifting. This simplifies designs while saving
LE resources. An input signal to each fast PLL can stall deserializer
parallel data outputs by one bit period. You can use an LE-based state
machine to signal the shift of receiver byte boundaries until a specified
pattern is detected to indicate byte alignment.
Power
Sequencing &
Hot Socketing
Because Stratix devices can be used in a mixed-voltage environment, they
have been designed specifically to tolerate any possible power-up
sequence. Therefore, the VCCIO and VCCINT power supplies may be
powered in any order.
Although you can power up or down the VCCIO and VCCINT power
supplies in any sequence, you should not power down any I/O banks
that contain configuration pins while leaving other I/O banks powered
on. For power up and power down, all supplies (VCCINT and all VCCIO
power planes) must be powered up and down within 100 ms of each
other. This prevents I/O pins from driving out.
Signals can be driven into Stratix devices before and during power up
without damaging the device. In addition, Stratix devices do not drive
out during power up. Once operating conditions are reached and the
device is configured, Stratix devices operate as specified by the user. For
more information, see Hot Socketing in the Selectable I/O Standards in
Stratix & Stratix GX Devices chapter in the Stratix Device Handbook,
Volume 2.
2–140
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
3. Configuration & Testing
S51003-1.3
IEEE Std. 1149.1
(JTAG)
Boundary-Scan
Support
All Stratix® devices provide JTAG BST circuitry that complies with the
IEEE Std. 1149.1a-1990 specification. JTAG boundary-scan testing can be
performed either before or after, but not during configuration. Stratix
devices can also use the JTAG port for configuration together with either
the Quartus® II software or hardware using either Jam Files (.jam) or Jam
Byte-Code Files (.jbc).
Stratix devices support IOE I/O standard setting reconfiguration through
the JTAG BST chain. The JTAG chain can update the I/O standard for all
input and output pins any time before or during user mode through the
CONFIG_IO instruction. You can use this ability for JTAG testing before
configuration when some of the Stratix pins drive or receive from other
devices on the board using voltage-referenced standards. Since the Stratix
device may not be configured before JTAG testing, the I/O pins may not
be configured for appropriate electrical standards for chip-to-chip
communication. Programming those I/O standards via JTAG allows you
to fully test the I/O connection to other devices.
The enhanced PLL reconfiguration bits are part of the JTAG chain before
configuration and after power-up. After device configuration, the PLL
reconfiguration bits are not part of the JTAG chain.
The JTAG pins support 1.5-V/1.8-V or 2.5-V/3.3-V I/O standards. The
TDO pin voltage is determined by the VCCIO of the bank where it resides.
The VCCSEL pin selects whether the JTAG inputs are 1.5-V, 1.8-V, 2.5-V, or
3.3-V compatible.
Stratix devices also use the JTAG port to monitor the logic operation of the
device with the SignalTap® II embedded logic analyzer. Stratix devices
support the JTAG instructions shown in Table 3–1.
The Quartus II software has an Auto Usercode feature where you can
choose to use the checksum value of a programming file as the JTAG user
code. If selected, the checksum is automatically loaded to the USERCODE
register. In the Settings dialog box in the Assignments menu, click Device
& Pin Options, then General, and then turn on the Auto Usercode
option.
Altera Corporation
July 2005
3–1
IEEE Std. 1149.1 (JTAG) Boundary-Scan Support
Table 3–1. Stratix JTAG Instructions
JTAG Instruction
Instruction Code
Description
SAMPLE/PRELOAD 00 0000 0101
Allows a snapshot of signals at the device pins to be captured and
examined during normal device operation, and permits an initial
data pattern to be output at the device pins. Also used by the
SignalTap II embedded logic analyzer.
EXTEST (1)
00 0000 0000
Allows the external circuitry and board-level interconnects to be
tested by forcing a test pattern at the output pins and capturing test
results at the input pins.
BYPASS
11 1111 1111
Places the 1-bit bypass register between the TDI and TDO pins,
which allows the BST data to pass synchronously through selected
devices to adjacent devices during normal device operation.
USERCODE
00 0000 0111
Selects the 32-bit USERCODE register and places it between the
TDI and TDO pins, allowing the USERCODE to be serially shifted
out of TDO.
IDCODE
00 0000 0110
Selects the IDCODE register and places it between TDI and TDO,
allowing the IDCODE to be serially shifted out of TDO.
HIGHZ (1)
00 0000 1011
Places the 1-bit bypass register between the TDI and TDO pins,
which allows the BST data to pass synchronously through selected
devices to adjacent devices during normal device operation, while
tri-stating all of the I/O pins.
CLAMP (1)
00 0000 1010
Places the 1-bit bypass register between the TDI and TDO pins,
which allows the BST data to pass synchronously through selected
devices to adjacent devices during normal device operation while
holding I/O pins to a state defined by the data in the boundary-scan
register.
ICR instructions
Used when configuring an Stratix device via the JTAG port with a
MasterBlasterTM, ByteBlasterMVTM, or ByteBlasterTM II download
cable, or when using a Jam File or Jam Byte-Code File via an
embedded processor or JRunner.
PULSE_NCONFIG
00 0000 0001
Emulates pulsing the nCONFIG pin low to trigger reconfiguration
even though the physical pin is unaffected.
CONFIG_IO
00 0000 1101
Allows configuration of I/O standards through the JTAG chain for
JTAG testing. Can be executed before, after, or during
configuration. Stops configuration if executed during configuration.
Once issued, the CONFIG_IO instruction will hold nSTATUS low
to reset the configuration device. nSTATUS is held low until the
device is reconfigured.
SignalTap II
instructions
Monitors internal device operation with the SignalTap II embedded
logic analyzer.
Note to Table 3–1:
(1)
Bus hold and weak pull-up resistor features override the high-impedance state of HIGHZ, CLAMP, and EXTEST.
3–2
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Configuration & Testing
The Stratix device instruction register length is 10 bits and the USERCODE
register length is 32 bits. Tables 3–2 and 3–3 show the boundary-scan
register length and device IDCODE information for Stratix devices.
Table 3–2. Stratix Boundary-Scan Register Length
Device
Boundary-Scan Register Length
EP1S10
1,317
EP1S20
1,797
EP1S25
2,157
EP1S30
2,253
EP1S40
2,529
EP1S60
3,129
EP1S80
3,777
Table 3–3. 32-Bit Stratix Device IDCODE
IDCODE (32 Bits) (1)
Device
Version (4 Bits)
Part Number (16 Bits)
Manufacturer Identity
(11 Bits)
LSB (1 Bit) (2)
EP1S10
0000
0010 0000 0000 0001
000 0110 1110
1
EP1S20
0000
0010 0000 0000 0010
000 0110 1110
1
EP1S25
0000
0010 0000 0000 0011
000 0110 1110
1
EP1S30
0000
0010 0000 0000 0100
000 0110 1110
1
EP1S40
0000
0010 0000 0000 0101
000 0110 1110
1
EP1S60
0000
0010 0000 0000 0110
000 0110 1110
1
EP1S80
0000
0010 0000 0000 0111
000 0110 1110
1
Notes to Tables 3–2 and 3–3:
(1)
(2)
The most significant bit (MSB) is on the left.
The IDCODE’s least significant bit (LSB) is always 1.
Altera Corporation
July 2005
3–3
Stratix Device Handbook, Volume 1
IEEE Std. 1149.1 (JTAG) Boundary-Scan Support
Figure 3–1 shows the timing requirements for the JTAG signals.
Figure 3–1. Stratix JTAG Waveforms
TMS
TDI
t JCP
t JCH
t JCL
t JPSU
t JPH
TCK
tJPZX
t JPXZ
t JPCO
TDO
tJSH
tJSSU
Signal
to Be
Captured
Signal
to Be
Driven
tJSCO
tJSZX
tJSXZ
Table 3–4 shows the JTAG timing parameters and values for Stratix
devices.
Table 3–4. Stratix JTAG Timing Parameters & Values
Symbol
Parameter
Min
Max
Unit
tJCP
TCK clock period
100
ns
tJCH
TCK clock high time
50
ns
tJCL
TCK clock low time
50
ns
tJPSU
JTAG port setup time
20
ns
tJPH
JTAG port hold time
45
ns
tJPCO
JTAG port clock to output
25
ns
tJPZX
JTAG port high impedance to valid output
25
ns
tJPXZ
JTAG port valid output to high impedance
25
ns
tJSSU
Capture register setup time
20
tJSH
Capture register hold time
45
tJSCO
Update register clock to output
35
ns
tJSZX
Update register high impedance to valid output
35
ns
tJSXZ
Update register valid output to high impedance
35
ns
3–4
Stratix Device Handbook, Volume 1
ns
ns
Altera Corporation
July 2005
Configuration & Testing
1
f
Stratix, Stratix II, Cyclone®, and Cyclone II devices must be
within the first 17 devices in a JTAG chain. All of these devices
have the same JTAG controller. If any of the Stratix, Stratix II,
Cyclone, and Cyclone II devices are in the 18th or after they will
fail configuration. This does not affect SignalTap II.
For more information on JTAG, see the following documents:
■
■
AN 39: IEEE Std. 1149.1 (JTAG) Boundary-Scan Testing in Altera Devices
Jam Programming & Test Language Specification
SignalTap II
Embedded Logic
Analyzer
Stratix devices feature the SignalTap II embedded logic analyzer, which
monitors design operation over a period of time through the IEEE Std.
1149.1 (JTAG) circuitry. You can analyze internal logic at speed without
bringing internal signals to the I/O pins. This feature is particularly
important for advanced packages, such as FineLine BGA® packages,
because it can be difficult to add a connection to a pin during the
debugging process after a board is designed and manufactured.
Configuration
The logic, circuitry, and interconnects in the Stratix architecture are
configured with CMOS SRAM elements. Altera® devices are
reconfigurable. Because every device is tested with a high-coverage
production test program, you do not have to perform fault testing and can
focus on simulation and design verification.
Stratix devices are configured at system power-up with data stored in an
Altera serial configuration device or provided by a system controller.
Altera offers in-system programmability (ISP)-capable configuration
devices that configure Stratix devices via a serial data stream. Stratix
devices can be configured in under 100 ms using 8-bit parallel data at
100 MHz. The Stratix device’s optimized interface allows
microprocessors to configure it serially or in parallel, and synchronously
or asynchronously. The interface also enables microprocessors to treat
Stratix devices as memory and configure them by writing to a virtual
memory location, making reconfiguration easy. After a Stratix device has
been configured, it can be reconfigured in-circuit by resetting the device
and loading new data. Real-time changes can be made during system
operation, enabling innovative reconfigurable computing applications.
Operating Modes
The Stratix architecture uses SRAM configuration elements that require
configuration data to be loaded each time the circuit powers up. The
process of physically loading the SRAM data into the device is called
configuration. During initialization, which occurs immediately after
Altera Corporation
July 2005
3–5
Stratix Device Handbook, Volume 1
Configuration
configuration, the device resets registers, enables I/O pins, and begins to
operate as a logic device. The I/O pins are tri-stated during power-up,
and before and during configuration. Together, the configuration and
initialization processes are called command mode. Normal device
operation is called user mode.
SRAM configuration elements allow Stratix devices to be reconfigured incircuit by loading new configuration data into the device. With real-time
reconfiguration, the device is forced into command mode with a device
pin. The configuration process loads different configuration data,
reinitializes the device, and resumes user-mode operation. You can
perform in-field upgrades by distributing new configuration files either
within the system or remotely.
PORSEL is a dedicated input pin used to select POR delay times of 2 ms
or 100 ms during power-up. When the PORSEL pin is connected to
ground, the POR time is 100 ms; when the PORSEL pin is connected to
VCC, the POR time is 2 ms.
The nIO_PULLUP pin enables a built-in weak pull-up resistor to pull all
user I/O pins to VCCIO before and during device configuration. If
nIO_PULLUP is connected to VCC during configuration, the weak pullups on all user I/O pins are disabled. If connected to ground, the pull-ups
are enabled during configuration. The nIO_PULLUP pin can be pulled to
1.5, 1.8, 2.5, or 3.3 V for a logic level high.
VCCSEL is a dedicated input that is used to choose whether all dedicated
configuration and JTAG input pins can accept 1.5 V/1.8 V or 2.5 V/3.3 V
during configuration. A logic low sets 3.3 V/2.5 V, and a logic high sets
1.8 V/1.5 V. VCCSEL affects the following pins: TDI, TMS, TCK, TRST,
MSEL0, MSEL1, MSEL2, nCONFIG, nCE, DCLK, PLL_ENA, CONF_DONE,
nSTATUS. The VCCSEL pin can be pulled to 1.5, 1.8, 2.5, or 3.3 V for a logic
level high.
The VCCSEL signal does not control the dual-purpose configuration pins
such as the DATA[7..0] and PPA pins (nWS, nRS, CS, nCS, and
RDYnBSY). During configuration, these dual-purpose pins will drive out
voltage levels corresponding to the VCCIO supply voltage that powers the
I/O bank containing the pin. After configuration, the dual-purpose pins
use I/O standards specified in the user design.
TDO and nCEO drive out at the same voltages as the VCCIO supply that
powers the I/O bank containing the pin. Users must select the VCCIO
supply for bank containing TDO accordingly. For example, when using
the ByteBlaster™ MV cable, the VCCIO for the bank containing TDO must
be powered up at 3.3 V.
3–6
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Configuration & Testing
Configuring Stratix FPGAs with JRunner
JRunner is a software driver that configures Altera FPGAs, including
Stratix FPGAs, through the ByteBlaster II or ByteBlasterMV cables in
JTAG mode. The programming input file supported is in Raw Binary File
(.rbf) format. JRunner also requires a Chain Description File (.cdf)
generated by the Quartus II software. JRunner is targeted for embedded
JTAG configuration. The source code is developed for the Windows NT
operating system (OS), but can be customized to run on other platforms.
For more information on the JRunner software driver, see the JRunner
Software Driver: An Embedded Solution to the JTAG Configuration
White Paper and the source files on the Altera web site (www.altera.com).
Configuration Schemes
You can load the configuration data for a Stratix device with one of five
configuration schemes (see Table 3–5), chosen on the basis of the target
application. You can use a configuration device, intelligent controller, or
the JTAG port to configure a Stratix device. A configuration device can
automatically configure a Stratix device at system power-up.
Multiple Stratix devices can be configured in any of five configuration
schemes by connecting the configuration enable (nCE) and configuration
enable output (nCEO) pins on each device.
Table 3–5. Data Sources for Configuration
Configuration Scheme
Data Source
Configuration device
Enhanced or EPC2 configuration device
Passive serial (PS)
MasterBlaster, ByteBlasterMV, or ByteBlaster II
download cable or serial data source
Passive parallel
asynchronous (PPA)
Parallel data source
Fast passive parallel
Parallel data source
JTAG
MasterBlaster, ByteBlasterMV, or ByteBlaster II
download cable, a microprocessor with a Jam or
JBC file, or JRunner
Partial Reconfiguration
The enhanced PLLs within the Stratix device family support partial
reconfiguration of their multiply, divide, and time delay settings without
reconfiguring the entire device. You can use either serial data from the
logic array or regular I/O pins to program the PLL’s counter settings in a
serial chain. This option provides considerable flexibility for frequency
Altera Corporation
July 2005
3–7
Stratix Device Handbook, Volume 1
Configuration
synthesis, allowing real-time variation of the PLL frequency and delay.
The rest of the device is functional while reconfiguring the PLL. See the
Stratix Architecture chapter of the Stratix Device Handbook, Volume 1 for
more information on Stratix PLLs.
Remote Update Configuration Modes
Stratix devices also support remote configuration using an Altera
enhanced configuration device (e.g., EPC16, EPC8, and EPC4 devices)
with page mode selection. Factory configuration data is stored in the
default page of the configuration device. This is the default configuration
that contains the design required to control remote updates and handle
or recover from errors. You write the factory configuration once into the
flash memory or configuration device. Remote update data can update
any of the remaining pages of the configuration device. If there is an error
or corruption in a remote update configuration, the configuration device
reverts back to the factory configuration information.
There are two remote configuration modes: remote and local
configuration. You can use the remote update configuration mode for all
three configuration modes: serial, parallel synchronous, and parallel
asynchronous. Configuration devices (for example, EPC16 devices) only
support serial and parallel synchronous modes. Asynchronous parallel
mode allows remote updates when an intelligent host is used to configure
the Stratix device. This host must support page mode settings similar to
an EPC16 device.
Remote Update Mode
When the Stratix device is first powered up in remote update
programming mode, it loads the configuration located at page address
“000.” The factory configuration should always be located at page
address “000,” and should never be remotely updated. The factory
configuration contains the required logic to perform the following
operations:
■
■
■
Determine the page address/load location for the next application’s
configuration data
Recover from a previous configuration error
Receive new configuration data and write it into the configuration
device
The factory configuration is the default and takes control if an error
occurs while loading the application configuration.
3–8
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Configuration & Testing
While in the factory configuration, the factory-configuration logic
performs the following operations:
■
■
■
Loads a remote update-control register to determine the page
address of the new application configuration
Determines whether to enable a user watchdog timer for the
application configuration
Determines what the watchdog timer setting should be if it is
enabled
The user watchdog timer is a counter that must be continually reset
within a specific amount of time in the user mode of an application
configuration to ensure that valid configuration occurred during a remote
update. Only valid application configurations designed for remote
update can reset the user watchdog timer in user mode. If a valid
application configuration does not reset the user watchdog timer in a
specific amount of time, the timer updates a status register and loads the
factory configuration. The user watchdog timer is automatically disabled
for factory configurations.
If an error occurs in loading the application configuration, the
configuration logic writes a status register to specify the cause of the error.
Once this occurs, the Stratix device automatically loads the factory
configuration, which reads the status register and determines the reason
for reconfiguration. Based on the reason, the factory configuration will
take appropriate steps and will write the remote update control register
to specify the next application configuration page to be loaded.
When the Stratix device successfully loads the application configuration,
it enters into user mode. The Stratix device then executes the main
application of the user. Intellectual property (IP), such as a Nios® (16-bit
ISA) and Nios® II (32-bit ISA) embedded processors, can help the Stratix
device determine when remote update is coming. The Nios embedded
processor or user logic receives incoming data, writes it to the
configuration device, and loads the factory configuration. The factory
configuration will read the remote update status register and determine
the valid application configuration to load. Figure 3–2 shows the Stratix
remote update. Figure 3–3 shows the transition diagram for remote
update mode.
Altera Corporation
July 2005
3–9
Stratix Device Handbook, Volume 1
Configuration
Figure 3–2. Stratix Device Remote Update
(1)
Watchdog
Timer
New Remote
Configuration Data
Configuration
Device
Application Configuration
Page 7
Page 6
Application Configuration
Stratix Device
Factory Configuration
Page 0
Configuration Device Updates
Stratix Device with Factory
Configuration (to Handle Update)
or New Application Configuration
Note to Figure 3–2:
(1)
When the Stratix device is configured with the factory configuration, it can handle update data from EPC16, EPC8,
or EPC4 configuration device pages and point to the next page in the configuration device.
3–10
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Configuration & Testing
Figure 3–3. Remote Update Transition Diagram Notes (1), (2)
Application 1
Configuration
Power-Up
Configuration
Error
Configuration
Error
Reload an
Application
Factory
Configuration
Reload an
Application
Configuration
Error
Application n
Configuration
Notes to Figure 3–3:
(1)
(2)
Remote update of Application Configuration is controlled by a Nios embedded processor or user logic programmed
in the Factory or Application configurations.
Up to seven pages can be specified allowing up to seven different configuration applications.
Altera Corporation
July 2005
3–11
Stratix Device Handbook, Volume 1
Stratix Automated Single Event Upset (SEU) Detection
Local Update Mode
Local update mode is a simplified version of the remote update. This
feature is intended for simple systems that need to load a single
application configuration immediately upon power up without loading
the factory configuration first. Local update designs have only one
application configuration to load, so it does not require a factory
configuration to determine which application configuration to use.
Figure 3–4 shows the transition diagram for local update mode.
Figure 3–4. Local Update Transition Diagram
Power-Up
or nCONFIG
nCONFIG
Application
Configuration
Configuration
Error
Configuration
Error
nCONFIG
Factory
Configuration
Stratix
Automated
Single Event
Upset (SEU)
Detection
Stratix devices offer on-chip circuitry for automated checking of single
event upset (SEU) detection. FPGA devices that operate at high elevations
or in close proximity to earth’s North or South Pole require periodic
checks to ensure continued data integrity. The error detection cyclic
redundancy check (CRC) feature controlled by the Device & Pin Options
dialog box in the Quartus II software uses a 32-bit CRC circuit to ensure
data reliability and is one of the best options for mitigating SEU.
3–12
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
Configuration & Testing
For Stratix, the CRC is computed by the Quartus II software and
downloaded into the device as a part of the configuration bit stream. The
CRC_ERROR pin reports a soft error when configuration SRAM data is
corrupted, triggering device reconfiguration.
Custom-Built Circuitry
Dedicated circuitry is built in the Stratix devices to perform error
detection automatically. You can use the built-in dedicated circuitry for
error detection using CRC feature in Stratix devices, eliminating the need
for external logic. This circuitry will perform error detection
automatically when enabled. This error detection circuitry in Stratix
devices constantly checks for errors in the configuration SRAM cells
while the device is in user mode. You can monitor one external pin for the
error and use it to trigger a re-configuration cycle. Select the desired time
between checks by adjusting a built-in clock divider.
Software Interface
In the Quartus II software version 4.1 and later, you can turn on the
automated error detection CRC feature in the Device & Pin Options
dialog box. This dialog box allows you to enable the feature and set the
internal frequency of the CRC between 400 kHz to 100 MHz. This controls
the rate that the CRC circuitry verifies the internal configuration SRAM
bits in the FPGA device.
For more information on CRC, see AN 357: Error Detection Using CRC in
Altera FPGA Devices.
Temperature
Sensing Diode
Stratix devices include a diode-connected transistor for use as a
temperature sensor in power management. This diode is used with an
external digital thermometer device such as a MAX1617A or MAX1619
from MAXIM Integrated Products. These devices steer bias current
through the Stratix diode, measuring forward voltage and converting this
reading to temperature in the form of an 8-bit signed number (7 bits plus
sign). The external device’s output represents the junction temperature of
the Stratix device and can be used for intelligent power management.
The diode requires two pins (tempdiodep and tempdioden) on the
Stratix device to connect to the external temperature-sensing device, as
shown in Figure 3–5. The temperature sensing diode is a passive element
and therefore can be used before the Stratix device is powered.
Altera Corporation
July 2005
3–13
Stratix Device Handbook, Volume 1
Temperature Sensing Diode
Figure 3–5. External Temperature-Sensing Diode
Stratix Device
Temperature-Sensing
Device
tempdiodep
tempdioden
Table 3–6 shows the specifications for bias voltage and current of the
Stratix temperature sensing diode.
Table 3–6. Temperature-Sensing Diode Electrical Characteristics
Parameter
Minimum
Typical
Maximum
Unit
IBIAS high
80
100
120
μA
IBIAS low
8
10
12
μA
VBP – VBN
VBN
Series resistance
3–14
Stratix Device Handbook, Volume 1
0.3
0.9
0.7
V
V
3
W
Altera Corporation
July 2005
Configuration & Testing
The temperature-sensing diode works for the entire operating range
shown in Figure 3–6.
Figure 3–6. Temperature vs. Temperature-Sensing Diode Voltage
0.95
0.90
100 μA Bias Current
10 μA Bias Current
0.85
0.80
0.75
Voltage
(Across Diode)
0.70
0.65
0.60
0.55
0.50
0.45
0.40
–55
–30
–5
20
45
70
95
120
Temperature ( C)
Altera Corporation
July 2005
3–15
Stratix Device Handbook, Volume 1
Temperature Sensing Diode
3–16
Stratix Device Handbook, Volume 1
Altera Corporation
July 2005
4. DC & Switching
Characteristics
S51004-3.4
Stratix® devices are offered in both commercial and industrial grades.
Industrial devices are offered in -6 and -7 speed grades and commercial
devices are offered in -5 (fastest), -6, -7, and -8 speed grades. This section
specifies the operation conditions for operating junction temperature,
VCCINT and VCCIO voltage levels, and input voltage requirements. The
voltage specifications in this section are specified at the pins of the device
(and not the power supply). If the device operates outside these ranges,
then all DC and AC specifications are not guaranteed. Furthermore, the
reliability of the device may be affected. The timing parameters in this
chapter apply to both commercial and industrial temperature ranges
unless otherwise stated.
Operating
Conditions
Tables 4–1 through 4–8 provide information on absolute maximum
ratings.
Table 4–1. Stratix Device Absolute Maximum Ratings Notes (1), (2)
Symbol
VCCINT
Parameter
Supply voltage
Conditions
With respect to ground
VCCIO
Minimum
Maximum
Unit
–0.5
2.4
V
–0.5
4.6
V
VI
DC input voltage (3)
–0.5
4.6
V
IOUT
DC output current, per pin
–25
40
mA
TSTG
Storage temperature
No bias
TJ
Junction temperature
BGA packages under bias
–65
150
°C
135
°C
Table 4–2. Stratix Device Recommended Operating Conditions (Part 1 of 2)
Symbol
VCCINT
Parameter
Supply voltage for internal
logic and input buffers
Altera Corporation
January 2006
Conditions
(4)
Minimum
Maximum
Unit
1.425
1.575
V
4–1
Operating Conditions
Table 4–2. Stratix Device Recommended Operating Conditions (Part 2 of 2)
Symbol
Parameter
Conditions
Minimum
Maximum
Unit
3.00 (3.135)
3.60 (3.465)
V
Supply voltage for output
buffers, 3.3-V operation
(4), (5)
Supply voltage for output
buffers, 2.5-V operation
(4)
2.375
2.625
V
Supply voltage for output
buffers, 1.8-V operation
(4)
1.71
1.89
V
Supply voltage for output
buffers, 1.5-V operation
(4)
1.4
1.6
V
VI
Input voltage
(3), (6)
–0.5
4.0
V
VO
Output voltage
0
VCCIO
V
TJ
Operating junction
temperature
0
85
°C
–40
100
°C
VCCIO
For commercial use
For industrial use
Table 4–3. Stratix Device DC Operating Conditions Note (7) (Part 1 of 2)
Symbol
Parameter
Conditions
Minimum
Typical
Maximum
Unit
II
Input pin leakage
current
VI = VCCIOmax to 0 V (8)
–10
10
μA
IOZ
Tri-stated I/O pin
leakage current
VO = VCCIOmax to 0 V (8)
–10
10
μA
ICC0
VCC supply current
(standby) (All
memory blocks in
power-down mode)
VI = ground, no load, no
toggling inputs
mA
EP1S10. VI = ground, no
load, no toggling inputs
37
mA
EP1S20. VI = ground, no
load, no toggling inputs
65
mA
EP1S25. VI = ground, no
load, no toggling inputs
90
mA
EP1S30. VI = ground, no
load, no toggling inputs
114
mA
EP1S40. VI = ground, no
load, no toggling inputs
145
mA
EP1S60. VI = ground, no
load, no toggling inputs
200
mA
EP1S80. VI = ground, no
load, no toggling inputs
277
mA
4–2
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–3. Stratix Device DC Operating Conditions Note (7) (Part 2 of 2)
Symbol
RCONF
Parameter
Conditions
Minimum
Value of I/O pin pull- VCCIO = 3.0 V (9)
up resistor before
VCCIO = 2.375 V (9)
and during
VCCIO = 1.71 V (9)
configuration
Typical
Maximum
Unit
20
50
kΩ
30
80
kΩ
60
150
kΩ
Table 4–4. LVTTL Specifications
Symbol
Parameter
Conditions
Minimum
Maximum
Unit
3.0
3.6
V
VCCIO
Output supply voltage
VI H
High-level input voltage
1.7
4.1
V
VIL
Low-level input voltage
–0.5
0.7
V
VOH
High-level output voltage
IOH = –4 to –24 mA (10)
VOL
Low-level output voltage
IOL = 4 to 24 mA (10)
2.4
V
0.45
V
Minimum
Maximum
Unit
3.0
3.6
V
Table 4–5. LVCMOS Specifications
Symbol
Parameter
Conditions
VCCIO
Output supply voltage
VIH
High-level input voltage
1.7
4.1
V
VIL
Low-level input voltage
–0.5
0.7
V
VOH
High-level output voltage
VCCIO = 3.0,
IOH = –0.1 mA
VOL
Low-level output voltage
VCCIO = 3.0,
IOL = 0.1 mA
VCCIO – 0.2
V
0.2
V
Minimum
Maximum
Unit
2.375
2.625
V
1.7
4.1
V
–0.5
0.7
V
Table 4–6. 2.5-V I/O Specifications
Symbol
Parameter
Conditions
VCCIO
Output supply voltage
VIH
High-level input voltage
VIL
Low-level input voltage
VOH
High-level output voltage
IOH = –1 mA (10)
VOL
Low-level output voltage
IOL = 1 mA (10)
Altera Corporation
January 2006
2.0
V
0.4
V
4–3
Stratix Device Handbook, Volume 1
Operating Conditions
Table 4–7. 1.8-V I/O Specifications
Symbol
Parameter
VCCIO
Output supply voltage
VI H
High-level input voltage
Conditions
VIL
Low-level input voltage
VOH
High-level output voltage
IOH = –2 to –8 mA (10)
VOL
Low-level output voltage
IOL = 2 to 8 mA (10)
Minimum
Maximum
Unit
1.65
1.95
V
0.65 × VCCIO
2.25
V
–0.3
0.35 × VCCIO
VCCIO – 0.45
V
V
0.45
V
Minimum
Maximum
Unit
1.4
1.6
V
0.65 × VCCIO
VCCIO + 0.3
V
–0.3
0.35 × VCCIO
V
Table 4–8. 1.5-V I/O Specifications
Symbol
Parameter
VCCIO
Output supply voltage
VI H
High-level input voltage
Conditions
VIL
Low-level input voltage
VOH
High-level output voltage
IOH = –2 mA (10)
VOL
Low-level output voltage
IOL = 2 mA (10)
0.75 × VCCIO
V
0.25 × VCCIO
V
Notes to Tables 4–1 through 4–8:
(1)
(2)
See the Operating Requirements for Altera Devices Data Sheet.
Conditions beyond those listed in Table 4–1 may cause permanent damage to a device. Additionally, device
operation at the absolute maximum ratings for extended periods of time may have adverse affects on the device.
(3) Minimum DC input is –0.5 V. During transitions, the inputs may undershoot to –2.0 V for input currents less than
100 mA and periods shorter than 20 ns, or overshoot to the voltage shown in Table 4–9, based on input duty cycle
for input currents less than 100 mA. The overshoot is dependent upon duty cycle of the signal. The DC case is
equivalent to 100% duty cycle.
(4) Maximum VCC rise time is 100 ms, and VCC must rise monotonically.
(5) VCCIO maximum and minimum conditions for LVPECL, LVDS, and 3.3-V PCML are shown in parentheses.
(6) All pins, including dedicated inputs, clock, I/O, and JTAG pins, may be driven before VCCINT and VCCIO are
powered.
(7) Typical values are for TA = 25°C, VCCINT = 1.5 V, and VCCIO = 1.5 V, 1.8 V, 2.5 V, and 3.3 V.
(8) This value is specified for normal device operation. The value may vary during power-up. This applies for all
VCCIO settings (3.3, 2.5, 1.8, and 1.5 V).
(9) Pin pull-up resistance values will lower if an external source drives the pin higher than VCCIO.
(10) Drive strength is programmable according to the values shown in the Stratix Architecture chapter of the Stratix
Device Handbook, Volume 1.
Table 4–9. Overshoot Input Voltage with Respect to Duty Cycle (Part 1 of 2)
4–4
Stratix Device Handbook, Volume 1
Vin (V)
Maximum Duty Cycle (%)
4.0
100
4.1
90
4.2
50
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–9. Overshoot Input Voltage with Respect to Duty Cycle (Part 2 of 2)
Vin (V)
Maximum Duty Cycle (%)
4.3
30
4.4
17
4.5
10
Figures 4–1 and 4–2 show receiver input and transmitter output
waveforms, respectively, for all differential I/O standards (LVDS, 3.3-V
PCML, LVPECL, and HyperTransport technology).
Figure 4–1. Receiver Input Waveforms for Differential I/O Standards
Single-Ended Waveform
Positive Channel (p) = VIH
VID
Negative Channel (n) = VIL
VCM
Ground
Differential Waveform
VID
p−n=0V
VID
Altera Corporation
January 2006
4–5
Stratix Device Handbook, Volume 1
Operating Conditions
Figure 4–2. Transmitter Output Waveforms for Differential I/O Standards
Single-Ended Waveform
Positive Channel (p) = VOH
VOD
Negative Channel (n) = VOL
VCM
Ground
Differential Waveform
VOD
p−n=0V
VOD
Tables 4–10 through 4–33 recommend operating conditions,
DC operating conditions, and capacitance for 1.5-V Stratix
devices.
Table 4–10. 3.3-V LVDS I/O Specifications (Part 1 of 2)
Symbol
Parameter
VCCIO
I/O supply voltage
VID (6)
Input differential voltage
swing (single-ended)
4–6
Stratix Device Handbook, Volume 1
Conditions
Minimum
Typical
Maximum
Unit
3.135
3.3
3.465
V
0.1 V ≤VCM < 1.1 V
W = 1 through 10
300
1,000
mV
1.1 V ≤VCM ≤1.6 V
W=1
200
1,000
mV
1.1 V ≤VCM ≤1.6 V
W = 2 through10
100
1,000
mV
1.6 V < VCM ≤1.8 V
W = 1 through 10
300
1,000
mV
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–10. 3.3-V LVDS I/O Specifications (Part 2 of 2)
Symbol
VICM
Parameter
Input common mode
voltage (6)
Conditions
Typical
Maximum
Unit
LVDS
0.3 V ≤VID ≤1.0 V
W = 1 through 10
100
1,100
mV
LVDS
0.3 V ≤VID ≤1.0 V
W = 1 through 10
1,600
1,800
mV
LVDS
0.2 V ≤VID ≤1.0 V
W=1
1,100
1,600
mV
LVDS
0.1 V ≤VID ≤1.0 V
W = 2 through 10
1,100
1,600
mV
550
mV
50
mV
1,375
mV
50
mV
110
Ω
VOD (1)
Output differential voltage
(single-ended)
RL = 100 Ω
Δ VOD
Change in VOD between
high and low
RL = 100 Ω
VOCM
Output common mode
voltage
RL = 100 Ω
Δ VOCM
Change in VOCM between
high and low
RL = 100 Ω
RL
Receiver differential input
discrete resistor (external
to Stratix devices)
Altera Corporation
January 2006
Minimum
250
375
1,125
90
1,200
100
4–7
Stratix Device Handbook, Volume 1
Operating Conditions
Table 4–11. 3.3-V PCML Specifications
Symbol
Parameter
Conditions
Minimum
Typical
Maximum
Unit
3.135
3.3
3.465
V
VCCIO
I/O supply voltage
VID (peakto-peak)
Input differential voltage
swing (single-ended)
300
600
mV
VICM
Input common mode
voltage
1.5
3.465
V
VOD
Output differential voltage
(single-ended)
300
500
mV
Δ VOD
Change in VOD between
high and low
50
mV
VOCM
Output common mode
voltage
3.3
V
Δ VOCM
Change in VOCM between
high and low
50
mV
2.5
370
2.85
VT
Output termination voltage
R1
Output external pull-up
resistors
45
VCCIO
50
55
V
Ω
R2
Output external pull-up
resistors
45
50
55
Ω
Minimum
Typical
Maximum
Unit
3.135
3.3
3.465
V
300
1,000
mV
1
2
V
Table 4–12. LVPECL Specifications
Symbol
Parameter
Conditions
VCCIO
I/O supply voltage
VID (peakto-peak)
Input differential voltage
swing (single-ended)
VICM
Input common mode
voltage
VOD
Output differential voltage
(single-ended)
RL = 100 Ω
525
700
970
mV
VOCM
Output common mode
voltage
RL = 100 Ω
1.5
1.7
1.9
V
RL
Receiver differential input
resistor
90
100
110
Ω
4–8
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–13. HyperTransport Technology Specifications
Symbol
Parameter
Conditions
Minimum
Typical
Maximum
Unit
2.375
2.5
2.625
V
VCCIO
I/O supply voltage
VID (peakto-peak)
Input differential voltage
swing (single-ended)
300
900
mV
VICM
Input common mode
voltage
300
900
mV
VOD
Output differential voltage
(single-ended)
RL = 100 Ω
820
mV
Δ VOD
Change in VOD between
high and low
RL = 100 Ω
50
mV
VOCM
Output common mode
voltage
RL = 100 Ω
780
mV
Δ VOCM
Change in VOCM between
high and low
RL = 100 Ω
50
mV
RL
Receiver differential input
resistor
380
485
440
650
90
100
110
Ω
Minimum
Typical
Maximum
Unit
3.0
3.3
3.6
V
Table 4–14. 3.3-V PCI Specifications
Symbol
Parameter
Conditions
VCCIO
Output supply voltage
VIH
High-level input voltage
0.5 ×
VCCIO
VCCIO +
0.5
V
VIL
Low-level input voltage
–0.5
0.3 ×
VCCIO
V
VOH
High-level output voltage
IOUT = –500 μA
VOL
Low-level output voltage
IOUT = 1,500 μA
Altera Corporation
January 2006
0.9 ×
VCCIO
V
0.1 ×
VCCIO
V
4–9
Stratix Device Handbook, Volume 1
Operating Conditions
Table 4–15. PCI-X 1.0 Specifications
Symbol
Parameter
Conditions
Minimum
Typical
Maximum
Unit
3.0
3.6
V
VCCIO
Output supply voltage
VIH
High-level input voltage
0.5 ×
VCCIO
VCCIO +
0.5
V
VIL
Low-level input voltage
–0.5
0.35 ×
VCCIO
V
VIPU
Input pull-up voltage
0.7 ×
VCCIO
V
VOH
High-level output voltage
IOUT = –500 μA
0.9 ×
VCCIO
V
VOL
Low-level output voltage
IOUT = 1,500 μA
0.1 ×
VCCIO
V
Table 4–16. GTL+ I/O Specifications
Symbol
Parameter
Conditions
Minimum
Typical
Maximum
Unit
VTT
Termination voltage
1.35
1.5
1.65
V
VREF
Reference voltage
0.88
1.0
1.12
V
VIH
High-level input voltage
VIL
Low-level input voltage
VOL
Low-level output voltage
VREF + 0.1
V
IOL = 34 mA (3)
VREF – 0.1
V
0.65
V
Table 4–17. GTL I/O Specifications
Symbol
Parameter
Conditions
Minimum
Typical
Maximum
Unit
VTT
Termination voltage
1.14
1.2
1.26
V
VREF
Reference voltage
0.74
0.8
0.86
V
VIH
High-level input voltage
VIL
Low-level input voltage
VOL
Low-level output voltage
4–10
Stratix Device Handbook, Volume 1
VREF +
0.05
IOL = 40 mA (3)
V
VREF –
0.05
V
0.4
V
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–18. SSTL-18 Class I Specifications
Symbol
Parameter
Conditions
Minimum
Typical
Maximum
Unit
VCCIO
Output supply voltage
1.65
1.8
1.95
V
VREF
Reference voltage
0.8
0.9
1.0
V
VREF – 0.04
VREF
VREF + 0.04
VTT
Termination voltage
VIH(DC)
High-level DC input voltage
VIL(DC)
Low-level DC input voltage
VIH(AC)
High-level AC input voltage
VREF +
0.125
VREF – 0.125
VREF +
0.275
VIL(AC)
Low-level AC input voltage
VOH
High-level output voltage
IOH = –6.7 mA
(3)
VOL
Low-level output voltage
IOL = 6.7 mA (3)
V
V
V
V
VREF – 0.275
VTT + 0.475
V
V
VTT – 0.475
V
Typical
Maximum
Unit
Table 4–19. SSTL-18 Class II Specifications
Symbol
Parameter
Conditions
Minimum
VCCIO
Output supply voltage
1.65
1.8
1.95
V
VREF
Reference voltage
0.8
0.9
1.0
V
VTT
Termination voltage
VREF – 0.04
VREF
VREF + 0.04
V
VIH(DC)
High-level DC input voltage
VIL(DC)
Low-level DC input voltage
VIH(AC)
High-level AC input voltage
VIL(AC)
Low-level AC input voltage
VOH
High-level output voltage
IOH = –13.4 mA
(3)
VOL
Low-level output voltage
IOL = 13.4 mA (3)
Altera Corporation
January 2006
VREF +
0.125
V
VREF – 0.125
VREF +
0.275
V
V
VREF – 0.275
VTT + 0.630
V
V
VTT – 0.630
V
4–11
Stratix Device Handbook, Volume 1
Operating Conditions
Table 4–20. SSTL-2 Class I Specifications
Symbol
Parameter
Conditions
Minimum
Typical
Maximum
Unit
2.375
2.5
2.625
V
VREF – 0.04
VREF
VREF + 0.04
V
1.15
1.25
VCCIO
Output supply voltage
VTT
Termination voltage
VREF
Reference voltage
1.35
V
VIH(DC)
High-level DC input voltage
VREF + 0.18
3.0
V
VIL(DC)
Low-level DC input voltage
–0.3
VREF – 0.18
V
VIH(AC)
High-level AC input voltage
VREF + 0.35
VIL(AC)
Low-level AC input voltage
VOH
High-level output voltage
IOH = –8.1 mA
(3)
VOL
Low-level output voltage
IOL = 8.1 mA (3)
V
VREF – 0.35
VTT + 0.57
V
V
VTT – 0.57
V
Maximum
Unit
Table 4–21. SSTL-2 Class II Specifications
Symbol
Parameter
Conditions
Minimum
Typical
VCCIO
Output supply voltage
VTT
Termination voltage
2.375
2.5
2.625
V
VREF – 0.04
VREF
VREF + 0.04
V
VREF
Reference voltage
1.15
1.25
1.35
V
VIH(DC)
High-level DC input voltage
VREF + 0.18
VCCIO + 0.3
V
VIL(DC)
Low-level DC input voltage
–0.3
VREF – 0.18
V
VIH(AC)
High-level AC input voltage
VREF + 0.35
VIL(AC)
Low-level AC input voltage
VOH
High-level output voltage
IOH = –16.4 mA
(3)
VOL
Low-level output voltage
IOL = 16.4 mA (3)
V
VREF – 0.35
VTT + 0.76
V
V
VTT – 0.76
V
Table 4–22. SSTL-3 Class I Specifications (Part 1 of 2)
Symbol
Parameter
VCCIO
Output supply voltage
Conditions
Minimum
Typical
Maximum
Unit
3.0
3.3
3.6
V
VTT
Termination voltage
VREF – 0.05
VREF
VREF + 0.05
V
VREF
Reference voltage
1.3
1.5
1.7
V
VIH(DC)
High-level DC input voltage
VREF + 0.2
VCCIO + 0.3
V
VIL(DC)
Low-level DC input voltage
–0.3
VREF – 0.2
V
VIH(AC)
High-level AC input voltage
VREF + 0.4
4–12
Stratix Device Handbook, Volume 1
V
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–22. SSTL-3 Class I Specifications (Part 2 of 2)
Symbol
Parameter
Conditions
VIL(AC)
Low-level AC input voltage
VOH
High-level output voltage
IOH = –8 mA (3)
VOL
Low-level output voltage
IOL = 8 mA (3)
Minimum
Typical
Maximum
Unit
VREF – 0.4
V
VTT + 0.6
V
VTT – 0.6
V
Table 4–23. SSTL-3 Class II Specifications
Symbol
Parameter
VCCIO
Output supply voltage
Conditions
Minimum
Typical
Maximum
Unit
3.0
3.3
3.6
V
VTT
Termination voltage
VREF – 0.05
VREF
VREF + 0.05
V
VREF
Reference voltage
1.3
1.5
1.7
V
VIH(DC)
High-level DC input voltage
VREF + 0.2
VCCIO + 0.3
V
VREF – 0.2
V
VIL(DC)
Low-level DC input voltage
–0.3
VIH(AC)
High-level AC input voltage
VREF + 0.4
VIL(AC)
Low-level AC input voltage
VOH
High-level output voltage
IOH = –16 mA (3)
VOL
Low-level output voltage
IOL = 16 mA (3)
V
VREF – 0.4
VT T + 0.8
V
V
VTT – 0.8
V
Maximum
Unit
Table 4–24. 3.3-V AGP 2× Specifications
Symbol
Parameter
Conditions
Minimum
Typical
3.15
3.3
VCCIO
Output supply voltage
3.45
V
VREF
Reference voltage
0.39 × VCCIO
0.41 × VCCIO
V
VIH
High-level input voltage (4)
0.5 × VCCIO
VCCIO + 0.5
V
VIL
Low-level input voltage (4)
0.3 × VCCIO
V
VOH
High-level output voltage
IOUT = –0.5 mA
VOL
Low-level output voltage
IOUT = 1.5 mA
0.9 × VCCIO
3.6
V
0.1 × VCCIO
V
Table 4–25. 3.3-V AGP 1× Specifications (Part 1 of 2)
Symbol
Parameter
VCCIO
Output supply voltage
VIH
High-level input voltage (4)
VIL
Low-level input voltage (4)
Altera Corporation
January 2006
Conditions
Minimum
Typical
Maximum
Unit
3.15
3.3
3.45
V
VCCIO + 0.5
V
0.3 × VCCIO
V
0.5 × VCCIO
4–13
Stratix Device Handbook, Volume 1
Operating Conditions
Table 4–25. 3.3-V AGP 1× Specifications (Part 2 of 2)
Symbol
Parameter
Conditions
VOH
High-level output voltage
IOUT = –0.5 mA
VOL
Low-level output voltage
IOUT = 1.5 mA
Minimum
Typical
0.9 × VCCIO
Maximum
Unit
3.6
V
0.1 × VCCIO
V
Table 4–26. 1.5-V HSTL Class I Specifications
Symbol
Parameter
Conditions
Minimum
Typical
Maximum
Unit
VCCIO
Output supply voltage
1.4
1.5
1.6
V
VREF
Input reference voltage
0.68
0.75
0.9
V
VTT
Termination voltage
0.7
0.75
0.8
V
VIH (DC)
DC high-level input voltage
VREF + 0.1
VIL (DC)
DC low-level input voltage
–0.3
VIH (AC)
AC high-level input voltage
VREF + 0.2
VIL (AC)
AC low-level input voltage
VOH
High-level output voltage
IOH = –8 mA (3)
VOL
Low-level output voltage
IOL = 8 mA (3)
V
VREF – 0.1
V
VREF – 0.2
V
V
VCCIO – 0.4
V
0.4
V
Table 4–27. 1.5-V HSTL Class II Specifications
Symbol
Parameter
Conditions
Minimum
Typical
Maximum
Unit
1.5
1.6
V
VCCIO
Output supply voltage
1.4
VREF
Input reference voltage
0.68
0.75
0.9
V
VTT
Termination voltage
0.7
0.75
0.8
V
VIH (DC)
DC high-level input voltage
VREF + 0.1
VIL (DC)
DC low-level input voltage
–0.3
VIH (AC)
AC high-level input voltage
VREF + 0.2
VIL (AC)
AC low-level input voltage
VOH
High-level output voltage
IOH = –16 mA (3)
VOL
Low-level output voltage
IOL = 16 mA (3)
4–14
Stratix Device Handbook, Volume 1
V
VREF – 0.1
V
VREF – 0.2
V
V
VCCIO – 0.4
V
0.4
V
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–28. 1.8-V HSTL Class I Specifications
Symbol
Parameter
Conditions
Minimum
Typical
Maximum
Unit
VCCIO
Output supply voltage
1.65
1.80
1.95
V
VREF
Input reference voltage
0.70
0.90
0.95
V
VTT
Termination voltage
VIH (DC)
DC high-level input voltage
VCCIO ×
0.5
V
VREF + 0.1
VIL (DC)
DC low-level input voltage
–0.5
VIH (AC)
AC high-level input voltage
VREF + 0.2
VIL (AC)
AC low-level input voltage
VOH
High-level output voltage
IOH = –8 mA (3)
VOL
Low-level output voltage
IOL = 8 mA (3)
V
VREF – 0.1
V
V
VREF – 0.2
VCCIO – 0.4
V
V
0.4
V
Maximum
Unit
Table 4–29. 1.8-V HSTL Class II Specifications
Symbol
Parameter
Conditions
Minimum
Typical
VCCIO
Output supply voltage
1.65
1.80
1.95
V
VREF
Input reference voltage
0.70
0.90
0.95
V
VTT
Termination voltage
VIH (DC)
DC high-level input voltage
VREF + 0.1
VIL (DC)
DC low-level input voltage
–0.5
VIH (AC)
AC high-level input voltage
VREF + 0.2
VIL (AC)
AC low-level input voltage
VOH
High-level output voltage
IOH = –16 mA (3)
VOL
Low-level output voltage
IOL = 16 mA (3)
VCCIO ×
0.5
V
V
VREF – 0.1
V
VREF – 0.2
V
V
VCCIO – 0.4
V
0.4
V
Table 4–30. 1.5-V Differential HSTL Class I & Class II Specifications
Symbol
Parameter
Conditions
Minimum
Typical
Maximum
Unit
1.5
1.6
V
VCCIO
I/O supply voltage
1.4
VDIF (DC)
DC input differential
voltage
0.2
VCM (DC)
DC common mode input
voltage
0.68
VDIF (AC)
AC differential input
voltage
0.4
Altera Corporation
January 2006
V
0.9
V
V
4–15
Stratix Device Handbook, Volume 1
Operating Conditions
Table 4–31. CTT I/O Specifications
Symbol
Parameter
Conditions
Minimum
Typical
Maximum
Unit
VCCIO
Output supply voltage
2.05
3.3
3.6
V
VTT/VREF
Termination and input
reference voltage
1.35
1.5
1.65
V
VIH
High-level input voltage
VREF + 0.2
VIL
Low-level input voltage
VOH
High-level output voltage
IOH = –8 mA
VOL
Low-level output voltage
IOL = 8 mA
IO
Output leakage current
(when output is high Z)
GND ≤VOUT ≤
VCCIO
V
VREF – 0.2
VREF + 0.4
V
V
VREF – 0.4
V
10
μA
–10
Table 4–32. Bus Hold Parameters
VCCIO Level
Parameter
Conditions
1.5 V
Min
1.8 V
Max
Min
Max
2.5 V
Min
Unit
3.3 V
Max
Min
Max
VIN > VIL
(maximum)
25
30
50
70
μA
High sustaining VIN < VIH
current
(minimum)
-25
–30
–50
–70
μA
Low sustaining
current
Low overdrive
current
0 V < VIN <
VCCIO
160
200
300
500
μA
High overdrive
current
0 V < VIN <
VCCIO
-160
–200
–300
–500
μA
2.0
V
Bus-hold trip
point
0.5
4–16
Stratix Device Handbook, Volume 1
1.0
0.68
1.07
0.7
1.7
0.8
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–33. Stratix Device Capacitance Note (5)
Symbol
Parameter
Minimum
Typical
Maximum
Unit
CIOTB
Input capacitance on I/O pins in I/O banks
3, 4, 7, and 8.
11.5
pF
CIOLR
Input capacitance on I/O pins in I/O banks
1, 2, 5, and 6, including high-speed
differential receiver and transmitter pins.
8.2
pF
CCLKTB
Input capacitance on top/bottom clock input
pins: CLK[4:7] and CLK[12:15].
11.5
pF
CCLKLR
Input capacitance on left/right clock inputs:
CLK1, CLK3, CLK8, CLK10.
7.8
pF
CCLKLR+
Input capacitance on left/right clock inputs:
CLK0, CLK2, CLK9, and CLK11.
4.4
pF
Notes to Tables 4–10 through 4–33:
(1)
(2)
(3)
(4)
(5)
(6)
When tx_outclock port of altlvds_tx megafunction is 717 MHz, VO D ( m i n ) = 235 mV on the output clock pin.
Pin pull-up resistance values will lower if an external source drives the pin higher than VCCIO.
Drive strength is programmable according to the values shown in the Stratix Architecture chapter of the Stratix
Device Handbook, Volume 1.
VREF specifies the center point of the switching range.
Capacitance is sample-tested only. Capacitance is measured using time-domain reflections (TDR). Measurement
accuracy is within ±0.5 pF.
VIO and VCM have multiple ranges and values for J=1 through 10.
Power
Consumption
Altera® offers two ways to calculate power for a design: the Altera web
power calculator and the PowerGaugeTM feature in the Quartus® II
software.
The interactive power calculator on the Altera web site is typically used
prior to designing the FPGA in order to get a magnitude estimate of the
device power. The Quartus II software PowerGauge feature allows you to
apply test vectors against your design for more accurate power
consumption modeling.
In both cases, these calculations should only be used as an estimation of
power, not as a specification.
Stratix devices require a certain amount of power-up current to
successfully power up because of the small process geometry on which
they are fabricated.
Table 4–34 shows the maximum power-up current (ICCINT) required to
power a Stratix device. This specification is for commercial operating
conditions. Measurements were performed with an isolated Stratix
device on the board to characterize the power-up current of an isolated
Altera Corporation
January 2006
4–17
Stratix Device Handbook, Volume 1
Power Consumption
device. Decoupling capacitors were not used in this measurement. To
factor in the current for decoupling capacitors, sum up the current for
each capacitor using the following equation:
I = C (dV/dt)
If the regulator or power supply minimum output current is more than
the Stratix device requires, then the device may consume more current
than the maximum current listed in Table 4–34. However, the device does
not require any more current to successfully power up than what is listed
in Table 4–34.
Table 4–34. Stratix Power-Up Current (ICCINT) Requirements Note (1)
Power-Up Current Requirement
Device
Unit
Typical
Maximum
EP1S10
250
700
mA
EP1S20
400
1,200
mA
EP1S25
500
1,500
mA
EP1S30
550
1,900
mA
EP1S40
650
2,300
mA
EP1S60
800
2,600
mA
EP1S80
1,000
3,000
mA
Note to Table 4–34:
(1)
The maximum test conditions are for 0° C and typical test conditions are for
40° C.
The exact amount of current consumed varies according to the process,
temperature, and power ramp rate. Stratix devices typically require less
current during power up than shown in Table 4–34. The user-mode
current during device operation is generally higher than the power-up
current.
The duration of the ICCINT power-up requirement depends on the VCCINT
voltage supply rise time. The power-up current consumption drops when
the VCCINT supply reaches approximately 0.75 V.
4–18
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Timing Model
The DirectDrive™ technology and MultiTrack™ interconnect ensure
predictable performance, accurate simulation, and accurate timing
analysis across all Stratix device densities and speed grades. This section
describes and specifies the performance, internal, external, and PLL
timing specifications.
All specifications are representative of worst-case supply voltage and
junction temperature conditions.
Preliminary & Final Timing
Timing models can have either preliminary or final status. The Quartus II
software issues an informational message during the design compilation
if the timing models are preliminary. Table 4–35 shows the status of the
Stratix device timing models.
Preliminary status means the timing model is subject to change. Initially,
timing numbers are created using simulation results, process data, and
other known parameters. These tests are used to make the preliminary
numbers as close to the actual timing parameters as possible.
Final timing numbers are based on actual device operation and testing.
These numbers reflect the actual performance of the device under worstcase voltage and junction temperature conditions.
Table 4–35. Stratix Device Timing Model Status
Device
Altera Corporation
January 2006
Preliminary
Final
EP1S10
v
EP1S20
v
EP1S25
v
EP1S30
v
EP1S40
v
EP1S60
v
EP1S80
v
4–19
Stratix Device Handbook, Volume 1
Timing Model
Performance
Table 4–36 shows Stratix performance for some common designs. All
performance values were obtained with Quartus II software compilation
of LPM, or MegaCore® functions for the FIR and FFT designs.
Table 4–36. Stratix Performance (Part 1 of 2) Notes (1), (2)
Resources Used
Applications
LE
16-to-1 multiplexer (1)
TriMatrix
memory
M-RAM
block
TriMatrix
DSP
LEs Memory
Blocks
Blocks
-5
Speed
Grade
-6
Speed
Grade
-7
Speed
Grade
-8
Speed
Grade
Units
22
407.83
324.56
288.68
228.67
MHz
0
0
32-to-1 multiplexer (3)
46
0
0
318.26
255.29
242.89
185.18
MHz
16-bit counter
16
0
0
422.11
422.11
390.01
348.67
MHz
64-bit counter
64
0
0
321.85
290.52
261.23
220.5
MHz
0
1
0
317.76
277.62
241.48
205.21
MHz
30
1
0
319.18
278.86
242.54
206.14
MHz
Simple dual-port RAM
128 × 36 bit
0
1
0
290.86
255.55
222.27
188.89
MHz
True dual-port RAM
128 × 18 bit
0
1
0
290.86
255.55
222.27
188.89
MHz
FIFO 128 × 36 bit
34
1
0
290.86
255.55
222.27
188.89
MHz
Single port
RAM 4K × 144 bit
1
1
0
255.95
223.06
194.06
164.93
MHz
Simple dual-port
RAM 4K × 144 bit
0
1
0
255.95
233.06
194.06
164.93
MHz
True dual-port
RAM 4K × 144 bit
0
1
0
255.95
233.06
194.06
164.93
MHz
Single port
RAM 8K × 72 bit
0
1
0
278.94
243.19
211.59
179.82
MHz
Simple dual-port
RAM 8K × 72 bit
0
1
0
255.95
223.06
194.06
164.93
MHz
True dual-port
RAM 8K × 72 bit
0
1
0
255.95
223.06
194.06
164.93
MHz
Single port
RAM 16K × 36 bit
0
1
0
280.66
254.32
221.28
188.00
MHz
Simple dual-port
RAM 16K × 36 bit
0
1
0
269.83
237.69
206.82
175.74
MHz
Simple dual-port RAM
TriMatrix
32 × 18 bit
memory
M512 block
FIFO 32 × 18 bit
TriMatrix
memory
M4K block
Performance
4–20
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–36. Stratix Performance (Part 2 of 2) Notes (1), (2)
Resources Used
Applications
TriMatrix
memory
M-RAM
block
DSP block
Larger
Designs
TriMatrix
DSP
LEs Memory
Blocks
Blocks
Performance
-5
Speed
Grade
-6
Speed
Grade
-7
Speed
Grade
-8
Speed
Grade
Units
True dual-port
RAM 16K × 36 bit
0
1
0
269.83
237.69
206.82
175.74
MHz
Single port
RAM 32K × 18 bit
0
1
0
275.86
244.55
212.76
180.83
MHz
Simple dual-port
RAM 32K × 18 bit
0
1
0
275.86
244.55
212.76
180.83
MHz
True dual-port
RAM 32K × 18 bit
0
1
0
275.86
244.55
212.76
180.83
MHz
Single port
RAM 64K × 9 bit
0
1
0
287.85
253.29
220.36
187.26
MHz
Simple dual-port
RAM 64K × 9 bit
0
1
0
287.85
253.29
220.36
187.26
MHz
True dual-port
RAM 64K × 9 bit
0
1
0
287.85
253.29
220.36
187.26
MHz
9 × 9-bit multiplier (3)
0
0
1
335.0
293.94
255.68
217.24
MHz
18 × 18-bit multiplier
(4)
0
0
1
278.78
237.41
206.52
175.50
MHz
36 × 36-bit multiplier
(4)
0
0
1
148.25
134.71
117.16
99.59
MHz
36 × 36-bit multiplier
(5)
0
0
1
278.78
237.41
206.52
175.5
MHz
18-bit, 4-tap FIR filter
0
0
1
278.78
237.41
206.52
175.50
MHz
8-bit, 16-tap parallel
FIR filter
58
0
4
141.26
133.49
114.88
100.28
MHz
8-bit, 1,024-point FFT
function
870
5
1
261.09
235.51
205.21
175.22
MHz
Notes to Table 4–36:
(1)
(2)
(3)
(4)
(5)
These design performance numbers were obtained using the Quartus II software.
Numbers not listed will be included in a future version of the data sheet.
This application uses registered inputs and outputs.
This application uses registered multiplier input and output stages within the DSP block.
This application uses registered multiplier input, pipeline, and output stages within the DSP
block.
Altera Corporation
January 2006
4–21
Stratix Device Handbook, Volume 1
Timing Model
Internal Timing Parameters
Internal timing parameters are specified on a speed grade basis
independent of device density. Tables 4–37 through 4–42 describe the
Stratix device internal timing microparameters for LEs, IOEs, TriMatrix™
memory structures, DSP blocks, and MultiTrack interconnects.
Table 4–37. LE Internal Timing Microparameter Descriptions
Symbol
Parameter
tSU
LE register setup time before clock
tH
LE register hold time after clock
tCO
LE register clock-to-output delay
tLUT
LE combinatorial LUT delay for data-in to data-out
tCLR
Minimum clear pulse width
tPRE
Minimum preset pulse width
tCLKHL
Register minimum clock high or low time. The maximum core
clock frequency can be calculated by 1/(2 × tCLKHL).
Table 4–38. IOE Internal Timing Microparameter Descriptions
Symbol
Parameter
tSU_R
Row IOE input register setup time
tSU_C
Column IOE input register setup time
tH
IOE input and output register hold time after clock
tCO_R
Row IOE input and output register clock-to-output delay
tC O _ C
Column IOE input and output register clock-to-output delay
tPIN2COMBOUT_R
Row input pin to IOE combinatorial output
tPIN2COMBOUT_C
Column input pin to IOE combinatorial output
tCOMBIN2PIN_R
Row IOE data input to combinatorial output pin
tCOMBIN2PIN_C
Column IOE data input to combinatorial output pin
tCLR
Minimum clear pulse width
tPRE
Minimum preset pulse width
tCLKHL
Register minimum clock high or low time. The maximum I/O
clock frequency can be calculated by 1/(2 × tCLKHL).
Performance may also be affected by I/O timing, use of PLL,
and I/O programmable settings.
4–22
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–39. DSP Block Internal Timing Microparameter Descriptions
Symbol
Altera Corporation
January 2006
Parameter
tSU
Input, pipeline, and output register setup time before clock
tH
Input, pipeline, and output register hold time after clock
tCO
Input, pipeline, and output register clock-to-output delay
tINREG2PIPE9
Input Register to DSP Block pipeline register in 9 × 9-bit
mode
tINREG2PIPE18
Input Register to DSP Block pipeline register in 18 × 18-bit
mode
tPIPE2OUTREG2ADD
DSP Block Pipeline Register to output register delay in TwoMultipliers Adder mode
tPIPE2OUTREG4ADD
DSP Block Pipeline Register to output register delay in FourMultipliers Adder mode
tPD9
Combinatorial input to output delay for 9 × 9
tPD18
Combinatorial input to output delay for 18 × 18
tPD36
Combinatorial input to output delay for 36 × 36
tCLR
Minimum clear pulse width
tCLKHL
Register minimum clock high or low time. This is a limit on
the min time for the clock on the registers in these blocks.
The actual performance is dependent upon the internal
point-to-point delays in the blocks and may give slower
performance as shown in Table 4–36 on page 4–20 and as
reported by the timing analyzer in the Quartus II software.
4–23
Stratix Device Handbook, Volume 1
Timing Model
Table 4–40. M512 Block Internal Timing Microparameter Descriptions
Symbol
Parameter
tM512RC
Synchronous read cycle time
tM512WC
Synchronous write cycle time
tM512WERESU
Write or read enable setup time before clock
tM512WEREH
Write or read enable hold time after clock
tM512CLKENSU
Clock enable setup time before clock
tM512CLKENH
Clock enable hold time after clock
tM512DATASU
Data setup time before clock
tM512DATAH
Data hold time after clock
tM512WADDRSU
Write address setup time before clock
tM512WADDRH
Write address hold time after clock
tM512RADDRSU
Read address setup time before clock
tM512RADDRH
Read address hold time after clock
tM512DATACO1
Clock-to-output delay when using output registers
tM512DATACO2
Clock-to-output delay without output registers
tM512CLKHL
Register minimum clock high or low time. This is a limit on
the min time for the clock on the registers in these blocks.
The actual performance is dependent upon the internal
point-to-point delays in the blocks and may give slower
performance as shown in Table 4–36 on page 4–20 and as
reported by the timing analyzer in the Quartus II software.
tM512CLR
Minimum clear pulse width
Table 4–41. M4K Block Internal Timing Microparameter Descriptions (Part
1 of 2)
Symbol
Parameter
tM4KRC
Synchronous read cycle time
tM4KWC
Synchronous write cycle time
tM4KWERESU
Write or read enable setup time before clock
tM4KWEREH
Write or read enable hold time after clock
tM4KCLKENSU
Clock enable setup time before clock
tM4KCLKENH
Clock enable hold time after clock
tM4KBESU
Byte enable setup time before clock
tM4KBEH
Byte enable hold time after clock
tM4KDATAASU
A port data setup time before clock
4–24
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–41. M4K Block Internal Timing Microparameter Descriptions (Part
2 of 2)
Symbol
Parameter
tM4KDATAAH
A port data hold time after clock
tM4KADDRASU
A port address setup time before clock
tM4KADDRAH
A port address hold time after clock
tM4KDATABSU
B port data setup time before clock
tM4KDATABH
B port data hold time after clock
tM4KADDRBSU
B port address setup time before clock
tM4KADDRBH
B port address hold time after clock
tM4KDATACO1
Clock-to-output delay when using output registers
tM4KDATACO2
Clock-to-output delay without output registers
tM4KCLKHL
Register minimum clock high or low time. This is a limit on
the min time for the clock on the registers in these blocks.
The actual performance is dependent upon the internal
point-to-point delays in the blocks and may give slower
performance as shown inTable 4–36 on page 4–20 and as
reported by the timing analyzer in the Quartus II software.
tM4KCLR
Minimum clear pulse width
Table 4–42. M-RAM Block Internal Timing Microparameter
Descriptions (Part 1 of 2)
Symbol
Altera Corporation
January 2006
Parameter
tMRAMRC
Synchronous read cycle time
tMRAMWC
Synchronous write cycle time
tMRAMWERESU
Write or read enable setup time before clock
tMRAMWEREH
Write or read enable hold time after clock
tMRAMCLKENSU
Clock enable setup time before clock
tMRAMCLKENH
Clock enable hold time after clock
tMRAMBESU
Byte enable setup time before clock
tMRAMBEH
Byte enable hold time after clock
tMRAMDATAASU
A port data setup time before clock
tMRAMDATAAH
A port data hold time after clock
tMRAMADDRASU
A port address setup time before clock
tMRAMADDRAH
A port address hold time after clock
tMRAMDATABSU
B port setup time before clock
4–25
Stratix Device Handbook, Volume 1
Timing Model
Table 4–42. M-RAM Block Internal Timing Microparameter
Descriptions (Part 2 of 2)
Symbol
Parameter
tMRAMDATABH
B port hold time after clock
tMRAMADDRBSU
B port address setup time before clock
tMRAMADDRBH
B port address hold time after clock
tMRAMDATACO1
Clock-to-output delay when using output registers
tMRAMDATACO2
Clock-to-output delay without output registers
tMRAMCLKHL
Register minimum clock high or low time. This is a limit on
the min time for the clock on the registers in these blocks.
The actual performance is dependent upon the internal
point-to-point delays in the blocks and may give slower
performance as shown in Table 4–36 on page 4–20 and as
reported by the timing analyzer in the Quartus II software.
tMRAMCLR
Minimum clear pulse width.
4–26
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Figure 4–3 shows the TriMatrix memory waveforms for the M512, M4K,
and M-RAM timing parameters shown in Tables 4–40 through 4–42.
Figure 4–3. Dual-Port RAM Timing Microparameter Waveform
wrclock
tWEREH
tWERESU
wren
tWADDRH
tWADDRSU
wraddress
an-1
an
a0
a1
a2
a3
a4
a5
a6
din4
din5
din6
tDATAH
data-in
din-1
din
tDATASU
rdclock
tWEREH
tWERESU
rden
tRC
rdaddress
bn
b1
b0
b2
b3
tDATACO1
reg_data-out
doutn-1
doutn-2
doutn
dout0
tDATACO2
unreg_data-out
doutn
doutn-1
dout0
Internal timing parameters are specified on a speed grade basis
independent of device density. Tables 4–44 through 4–50 show the
internal timing microparameters for LEs, IOEs, TriMatrix memory
structures, DSP blocks, and MultiTrack interconnects.
Table 4–43. Routing Delay Internal Timing Microparameter
Descriptions (Part 1 of 2)
Symbol
Altera Corporation
January 2006
Parameter
tR4
Delay for an R4 line with average loading; covers a distance of four
LAB columns.
tR8
Delay for an R8 line with average loading; covers a distance of eight
LAB columns.
tR24
Delay for an R24 line with average loading; covers a distance of 24
LAB columns.
4–27
Stratix Device Handbook, Volume 1
Timing Model
Table 4–43. Routing Delay Internal Timing Microparameter
Descriptions (Part 2 of 2)
Symbol
Parameter
tC4
Delay for a C4 line with average loading; covers a distance of four
LAB rows.
tC8
Delay for a C8 line with average loading; covers a distance of eight
LAB rows.
tC16
Delay for a C16 line with average loading; covers a distance of 16
LAB rows.
tLOCAL
Local interconnect delay, for connections within a LAB, and for the
final routing hop of connections to LABs, DSP blocks, RAM blocks
and I/Os.
Table 4–44. LE Internal Timing Microparameters
-5
-6
-7
-8
Parameter
Unit
Min
Max
Min
Max
Min
Max
Min
Max
tSU
10
10
11
13
ps
tH
100
100
114
135
ps
tCO
156
tLUT
176
366
202
459
238
527
ps
621
ps
tCLR
100
100
114
135
ps
tPRE
100
100
114
135
ps
tCLKHL
1000
1111
1190
1400
ps
Table 4–45. IOE Internal TSU Microparameter by Device Density (Part 1 of 2)
-5
Device
Min
EP1S10
EP1S20
EP1S25
EP1S30
-6
-7
-8
Unit
Symbol
Max
Min
Max
Min
Max
Min
Max
tSU_R
76
80
80
80
ps
tSU_C
176
80
80
80
ps
tSU_R
76
80
80
80
ps
tSU_C
76
80
80
80
ps
tSU_R
276
280
280
280
ps
tSU_C
276
280
280
280
ps
tSU_R
76
80
80
80
ps
tSU_C
176
180
180
180
ps
4–28
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–45. IOE Internal TSU Microparameter by Device Density (Part 2 of 2)
-5
Device
Min
EP1S40
EP1S60
EP1S80
-6
-7
-8
Unit
Symbol
Max
Min
Max
Min
Max
Min
Max
tSU_R
76
80
80
80
ps
tSU_C
376
380
380
380
ps
tSU_R
276
280
280
280
ps
tS U _ C
276
280
280
280
ps
tSU_R
426
430
430
430
ps
tSU_C
76
80
80
80
ps
Table 4–46. IOE Internal Timing Microparameters
-5
-6
-7
-8
Symbol
Unit
Min
tH
Max
68
Min
Max
71
Min
Max
82
Min
Max
96
ps
tCO_R
171
179
206
242
ps
tCO_C
171
179
206
242
ps
tPIN2COMBOUT_R
1,234
1,295
1,490
1,753
ps
tPIN2COMBOUT_C
1,087
1,141
1,312
1,544
ps
tCOMBIN2PIN_R
3,894
4,089
4,089
4,089
ps
tCOMBIN2PIN_C
4,299
4,494
4,494
4,494
ps
tCLR
276
tPRE
tCLKHL
289
333
392
ps
260
273
313
369
ps
1,000
1,111
1,190
1,400
ps
Table 4–47. DSP Block Internal Timing Microparameters (Part 1 of 2)
-5
-6
-7
-8
Symbol
Unit
Min
tSU
0
tH
67
tCO
Max
Min
Max
0
Min
Max
0
75
Min
Max
0
86
ps
101
ps
142
158
181
214
ps
tINREG2PIPE9
2,613
2,982
3,429
4,035
ps
tINREG2PIPE18
3,390
3,993
4,591
5,402
ps
Altera Corporation
January 2006
4–29
Stratix Device Handbook, Volume 1
Timing Model
Table 4–47. DSP Block Internal Timing Microparameters (Part 2 of 2)
-5
-6
-7
-8
Symbol
Unit
Min
Max
Min
Max
Min
Max
Min
Max
tPIPE2OUTREG2ADD
2,002
2,203
2,533
2,980
ps
tPIPE2OUTREG4ADD
2,899
3,189
3,667
4,314
ps
tPD9
3,709
4,081
4,692
5,520
ps
tPD18
4,795
5,275
6,065
7,135
ps
tPD36
7,495
tCLR
tCLKHL
8,245
9,481
11,154
ps
450
500
575
676
ps
1,350
1,500
1,724
2,029
ps
Table 4–48. M512 Block Internal Timing Microparameters
-5
-6
-7
-8
Symbol
Unit
Min
Max
Min
Max
Min
Max
Min
Max
tM512RC
3,340
3,816
4,387
5,162
ps
tM512WC
3,138
3,590
4,128
4,860
ps
tM512WERESU
110
123
141
166
ps
tM512WEREH
34
38
43
51
ps
tM512CLKENSU
215
215
247
290
ps
tM512CLKENH
–70
–70
–81
–95
ps
tM512DATASU
110
123
141
166
ps
tM512DATAH
34
38
43
51
ps
tM512WADDRSU
110
123
141
166
ps
tM512WADDRH
34
38
43
51
ps
tM512RADDRSU
110
123
141
166
ps
tM512RADDRH
34
38
43
51
ps
tM512DATACO1
424
472
541
637
ps
tM512DATACO2
3,366
3,846
4,421
5,203
ps
tM512CLKHL
tM512CLR
1,000
1,111
1,190
1,400
ps
170
189
217
255
ps
4–30
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–49. M4K Block Internal Timing Microparameters
-5
-6
-7
-8
Symbol
Unit
Min
Max
tM4KRC
Min
Max
3,807
tM4KWC
Min
Max
4,320
2,556
Min
Max
4,967
2,840
5,844
3,265
3,842
ps
ps
tM4KWERESU
131
149
171
202
ps
tM4KWEREH
34
38
43
51
ps
tM4KCLKENSU
193
215
247
290
ps
tM4KCLKENH
–63
–70
–81
–95
ps
tM4KBESU
131
149
171
202
ps
tM4KBEH
34
38
43
51
ps
tM4KDATAASU
131
149
171
202
ps
tM4KDATAAH
34
38
43
51
ps
tM4KADDRASU
131
149
171
202
ps
tM4KADDRAH
34
38
43
51
ps
tM4KDATABSU
131
149
171
202
ps
tM4KDATABH
34
38
43
51
ps
tM4KADDRBSU
131
149
171
202
ps
tM4KADDRBH
34
38
43
51
ps
tM4KDATACO1
571
tM4KDATACO2
tM4KCLKHL
tM4KCLR
635
3,984
729
4,507
858
5,182
6,097
ps
ps
1,000
1,111
1,190
1,400
ps
170
189
217
255
ps
Table 4–50. M-RAM Block Internal Timing Microparameters (Part 1 of 2)
-5
-6
-7
-8
Symbol
Unit
Min
tMRAMRC
Max
Min
4,364
tMRAMWC
Max
Min
4,838
3,654
Max
Min
5,562
4,127
Max
6,544
4,746
5,583
ps
ps
tMRAMWERESU
25
25
28
33
ps
tMRAMWEREH
18
20
23
27
ps
tMRAMCLKENSU
99
111
127
150
ps
tMRAMCLKENH
–48
–53
–61
–72
ps
Altera Corporation
January 2006
4–31
Stratix Device Handbook, Volume 1
Timing Model
Table 4–50. M-RAM Block Internal Timing Microparameters (Part 2 of 2)
-5
-6
-7
-8
Symbol
Unit
Min
Max
Min
Max
Min
Max
Min
Max
tMRAMBESU
25
25
28
33
ps
tMRAMBEH
18
20
23
27
ps
tMRAMDATAASU
25
25
28
33
ps
tMRAMDATAAH
18
20
23
27
ps
tMRAMADDRASU
25
25
28
33
ps
tMRAMADDRAH
18
20
23
27
ps
tMRAMDATABSU
25
25
28
33
ps
tMRAMDATABH
18
20
23
27
ps
tMRAMADDRBSU
25
25
28
33
ps
tMRAMADDRBH
18
20
23
27
ps
tMRAMDATACO1
1,038
tMRAMDATACO2
1,053
4,362
tMRAMCLKHL
1,210
4,939
1,424
5,678
ps
6,681
ps
1,000
1,111
1,190
1,400
ps
135
150
172
202
ps
tMRAMCLR
Table 4–51. Routing Delay Internal Timing Parameters
-5
-6
-7
-8
Unit
Symbol
Min
Max
Min
Max
Min
Max
Min
Max
tR 4
268
295
339
390
ps
tR 8
371
349
401
461
ps
tR 2 4
465
512
588
676
ps
tC 4
440
484
557
641
ps
tC 8
577
634
730
840
ps
tC 1 6
445
489
563
647
ps
tL O C A L
313
345
396
455
ps
Routing delays vary depending on the load on that specific routing line.
The Quartus II software reports the routing delay information when
running the timing analysis for a design.
4–32
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
External Timing Parameters
External timing parameters are specified by device density and speed
grade. Figure 4–4 shows the pin-to-pin timing model for bidirectional
IOE pin timing. All registers are within the IOE.
Figure 4–4. External Timing in Stratix Devices
OE Register
D
PRN
Q
Dedicated
Clock
CLRN
Output Register
D
PRN
Q
tINSU
tINH
tOUTCO
tXZ
tZX
Bidirectional
Pin
CLRN
Input Register
D
PRN
Q
CLRN
All external timing parameters reported in this section are defined with
respect to the dedicated clock pin as the starting point. All external I/O
timing parameters shown are for 3.3-V LVTTL I/O standard with the
24-mA current strength and fast slew rate. For external I/O timing using
standards other than LVTTL or for different current strengths, use the I/O
standard input and output delay adders in Tables 4–103 through 4–108.
Altera Corporation
January 2006
4–33
Stratix Device Handbook, Volume 1
Timing Model
Table 4–52 shows the external I/O timing parameters when using fast
regional clock networks.
Table 4–52. Stratix Fast Regional Clock External I/O Timing Parameters
Notes (1), (2)
Symbol
Parameter
tINSU
Setup time for input or bidirectional pin using IOE input register with
fast regional clock fed by FCLK pin
tINH
Hold time for input or bidirectional pin using IOE input register with
fast regional clock fed by FCLK pin
tOUTCO
Clock-to-output delay output or bidirectional pin using IOE output
register with fast regional clock fed by FCLK pin
tXZ
Synchronous IOE output enable register to output pin disable delay
using fast regional clock fed by FCLK pin
tZX
Synchronous IOE output enable register to output pin enable delay
using fast regional clock fed by FCLK pin
Notes to Table 4–52:
(1)
(2)
These timing parameters are sample-tested only.
These timing parameters are for column and row IOE pins. You should use the
Quartus II software to verify the external timing for any pin.
Table 4–53 shows the external I/O timing parameters when using
regional clock networks.
Table 4–53. Stratix Regional Clock External I/O Timing Parameters (Part 1
of 2) Notes (1), (2)
Symbol
Parameter
tINSU
Setup time for input or bidirectional pin using IOE input register with
regional clock fed by CLK pin
tINH
Hold time for input or bidirectional pin using IOE input register with
regional clock fed by CLK pin
tOUTCO
Clock-to-output delay output or bidirectional pin using IOE output
register with regional clock fed by CLK pin
tINSUPLL
Setup time for input or bidirectional pin using IOE input register with
regional clock fed by Enhanced PLL with default phase setting
tINHPLL
Hold time for input or bidirectional pin using IOE input register with
regional clock fed by Enhanced PLL with default phase setting
tOUTCOPLL
Clock-to-output delay output or bidirectional pin using IOE output
register with regional clock Enhanced PLL with default phase setting
4–34
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–53. Stratix Regional Clock External I/O Timing Parameters (Part 2
of 2) Notes (1), (2)
Symbol
Parameter
tXZPLL
Synchronous IOE output enable register to output pin disable delay
using regional clock fed by Enhanced PLL with default phase setting
tZXPLL
Synchronous IOE output enable register to output pin enable delay
using regional clock fed by Enhanced PLL with default phase setting
Notes to Table 4–53:
(1)
(2)
These timing parameters are sample-tested only.
These timing parameters are for column and row IOE pins. You should use the
Quartus II software to verify the external timing for any pin.
Table 4–54 shows the external I/O timing parameters when using global
clock networks.
Table 4–54. Stratix Global Clock External I/O Timing Parameters Notes (1),
(2)
Symbol
Parameter
tINSU
Setup time for input or bidirectional pin using IOE input register with
global clock fed by CLK pin
tINH
Hold time for input or bidirectional pin using IOE input register with
global clock fed by CLK pin
tOUTCO
Clock-to-output delay output or bidirectional pin using IOE output
register with global clock fed by CLK pin
tINSUPLL
Setup time for input or bidirectional pin using IOE input register with
global clock fed by Enhanced PLL with default phase setting
tINHPLL
Hold time for input or bidirectional pin using IOE input register with
global clock fed by Enhanced PLL with default phase setting
tOUTCOPLL
Clock-to-output delay output or bidirectional pin using IOE output
register with global clock Enhanced PLL with default phase setting
tXZPLL
Synchronous IOE output enable register to output pin disable delay
using global clock fed by Enhanced PLL with default phase setting
tZXPLL
Synchronous IOE output enable register to output pin enable delay
using global clock fed by Enhanced PLL with default phase setting
Notes to Table 4–54:
(1)
(2)
Altera Corporation
January 2006
These timing parameters are sample-tested only.
These timing parameters are for column and row IOE pins. You should use the
Quartus II software to verify the external timing for any pin.
4–35
Stratix Device Handbook, Volume 1
Timing Model
Stratix External I/O Timing
These timing parameters are for both column IOE and row IOE pins. In
EP1S30 devices and above, you can decrease the tSU time by using the
FPLLCLK, but may get positive hold time in EP1S60 and EP1S80 devices.
You should use the Quartus II software to verify the external devices for
any pin.
Tables 4–55 through 4–60 show the external timing parameters on column
and row pins for EP1S10 devices.
Table 4–55. EP1S10 External I/O Timing on Column Pins Using Fast Regional Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
Max
Max
tINSU
2.238
2.325
2.668
NA
ns
tINH
0.000
0.000
0.000
NA
ns
tOUTCO
2.240
4.549
2.240
4.836
2.240
5.218
NA
NA
ns
tXZ
2.180
4.423
2.180
4.704
2.180
5.094
NA
NA
ns
tZX
2.180
4.423
2.180
4.704
2.180
5.094
NA
NA
ns
Table 4–56. EP1S10 External I/O Timing on Column Pins Using Regional Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
Min
Min
Min
-8 Speed Grade
Parameter
Unit
Max
Max
2.054
Max
tINSU
1.992
2.359
tINH
0.000
tOUTCO
2.395
4.795
2.395
5.107
2.395
5.527
NA
NA
ns
tXZ
2.335
4.669
2.335
4.975
2.335
5.403
NA
NA
ns
tZX
2.335
4.669
2.335
4.975
2.335
5.403
NA
NA
ns
tINSUPLL
0.975
0.985
1.097
NA
tINHPLL
0.000
0.000
0.000
NA
NA
ns
tOUTCOPLL
1.262
2.636
1.262
2.680
1.262
2.769
NA
NA
ns
tXZPLL
1.202
2.510
1.202
2.548
1.202
2.645
NA
NA
ns
tZXPLL
1.202
2.510
1.202
2.548
1.202
2.645
NA
NA
ns
0.000
4–36
Stratix Device Handbook, Volume 1
NA
0.000
ns
NA
ns
ns
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–57. EP1S10 External I/O Timing on Column Pins Using Global Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
1.692
Max
1.940
Max
tINSU
1.647
tINH
0.000
tOUTCO
2.619
5.184
2.619
5.515
2.619
5.999
NA
NA
ns
tXZ
2.559
5.058
2.559
5.383
2.559
5.875
NA
NA
ns
tZX
2.559
5.058
2.559
5.383
2.559
5.875
NA
NA
ns
tINSUPLL
1.239
1.229
1.374
NA
ns
tINHPLL
0.000
0.000
0.000
NA
ns
tOUTCOPLL
1.109
2.372
1.109
2.436
1.109
2.492
NA
NA
ns
tXZPLL
1.049
2.246
1.049
2.304
1.049
2.368
NA
NA
ns
tZXPLL
1.049
2.246
1.049
2.304
1.049
2.368
NA
NA
ns
0.000
NA
0.000
ns
NA
ns
Table 4–58. EP1S10 External I/O Timing on Row Pin Using Fast Regional Clock Network Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
2.759
Max
tINSU
2.212
tINH
0.000
tOUTCO
2.391
4.838
2.391
5.159
2.391
5.569
NA
NA
ns
tXZ
2.418
4.892
2.418
5.215
2.418
5.637
NA
NA
ns
tZX
2.418
4.892
2.418
5.215
2.418
5.637
NA
NA
ns
Altera Corporation
January 2006
2.403
Max
0.000
NA
0.000
ns
NA
ns
4–37
Stratix Device Handbook, Volume 1
Timing Model
Table 4–59. EP1S10 External I/O Timing on Row Pins Using Regional Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
2.336
Max
2.685
Max
tINSU
2.161
tINH
0.000
tOUTCO
2.434
4.889
2.434
5.226
2.434
5.643
NA
NA
ns
tXZ
2.461
4.493
2.461
5.282
2.461
5.711
NA
NA
ns
tZX
2.461
4.493
2.461
5.282
2.461
5.711
NA
NA
ns
tINSUPLL
1.057
1.172
1.315
NA
ns
tINHPLL
0.000
0.000
0.000
NA
ns
tOUTCOPLL
1.327
2.773
1.327
2.848
1.327
2.940
NA
NA
ns
tXZPLL
1.354
2.827
1.354
2.904
1.354
3.008
NA
NA
ns
tZXPLL
1.354
2.827
1.354
2.904
1.354
3.008
NA
NA
ns
0.000
NA
0.000
ns
NA
ns
Table 4–60. EP1S10 External I/O Timing on Row Pins Using Global Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
1.944
Max
2.232
Max
tINSU
1.787
tINH
0.000
tOUTCO
2.647
5.263
2.647
5.618
2.647
6.069
NA
NA
ns
tXZ
2.674
5.317
2.674
5.674
2.674
6.164
NA
NA
ns
tZX
2.674
5.317
2.674
5.674
2.674
6.164
NA
NA
ns
tINSUPLL
1.371
1.1472
1.654
NA
ns
tINHPLL
0.000
0.000
0.000
NA
ns
tOUTCOPLL
1.144
2.459
1.144
2.548
1.144
2.601
NA
NA
ns
tXZPLL
1.171
2.513
1.171
2.604
1.171
2.669
NA
NA
ns
tZXPLL
1.171
2.513
1.171
2.604
1.171
2.669
NA
NA
ns
0.000
NA
0.000
ns
NA
ns
Note to Tables 4–55 to 4–60:
(1)
Only EP1S25, EP1S30, and EP1S40 have speed grade of -8.
4–38
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Tables 4–61 through 4–66 show the external timing parameters on column
and row pins for EP1S20 devices.
Table 4–61. EP1S20 External I/O Timing on Column Pins Using Fast Regional Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
2.245
Max
2.576
Max
tINSU
2.065
tINH
0.000
tOUTCO
2.283
4.622
2.283
4.916
2.283
5.310
NA
NA
ns
tXZ
2.223
4.496
2.223
4.784
2.223
5.186
NA
NA
ns
tZX
2.223
4.496
2.223
4.784
2.223
5.186
NA
NA
ns
0.000
NA
0.000
ns
NA
ns
Table 4–62. EP1S20 External I/O Timing on Column Pins Using Regional Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
Max
Max
tINSU
1.541
1.680
1.931
NA
ns
tINH
0.000
0.000
0.000
NA
ns
tOUTCO
2.597
5.146
2.597
5.481
2.597
5.955
NA
NA
ns
tXZ
2.537
5.020
2.537
5.349
2.537
5.831
NA
NA
ns
tZX
2.537
5.020
2.537
5.349
2.537
5.831
NA
NA
ns
tINSUPLL
0.777
tINHPLL
0.000
tOUTCOPLL
1.296
2.690
1.296
2.801
1.296
2.876
NA
NA
ns
tXZPLL
1.236
2.564
1.236
2.669
1.236
2.752
NA
NA
ns
tZXPLL
1.236
2.564
1.236
2.669
1.236
2.752
NA
NA
ns
Altera Corporation
January 2006
0.818
0.937
0.000
NA
0.000
ns
NA
ns
4–39
Stratix Device Handbook, Volume 1
Timing Model
Table 4–63. EP1S20 External I/O Timing on Column Pins Using Global Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
1.479
Max
1.699
Max
tINSU
1.351
tINH
0.000
tOUTCO
2.732
5.380
2.732
5.728
2.732
6.240
NA
NA
ns
tXZ
2.672
5.254
2.672
5.596
2.672
6.116
NA
NA
ns
tZX
2.672
5.254
2.672
5.596
2.672
6.116
NA
NA
tINSUPLL
0.923
0.971
1.098
NA
ns
tINHPLL
0.000
0.000
0.000
NA
ns
tOUTCOPLL
1.210
2.544
1.210
2.648
1.210
2.715
NA
NA
ns
tXZPLL
1.150
2.418
1.150
2.516
1.150
2.591
NA
NA
ns
tZXPLL
1.150
2.418
1.150
2.516
1.150
2.591
NA
NA
ns
0.000
NA
0.000
ns
NA
ns
ns
Table 4–64. EP1S20 External I/O Timing on Row Pins Using Fast Regional Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
2.207
Max
2.535
Max
tINSU
2.032
tINH
0.000
tOUTCO
2.492
5.018
2.492
5.355
2.492
5.793
NA
NA
ns
tXZ
2.519
5.072
2.519
5.411
2.519
5.861
NA
NA
ns
tZX
2.519
5.072
2.519
5.411
2.519
5.861
NA
NA
ns
0.000
4–40
Stratix Device Handbook, Volume 1
NA
0.000
ns
NA
ns
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–65. EP1S20 External I/O Timing on Row Pins Using Regional Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
1.967
Max
2.258
Max
tINSU
1.815
tINH
0.000
tOUTCO
2.633
5.235
2.663
5.595
2.663
6.070
NA
NA
ns
tXZ
2.660
5.289
2.660
5.651
2.660
6.138
NA
NA
ns
tZX
2.660
5.289
2.660
5.651
2.660
6.138
NA
NA
tINSUPLL
1.060
1.112
1.277
NA
ns
tINHPLL
0.000
0.000
0.000
NA
ns
tOUTCOPLL
1.325
2.770
1.325
2.908
1.325
2.978
NA
NA
ns
tXZPLL
1.352
2.824
1.352
2.964
1.352
3.046
NA
NA
ns
tZXPLL
1.352
2.824
1.352
2.964
1.352
3.046
NA
NA
ns
0.000
NA
0.000
ns
NA
ns
ns
Table 4–66. EP1S20 External I/O Timing on Row Pins Using Global Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
1.887
Max
2.170
Max
tINSU
1.742
tINH
0.000
tOUTCO
2.674
5.308
2.674
5.675
2.674
6.158
NA
NA
ns
tXZ
2.701
5.362
2.701
5.731
2.701
6.226
NA
NA
ns
tZX
2.701
5.362
2.701
5.731
2.701
6.226
NA
NA
tINSUPLL
1.353
1.418
1.613
NA
ns
tINHPLL
0.000
0.000
0.000
NA
ns
tOUTCOPLL
1.158
2.447
1.158
2.602
1.158
2.642
NA
NA
ns
tXZPLL
1.185
2.531
1.158
2.602
1.185
2.710
NA
NA
ns
tZXPLL
1.185
2.531
1.158
2.602
1.185
2.710
NA
NA
ns
0.000
NA
0.000
ns
NA
ns
ns
Note to Tables 4–61 to 4–66:
(1)
Only EP1S25, EP1S30, and EP1S40 have a speed grade of -8.
Altera Corporation
January 2006
4–41
Stratix Device Handbook, Volume 1
Timing Model
Tables 4–67 through 4–72 show the external timing parameters on column
and row pins for EP1S25 devices.
Table 4–67. EP1S25 External I/O Timing on Column Pins Using Fast Regional Clock Networks
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
2.613
Max
2.968
Max
tINSU
2.412
tINH
0.000
tOUTCO
2.196
4.475
2.196
4.748
2.196
5.118
2.196
5.603
ns
tXZ
2.136
4.349
2.136
4.616
2.136
4.994
2.136
5.488
ns
tZX
2.136
4.349
2.136
4.616
2.136
4.994
2.136
5.488
ns
0.000
3.468
0.000
ns
0.000
ns
Table 4–68. EP1S25 External I/O Timing on Column Pins Using Regional Clock Networks
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
Max
Max
tINSU
1.535
1.661
1.877
2.125
ns
tINH
0.000
0.000
0.000
0.000
ns
tOUTCO
2.739
5.396
2.739
5.746
2.739
6.262
2.739
6.946
ns
tXZ
2.679
5.270
2.679
5.614
2.679
6.138
2.679
6.831
ns
tZX
2.679
5.270
2.679
5.614
2.679
6.138
2.679
6.831
ns
tINSUPLL
0.934
tINHPLL
0.000
tOUTCOPLL
1.316
2.733
1.316
2.839
1.316
2.921
1.316
3.110
ns
tXZPLL
1.256
2.607
1.256
2.707
1.256
2.797
1.256
2.995
ns
tZXPLL
1.256
2.607
1.256
2.707
1.256
2.797
1.256
2.995
ns
0.980
1.092
0.000
4–42
Stratix Device Handbook, Volume 1
1.231
0.000
ns
0.000
ns
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–69. EP1S25 External I/O Timing on Column Pins Using Global Clock Networks
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
1.471
Max
1.657
Max
tINSU
1.371
tINH
0.000
tOUTCO
2.809
5.516
2.809
5.890
2.809
6.429
2.809
7.155
ns
tXZ
2.749
5.390
2.749
5.758
2.749
6.305
2.749
7.040
ns
tZX
2.749
5.390
2.749
5.758
2.749
6.305
2.749
7.040
tINSUPLL
1.271
1.327
1.491
1.677
ns
tINHPLL
0.000
0.000
0.000
0.000
ns
tOUTCOPLL
1.124
2.396
1.124
2.492
1.124
2.522
1.124
2.602
ns
tXZPLL
1.064
2.270
1.064
2.360
1.064
2.398
1.064
2.487
ns
tZXPLL
1.064
2.270
1.064
2.360
1.064
2.398
1.064
2.487
ns
0.000
1.916
0.000
ns
0.000
ns
ns
Table 4–70. EP1S25 External I/O Timing on Row Pins Using Fast Regional Clock Networks
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
Min
Min
Min
-8 Speed Grade
Parameter
Unit
Max
Max
2.990
Min
Max
tINSU
2.429
tINH
0.000
tOUTCO
2.376
4.821
2.376
5.131
2.376
5.538
2.376
6.063
ns
tXZ
2.403
4.875
2.403
5.187
2.403
5.606
2.403
6.145
ns
tZX
2.403
4.875
2.403
5.187
2.403
5.606
2.403
6.145
ns
Altera Corporation
January 2006
2.631
Max
0.000
3.503
0.000
ns
0.000
ns
4–43
Stratix Device Handbook, Volume 1
Timing Model
Table 4–71. EP1S25 External I/O Timing on Row Pins Using Regional Clock Networks
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
1.927
Max
2.182
Max
tINSU
1.793
tINH
0.000
tOUTCO
2.759
5.457
2.759
5.835
2.759
6.346
2.759
7.024
ns
tXZ
2.786
5.511
2.786
5.891
2.786
6.414
2.786
7.106
ns
tZX
2.786
5.511
2.786
5.891
2.786
6.414
2.786
7.106
tINSUPLL
1.169
1.221
1.373
1.600
ns
tINHPLL
0.000
0.000
0.000
0.000
ns
tOUTCOPLL
1.375
2.861
1.375
2.999
1.375
3.082
1.375
3.174
ns
tXZPLL
1.402
2.915
1.402
3.055
1.402
3.150
1.402
3.256
ns
tZXPLL
1.402
2.915
1.402
3.055
1.402
3.150
1.402
3.256
ns
0.000
2.542
0.000
ns
0.000
ns
ns
Table 4–72. EP1S25 External I/O Timing on Row Pins Using Global Clock Networks
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
1.779
Max
2.012
Max
tINSU
1.665
tINH
0.000
tOUTCO
2.834
5.585
2.834
5.983
2.834
6.516
2.834
7.194
ns
tXZ
2.861
5.639
2.861
6.039
2.861
6.584
2.861
7.276
ns
tZX
2.861
5.639
2.861
6.039
2.861
6.584
2.861
7.276
tINSUPLL
1.538
1.606
1.816
2.121
ns
tINHPLL
0.000
0.000
0.000
0.000
ns
tOUTCOPLL
1.164
2.492
1.164
2.614
1.164
2.639
1.164
2.653
ns
tXZPLL
1.191
2.546
1.191
2.670
1.191
2.707
1.191
2.735
ns
tZXPLL
1.191
2.546
1.191
2.670
1.191
2.707
1.191
2.735
ns
0.000
4–44
Stratix Device Handbook, Volume 1
2.372
0.000
ns
0.000
ns
ns
Altera Corporation
January 2006
DC & Switching Characteristics
Tables 4–73 through 4–78 show the external timing parameters on column
and row pins for EP1S30 devices.
Table 4–73. EP1S30 External I/O Timing on Column Pins Using Fast Regional Clock Networks
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Unit
Parameter
Max
Max
2.680
Max
3.062
Max
tINSU
2.502
tINH
0.000
tOUTCO
2.473
4.965
2.473
5.329
2.473
5.784
2.473
6.392
ns
tXZ
2.413
4.839
2.413
5.197
2.413
5.660
2.413
6.277
ns
tZX
2.413
4.839
2.413
5.197
2.413
5.660
2.413
6.277
ns
0.000
3.591
0.000
ns
0.000
ns
Table 4–74. EP1S30 External I/O Timing on Column Pins Using Regional Clock Networks
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
Max
Max
tINSU
2.286
2.426
2.769
3.249
ns
tINH
0.000
0.000
0.000
0.000
ns
tOUTCO
2.641
5.225
2.641
5.629
2.641
6.130
2.641
6.796
ns
tXZ
2.581
5.099
2.581
5.497
2.581
6.006
2.581
6.681
ns
tZX
2.581
5.099
2.581
5.497
2.581
6.006
2.581
6.681
ns
tINSUPLL
1.200
tINHPLL
0.000
tOUTCOPLL
1.108
2.367
1.108
2.534
1.108
2.569
1.108
2.517
ns
tXZPLL
1.048
2.241
1.048
2.402
1.048
2.445
1.048
2.402
ns
tZXPLL
1.048
2.241
1.048
2.402
1.048
2.445
1.048
2.402
ns
1.185
1.344
0.000
1.662
0.000
ns
0.000
Table 4–75. EP1S30 External I/O Timing on Column Pins Using Global Clock Networks
ns
(Part 1 of 2)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
Max
Max
tINSU
1.935
2.029
2.310
2.709
ns
tINH
0.000
0.000
0.000
0.000
ns
tOUTCO
2.814
Altera Corporation
January 2006
5.532
2.814
5.980
2.814
6.536
2.814
7.274
ns
4–45
Stratix Device Handbook, Volume 1
Timing Model
Table 4–75. EP1S30 External I/O Timing on Column Pins Using Global Clock Networks
(Part 2 of 2)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Max
Min
Max
Min
Max
Min
Max
tXZ
2.754
5.406
2.754
5.848
2.754
6.412
2.754
7.159
tZX
2.754
5.406
2.754
5.848
2.754
6.412
2.754
7.159
tINSUPLL
1.265
1.236
1.403
1.756
ns
tINHPLL
0.000
0.000
0.000
0.000
ns
tOUTCOPLL
1.068
2.302
1.068
2.483
1.068
2.510
1.068
2.423
ns
tXZPLL
1.008
2.176
1.008
2.351
1.008
2.386
1.008
2.308
ns
tZXPLL
1.008
2.176
1.008
2.351
1.008
2.386
1.008
2.308
ns
Parameter
Unit
ns
ns
Table 4–76. EP1S30 External I/O Timing on Row Pins Using Fast Regional Clock Networks
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameters
Unit
Max
Max
Max
Max
tINSU
2.616
2.808
3.223
3.797
ns
tINH
0.000
0.000
0.000
0.000
ns
tOUTCO
2.542
5.114
2.542
5.502
2.542
5.965
2.542
6.581
ns
tXZ
2.569
5.168
2.569
5.558
2.569
6.033
2.569
6.663
ns
tZX
2.569
5.168
2.569
5.558
2.569
6.033
2.569
6.663
ns
4–46
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–77. EP1S30 External I/O Timing on Row Pins Using Regional Clock Networks
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
2.467
Max
2.828
Max
tINSU
2.322
tINH
0.000
tOUTCO
2.731
5.408
2.731
5.843
2.731
6.360
2.731
7.036
ns
tXZ
2.758
5.462
2.758
5.899
2.758
6.428
2.758
7.118
ns
tZX
2.758
5.462
2.758
5.899
2.758
6.428
2.758
7.118
tINSUPLL
1.291
1.283
1.469
1.832
ns
tINHPLL
0.000
0.000
0.000
0.000
ns
tOUTCOPLL
1.192
2.539
1.192
2.737
1.192
2.786
1.192
2.742
ns
tXZPLL
1.219
2.539
1.219
2.793
1.219
2.854
1.219
2.824
ns
tZXPLL
1.219
2.539
1.219
2.793
1.219
2.854
1.219
2.824
ns
0.000
3.342
0.000
ns
0.000
ns
ns
Table 4–78. EP1S30 External I/O Timing on Row Pins Using Global Clock Networks
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
2.398
Max
tINSU
1.995
tINH
0.000
tOUTCO
2.917
5.735
2.917
6.221
2.917
6.790
2.917
7.548
ns
tXZ
2.944
5.789
2.944
6.277
2.944
6.858
2.944
7.630
ns
tZX
2.944
5.789
2.944
6.277
2.944
6.858
2.944
7.630
tINSUPLL
1.337
1.312
1.508
1.902
ns
tINHPLL
0.000
0.000
0.000
0.000
ns
tOUTCOPLL
1.164
2.493
1.164
2.708
1.164
2.747
1.164
2.672
ns
tXZPLL
1.191
2.547
1.191
2.764
1.191
2.815
1.191
2.754
ns
tZXPLL
1.191
2.547
1.191
2.764
1.191
2.815
1.191
2.754
ns
Altera Corporation
January 2006
2.089
Max
0.000
2.830
0.000
ns
0.000
ns
ns
4–47
Stratix Device Handbook, Volume 1
Timing Model
Tables 4–79 through 4–84 show the external timing parameters on column
and row pins for EP1S40 devices.
Table 4–79. EP1S40 External I/O Timing on Column Pins Using Fast Regional Clock Networks
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
2.907
Max
3.290
Max
tINSU
2.696
tINH
0.000
tOUTCO
2.506
5.015
2.506
5.348
2.506
5.809
2.698
7.286
ns
tXZ
2.446
4.889
2.446
5.216
2.446
5.685
2.638
7.171
ns
tZX
2.446
4.889
2.446
5.216
2.446
5.685
2.638
7.171
ns
0.000
2.899
0.000
ns
0.000
ns
Table 4–80. EP1S40 External I/O Timing on Column Pins Using Regional Clock Networks
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
Max
Max
tINSU
2.413
2.581
2.914
2.938
ns
tINH
0.000
0.000
0.000
0.000
ns
tOUTCO
2.668
5.254
2.668
5.628
2.668
6.132
2.869
7.307
ns
tXZ
2.608
5.128
2.608
5.496
2.608
6.008
2.809
7.192
ns
tZX
2.608
5.128
2.608
5.496
2.608
6.008
2.809
7.192
ns
tINSUPLL
1.385
tINHPLL
0.000
tOUTCOPLL
1.117
2.382
1.117
2.552
1.117
2.504
1.117
2.542
ns
tXZPLL
1.057
2.256
1,057
2.420
1.057
2.380
1.057
2.427
ns
tZXPLL
1.057
2.256
1,057
2.420
1.057
2.380
1.057
2.427
ns
1.376
1.609
0.000
4–48
Stratix Device Handbook, Volume 1
1.837
0.000
ns
0.000
ns
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–81. EP1S40 External I/O Timing on Column Pins Using Global Clock Networks
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
2.268
Max
2.558
Max
tINSU
2.126
tINH
0.000
tOUTCO
2.856
5.585
2.856
5.987
2.856
6.541
2.847
7.253
ns
tXZ
2.796
5.459
2.796
5.855
2.796
6.417
2.787
7.138
ns
tZX
2.796
5.459
2.796
5.855
2.796
6.417
2.787
7.138
tINSUPLL
1.466
1.455
1.711
1.906
ns
tINHPLL
0.000
0.000
0.000
0.000
ns
tOUTCOPLL
1.092
2.345
1.092
2.510
1.092
2.455
1.089
2.473
ns
tXZPLL
1.032
2.219
1.032
2.378
1.032
2.331
1.029
2.358
ns
tZXPLL
1.032
2.219
1.032
2.378
1.032
2.331
1.029
2.358
ns
0.000
2.930
0.000
ns
0.000
ns
ns
Table 4–82. EP1S40 External I/O Timing on Row Pins Using Fast Regional Clock Networks
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
3.083
Max
tINSU
2.472
tINH
0.000
tOUTCO
2.631
5.258
2.631
5.625
2.631
6.105
2.745
7.324
ns
tXZ
2.658
5.312
2.658
5.681
2.658
6.173
2.772
7.406
ns
tZX
2.658
5.312
2.658
5.681
2.658
6.173
2.772
7.406
ns
Altera Corporation
January 2006
2.685
Max
0.000
3.056
0.000
ns
0.000
ns
4–49
Stratix Device Handbook, Volume 1
Timing Model
Table 4–83. EP1S40 External I/O Timing on Row Pins Using Regional Clock Networks
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
2.526
Max
2.898
Max
tINSU
2.349
tINH
0.000
tOUTCO
2.725
5.381
2.725
5.784
2.725
6.290
2.725
7.426
ns
tXZ
2.752
5.435
2.752
5.840
2.752
6.358
2.936
7.508
ns
tZX
2.752
5.435
2.752
5.840
2.752
6.358
2.936
7.508
tINSUPLL
1.328
1.322
1.605
1.883
ns
tINHPLL
0.000
0.000
0.000
0.000
ns
tOUTCOPLL
1.169
2.502
1.169
2.698
1.169
2.650
1.169
2.691
ns
tXZPLL
1.196
2.556
1.196
2.754
1.196
2.718
1.196
2.773
ns
tZXPLL
1.196
2.556
1.196
2.754
1.196
2.718
1.196
2.773
ns
0.000
2.952
0.000
ns
0.000
ns
ns
Table 4–84. EP1S40 External I/O Timing on Row Pins Using Global Clock Networks
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
2.171
Max
2.491
Max
tINSU
2.020
tINH
0.000
tOUTCO
2.912
5.710
2.912
6.139
2.912
6.697
2.931
7.480
ns
tXZ
2.939
5.764
2.939
6.195
2.939
6.765
2.958
7.562
ns
tZX
2.939
5.764
2.939
6.195
2.939
6.765
2.958
7.562
tINSUPLL
1.370
1.368
1.654
1.881
ns
tINHPLL
0.000
0.000
0.000
0.000
ns
tOUTCOPLL
1.144
2.460
1.144
2.652
1.144
2.601
1.170
2.693
ns
tXZPLL
1.171
2.514
1.171
2.708
1.171
2.669
1.197
2.775
ns
tZXPLL
1.171
2.514
1.171
2.708
1.171
2.669
1.197
2.775
ns
0.000
4–50
Stratix Device Handbook, Volume 1
2.898
0.000
ns
0.000
ns
ns
Altera Corporation
January 2006
DC & Switching Characteristics
Tables 4–85 through 4–90 show the external timing parameters on column
and row pins for EP1S60 devices.
Table 4–85. EP1S60 External I/O Timing on Column Pins Using Fast Regional Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
3.277
Max
3.733
Max
tINSU
3.029
tINH
0.000
tOUTCO
2.446
4.871
2.446
5.215
2.446
5.685
NA
NA
ns
tXZ
2.386
4.745
2.386
5.083
2.386
5.561
NA
NA
ns
tZX
2.386
4.745
2.386
5.083
2.386
5.561
NA
NA
ns
0.000
NA
0.000
ns
NA
ns
Table 4–86. EP1S60 External I/O Timing on Column Pins Using Regional Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
Max
Max
tINSU
2.491
2.691
3.060
NA
ns
tINH
0.000
0.000
0.000
NA
ns
tOUTCO
2.767
5.409
2.767
5.801
2.767
6.358
NA
NA
ns
tXZ
2.707
5.283
2.707
5.669
2.707
6.234
NA
NA
ns
tZX
2.707
5.283
2.707
5.669
2.707
6.234
NA
NA
ns
tINSUPLL
1.233
tINHPLL
0.000
tOUTCOPLL
1.078
2.278
1.078
2.395
1.078
2.428
NA
NA
ns
tXZPLL
1.018
2.152
1.018
2.263
1.018
2.304
NA
NA
ns
tZXPLL
1.018
2.152
1.018
2.263
1.018
2.304
NA
NA
ns
Altera Corporation
January 2006
1.270
1.438
0.000
NA
0.000
ns
NA
ns
4–51
Stratix Device Handbook, Volume 1
Timing Model
Table 4–87. EP1S60 External I/O Timing on Column Pins Using Global Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
2.152
Max
2.441
Max
tINSU
2.000
tINH
0.000
tOUTCO
3.051
5.900
3.051
6.340
3.051
6.977
NA
NA
ns
tXZ
2.991
5.774
2.991
6.208
2.991
6.853
NA
NA
ns
tZX
2.991
5.774
2.991
6.208
2.991
6.853
NA
NA
tINSUPLL
1.315
1.362
1.543
NA
ns
tINHPLL
0.000
0.000
0.000
NA
ns
tOUTCOPLL
1.029
2.196
1.029
2.303
1.029
2.323
NA
NA
ns
tXZPLL
0.969
2.070
0.969
2.171
0.969
2.199
NA
NA
ns
tZXPLL
0.969
2.070
0.969
2.171
0.969
2.199
NA
NA
ns
0.000
NA
0.000
ns
NA
ns
ns
Table 4–88. EP1S60 External I/O Timing on Row Pins Using Fast Regional Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
3.393
Max
3.867
Max
tINSU
3.144
tINH
0.000
tOUTCO
2.643
5.275
2.643
5.654
2.643
6.140
NA
NA
ns
tXZ
2.670
5.329
2.670
5.710
2.670
6.208
NA
NA
ns
tZX
2.670
5.329
2.670
5.710
2.670
6.208
NA
NA
ns
0.000
4–52
Stratix Device Handbook, Volume 1
NA
0.000
ns
NA
ns
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–89. EP1S60 External I/O Timing on Row Pins Using Regional Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
2.990
Max
3.407
Max
tINSU
2.775
tINH
0.000
tOUTCO
2.867
5.644
2.867
6.057
2.867
6.600
NA
NA
ns
tXZ
2.894
5.698
2.894
6.113
2.894
6.668
NA
NA
ns
tZX
2.894
5.698
2.894
6.113
2.894
6.668
NA
NA
tINSUPLL
1.523
1.577
1.791
NA
ns
tINHPLL
0.000
0.000
0.000
NA
ns
tOUTCOPLL
1.174
2.507
1.174
2.643
1.174
2.664
NA
NA
ns
tXZPLL
1.201
2.561
1.201
2.699
1.201
2.732
NA
NA
ns
tZXPLL
1.201
2.561
1.201
2.699
1.201
2.732
NA
NA
ns
0.000
NA
0.000
ns
NA
ns
ns
Table 4–90. EP1S60 External I/O Timing on Row Pins Using Global Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
2.393
Max
2.721
Max
tINSU
2.232
tINH
0.000
tOUTCO
3.182
6.187
3.182
6.654
3.182
7.286
NA
NA
ns
tXZ
3.209
6.241
3.209
6.710
3.209
7.354
NA
NA
ns
tZX
3.209
6.241
3.209
6.710
3.209
7.354
NA
NA
tINSUPLL
1.651
1.612
1.833
NA
ns
tINHPLL
0.000
0.000
0.000
NA
ns
tOUTCOPLL
1.154
2.469
1.154
2.608
1.154
2.622
NA
NA
ns
tXZPLL
1.181
2.523
1.181
2.664
1.181
2.690
NA
NA
ns
tZXPLL
1.181
2.523
1.181
2.664
1.181
2.690
NA
NA
ns
0.000
NA
0.000
ns
NA
ns
ns
Note to Tables 4–85 to 4–90:
(1)
Only EP1S25, EP1S30, and EP1S40 devices have the -8 speed grade.
Altera Corporation
January 2006
4–53
Stratix Device Handbook, Volume 1
Timing Model
Tables 4–91 through 4–96 show the external timing parameters on column
and row pins for EP1S80 devices.
Table 4–91. EP1S80 External I/O Timing on Column Pins Using Fast Regional Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
2.528
Max
2.900
Max
tINSU
2.328
tINH
0.000
tOUTCO
2.422
4.830
2.422
5.169
2.422
5.633
NA
NA
ns
tXZ
2.362
4.704
2.362
5.037
2.362
5.509
NA
NA
ns
tZX
2.362
4.704
2.362
5.037
2.362
5.509
NA
NA
ns
0.000
NA
0.000
ns
NA
ns
Table 4–92. EP1S80 External I/O Timing on Column Pins Using Regional Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
Max
Max
tINSU
1.760
1.912
2.194
NA
ns
tINH
0.000
0.000
0.000
NA
ns
tOUTCO
2.761
5.398
2.761
5.785
2.761
6.339
NA
NA
ns
tXZ
2.701
5.272
2.701
5.653
2.701
6.215
NA
NA
ns
tZX
2.701
5.272
2.701
5.653
2.701
6.215
NA
NA
ns
tINSUPLL
0.462
tINHPLL
0.000
tOUTCOPLL
1.661
2.849
1.661
2.859
1.661
2.881
NA
NA
ns
tXZPLL
1.601
2.723
1.601
2.727
1.601
2.757
NA
NA
ns
tZXPLL
1.601
2.723
1.601
2.727
1.601
2.757
NA
NA
ns
0.606
0.785
0.000
4–54
Stratix Device Handbook, Volume 1
NA
0.000
ns
NA
ns
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–93. EP1S80 External I/O Timing on Column Pins Using Global Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Parameter
Unit
Max
Max
0.976
Max
1.118
Max
tINSU
0.884
tINH
0.000
tOUTCO
3.267
6.274
3.267
6.721
3.267
7.415
NA
NA
ns
tXZ
3.207
6.148
3.207
6.589
3.207
7.291
NA
NA
ns
tZX
3.207
6.148
3.207
6.589
3.207
7.291
NA
NA
tINSUPLL
0.506
0.656
0.838
NA
ns
tINHPLL
0.000
0.000
0.000
NA
ns
tOUTCOPLL
1.635
2.805
1.635
2.809
1.635
2.828
NA
NA
ns
tXZPLL
1.575
2.679
1.575
2.677
1.575
2.704
NA
NA
ns
tZXPLL
1.575
2.679
1.575
2.677
1.575
2.704
NA
NA
ns
0.000
NA
0.000
ns
NA
ns
ns
Table 4–94. EP1S80 External I/O Timing on Row Pins Using Fast Regional Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Parameter
Unit
Min
Max
Min
Max
Min
Max
Min
Max
tINSU
2.792
2.993
3.386
NA
ns
tINH
0.000
0.000
0.000
NA
ns
tOUTCO
2.619
5.235
2.619
5.609
2.619
6.086
NA
NA
ns
tXZ
2.646
5.289
2.646
5.665
2.646
6.154
NA
NA
ns
tZX
2.646
5.289
2.646
5.665
2.646
6.154
NA
NA
ns
Altera Corporation
January 2006
4–55
Stratix Device Handbook, Volume 1
Timing Model
Table 4–95. EP1S80 External I/O Timing on Row Pins Using Regional Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Parameter
Unit
Min
Max
Min
Max
Min
Max
Min
Max
tINSU
2.295
2.454
2.767
NA
ns
tINH
0.000
0.000
0.000
NA
ns
tOUTCO
2.917
5.732
2.917
6.148
2.917
6.705
NA
NA
ns
tXZ
2.944
5.786
2.944
6.204
2.944
6.773
NA
NA
ns
tZX
2.944
5.786
2.944
6.204
2.944
6.773
NA
NA
ns
tINSUPLL
1.011
tINHPLL
0.000
tOUTCOPLL
1.808
3.169
1.808
3.209
1.808
3.233
NA
NA
ns
tXZPLL
1.835
3.223
1.835
3.265
1.835
3.301
NA
NA
ns
tZXPLL
1.835
3.223
1.835
3.265
1.835
3.301
NA
NA
ns
1.161
1.372
0.000
NA
0.000
ns
NA
ns
Table 4–96. EP1S80 External I/O Timing on Rows Using Pin Global Clock Networks Note (1)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Min
Min
Min
Min
Symbol
Unit
Max
Max
Max
Max
tINSU
1.362
1.451
1.613
NA
ns
tINH
0.000
0.000
0.000
NA
ns
tOUTCO
3.457
6.665
3.457
7.151
3.457
7.859
NA
NA
ns
tXZ
3.484
6.719
3.484
7.207
3.484
7.927
NA
NA
ns
tZX
3.484
6.719
3.484
7.207
3.484
7.927
NA
NA
ns
tINSUPLL
o.994
tINHPLL
0.000
tOUTCOPLL
1.821
3.186
1.821
3.227
1.821
3.254
NA
NA
ns
tXZPLL
1.848
3.240
1.848
3.283
1.848
3.322
NA
NA
ns
tZXPLL
1.848
3.240
1.848
3.283
1.848
3.322
NA
NA
ns
1.143
1.351
0.000
NA
0.000
ns
NA
ns
Note to Tables 4–91 to 4–96:
(1)
Only EP1S25, EP1S30, and EP1S40 devices have the -8 speed grade.
4–56
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Definition of I/O Skew
I/O skew is defined as the absolute value of the worst-case difference in
clock-to-out times (tCO) between any two output registers fed by a
common clock source.
I/O bank skew is made up of the following components:
■
■
Clock network skews: This is the difference between the arrival times
of the clock at the clock input port of the two IOE registers.
Package skews: This is the package trace length differences between
(I/O pad A to I/O pin A) and (I/O pad B to I/O pin B).
Figure 4–5 shows an example of two IOE registers located in the same
bank, being fed by a common clock source. The clock can come from an
input pin or from a PLL output.
Figure 4–5. I/O Skew within an I/O Bank
I/O Bank
I/O Pin A
Common Source of GCLK
I/O Pin B
Fast Edge
I/O Pin A
Slow Edge
I/O Pin B
I/O Skew
Altera Corporation
January 2006
I/O Skew
4–57
Stratix Device Handbook, Volume 1
Timing Model
Figure 4–6 shows the case where four IOE registers are located in two
different I/O banks.
Figure 4–6. I/O Skew Across Two I/O Banks
I/O Bank
I/O Pin A
I/O Pin B
Common Source of GCLK
I/O Pin C
I/O Pin D
I/O Bank
I/O Pin Skew across
two Banks
I/O Pin A
I/O Pin B
I/O Pin C
I/O Pin D
Table 4–97 defines the timing parameters used to define the timing for
horizontal I/O pins (side banks 1, 2, 5, 6) and vertical I/O pins (top and
bottom banks 3, 4, 7, 8). The timing parameters define the skew within an
I/O bank, across two neighboring I/O banks on the same side of the
device, across all horizontal I/O banks, across all vertical I/O banks, and
the skew for the overall device.
Table 4–97. Output Pin Timing Skew Definitions (Part 1 of 2)
Symbol
Definition
tSB_HIO
Row I/O (HIO) within one I/O bank (1)
tSB_VIO
Column I/O (VIO) within one I/O bank (1)
tSS_HIO
Row I/O (HIO) same side of the device, across two
banks (2)
tSS_VIO
Column I/O (VIO) same side of the device, across two
banks (2)
4–58
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–97. Output Pin Timing Skew Definitions (Part 2 of 2)
Symbol
Definition
tLR_HIO
Across all HIO banks (1, 2, 5, 6); across four similar
type I/O banks
tTB_VIO
Across all VIO banks (3, 4, 7, 8); across four similar
type I/O banks
tOVERALL
Output timing skew for all I/O pins on the device.
Notes to Table 4–97:
(1)
(2)
See Figure 4–5 on page 4–57.
See Figure 4–6 on page 4–58.
Table 4–98 shows the I/O skews when using the same global or regional
clock to feed IOE registers in I/O banks around each device. These values
can be used for calculating the timing budget on the output (write) side
of a memory interface. These values already factor in the package skew.
Table 4–98. Output Skew for Stratix by Device Density
Skew (ps) (1)
Symbol
EP1S10 to EP1S30
EP1S40
EP1S60 & EP1S80
tSB_HIO
90
290
500
tSB_VIO
160
290
500
tSS_HIO
90
460
600
tSS_VIO
180
520
630
tLR_HIO
150
490
600
tTB_VIO
190
580
670
tOVERALL
430
630
880
Note to Table 4–98:
(1)
Altera Corporation
January 2006
The skew numbers in Table 4–98 account for worst case package skews.
4–59
Stratix Device Handbook, Volume 1
Timing Model
Skew on Input Pins
Table 4–99 shows the package skews that were considered to get the
worst case I/O skew value. You can use these values, for example, when
calculating the timing budget on the input (read) side of a memory
interface.
Table 4–99. Package Skew on Input Pins
Package Parameter
Pins in the same I/O bank
Worst-Case Skew (ps)
50
Pins in top/bottom (vertical I/O) banks
50
Pins in left/right side (horizontal I/O) banks
50
Pins across the entire device
100
PLL Counter & Clock Network Skews
Table 4–100 shows the clock skews between different clock outputs from
the Stratix device PLL.
Table 4–100. PLL Counter & Clock Network Skews
Parameter
Worst-Case Skew (ps)
Clock skew between two external clock outputs driven
by the same counter
100
Clock skew between two external clock outputs driven
by the different counters with the same settings
150
Dual-purpose PLL dedicated clock output used as I/O
pin vs. regular I/O pin
270 (1)
Clock skew between any two outputs of the PLL that
drive global clock networks
150
Note to Table 4–100:
(1)
The Quartus II software models 270 ps of delay on the PLL dedicated clock
output (PLL6_OUT[3..0]p/n and PLL5_OUT[3..0]p/n) pins both when
used as clocks and when used as I/O pins.
I/O Timing Measurement Methodology
Different I/O standards require different baseline loading techniques for
reporting timing delays. Altera characterizes timing delays with the
required termination and loading for each I/O standard. The timing
information is specified from the input clock pin up to the output pin of
4–60
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
the FPGA device. The Quartus II software calculates the I/O timing for
each I/O standard with a default baseline loading as specified by the I/O
standard.
Altera measures clock-to-output delays (tCO) at worst-case process,
minimum voltage, and maximum temperature (PVT) for the 3.3-V LVTTL
I/O standard with 24 mA (default case) current drive strength setting and
fast slew rate setting. I/O adder delays are measured to calculate the tCO
change at worst-case PVT across all I/O standards and current drive
strength settings with the default loading shown in Table 4–101 on
page 4–62. Timing derating data for additional loading is taken for tCO
across worst-case PVT for all I/O standards and drive strength settings.
These three pieces of data are used to predict the timing at the output pin.
tCO at pin = tOUTCO max for 3.3-V 24 mA LVTTL + I/O Adder +
Output Delay Adder for Loading
Simulation using IBIS models is required to determine the delays on the
PCB traces in addition to the output pin delay timing reported by the
Quartus II software and the timing model in the device handbook.
1.
Simulate the output driver of choice into the generalized test setup
using values from Table 4–101 on page 4–62.
2.
Record the time to VMEAS.
3.
Simulate the output driver of choice into the actual PCB trace and
load, using the appropriate IBIS input buffer model or an equivalent
capacitance value to represent the load.
4.
Record the time to VMEAS.
5.
Compare the results of steps 2 and 4. The increase or decrease in
delay should be added to or subtracted from the I/O Standard
Output Adder delays to yield the actual worst-case propagation
delay (clock-to-input) of the PCB trace.
The Quartus II software reports maximum timing with the conditions
shown in Table 4–101 on page 4–62 using the proceeding equation.
Figure 4–7 on page 4–62 shows the model of the circuit that is represented
by the Quartus II output timing.
Altera Corporation
January 2006
4–61
Stratix Device Handbook, Volume 1
Timing Model
Figure 4–7. Output Delay Timing Reporting Setup Modeled by Quartus II
VCCIO
Single-Ended Outputs
VCCIO
VTT
RUP
Output
Buffer
RT
RS
OUTPUT
VMEAS
CL
RDN
GND
GND
GND
Notes to Figure 4–7:
(1)
(2)
Output pin timing is reported at the output pin of the FPGA device. Additional
delays for loading and board trace delay need to be accounted for with IBIS model
simulations.
VCCINT is 1.42-V unless otherwise specified.
Table 4–101. Reporting Methodology For Maximum Timing For Single-Ended Output Pins (Part 1 of 2)
Notes (1), (2), (3)
Measurement
Point
Loading and Termination
I/O Standard
RUP
RDN
RS
RT
Ω
Ω
Ω
3.3-V LVTTL
–
–
2.5-V LVTTL
–
1.8-V LVTTL
–
Ω
VCCIO
(V)
VTT
(V)
CL
(pF)
VMEAS
0
–
2.950
2.95
10
1.500
–
0
–
2.370
2.37
10
1.200
–
0
–
1.650
1.65
10
0.880
1.5-V LVTTL
–
–
0
–
1.400
1.40
10
0.750
3.3-V LVCMOS
–
–
0
–
2.950
2.95
10
1.500
2.5-V LVCMOS
–
–
0
–
2.370
2.37
10
1.200
1.8-V LVCMOS
–
–
0
–
1.650
1.65
10
0.880
1.5-V LVCMOS
–
–
0
–
1.400
1.40
10
0.750
3.3-V GTL
–
–
0
25
2.950
1.14
30
0.740
2.5-V GTL
–
–
0
25
2.370
1.14
30
0.740
3.3-V GTL+
–
–
0
25
2.950
1.35
30
0.880
2.5-V GTL+
–
–
0
25
2.370
1.35
30
0.880
3.3-V SSTL-3 Class II
–
–
25
25
2.950
1.25
30
1.250
4–62
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–101. Reporting Methodology For Maximum Timing For Single-Ended Output Pins (Part 2 of 2)
Notes (1), (2), (3)
Measurement
Point
Loading and Termination
I/O Standard
RUP
RDN
RS
RT
Ω
Ω
Ω
Ω
VCCIO
(V)
VTT
(V)
CL
(pF)
VMEAS
3.3-V SSTL-3 Class I
–
–
25
50
2.950
1.250
30
1.250
2.5-V SSTL-2 Class II
–
–
25
25
2.370
1.110
30
1.110
2.5-V SSTL-2 Class I
–
–
25
50
2.370
1.110
30
1.110
1.8-V SSTL-18 Class II
–
–
25
25
1.650
0.760
30
0.760
1.8-V SSTL-18 Class I
–
–
25
50
1.650
0.760
30
0.760
1.5-V HSTL Class II
–
–
0
25
1.400
0.700
20
0.680
1.5-V HSTL Class I
–
–
0
50
1.400
0.700
20
0.680
1.8-V HSTL Class II
–
–
0
25
1.650
0.700
20
0.880
1.8-V HSTL Class I
–
–
0
50
1.650
0.700
20
0.880
3.3-V PCI (4)
–/25
25/–
0
–
2.950
2.950
10
0.841/1.814
3.3-V PCI-X 1.0 (4)
–/25
25/–
0
–
2.950
2.950
10
0.841/1.814
3.3-V Compact PCI (4)
–/25
25/–
0
–
2.950
2.950
10
0.841/1.814
3.3-V AGP 1X (4)
–/25
25/–
0
–
2.950
2.950
10
0.841/1.814
–
–
25
50
2.050
1.350
30
1.350
3.3-V CTT
Notes to Table 4–101:
(1)
(2)
(3)
(4)
Input measurement point at internal node is 0.5 × VCCINT.
Output measuring point for data is VMEAS.
Input stimulus edge rate is 0 to VCCINT in 0.5 ns (internal signal) from the driver preceding the IO buffer.
The first value is for output rising edge and the second value is for output falling edge. The hyphen (-) indicates
infinite resistance or disconnection.
Altera Corporation
January 2006
4–63
Stratix Device Handbook, Volume 1
Timing Model
Table 4–102 shows the reporting methodology used by the Quartus II
software for minimum timing information for output pins.
Table 4–102. Reporting Methodology For Minimum Timing For Single-Ended Output Pins (Part 1 of 2)
Notes (1), (2), (3)
Measurement
Point
Loading and Termination
I/O Standard
RUP
RDN
RS
RT
Ω
Ω
Ω
Ω
VCCIO
(V)
VTT
(V)
CL
(pF)
VMEAS
3.3-V LVTTL
–
–
0
–
3.600
3.600
10
1.800
2.5-V LVTTL
–
–
0
–
2.630
2.630
10
1.200
1.8-V LVTTL
–
–
0
–
1.950
1.950
10
0.880
1.5-V LVTTL
–
–
0
–
1.600
1.600
10
0.750
3.3-V LVCMOS
–
–
0
–
3.600
3.600
10
1.800
2.5-V LVCMOS
–
–
0
–
2.630
2.630
10
1.200
1.8-V LVCMOS
–
–
0
–
1.950
1.950
10
0.880
1.5-V LVCMOS
–
–
0
–
1.600
1.600
10
0.750
3.3-V GTL
–
–
0
25
3.600
1.260
30
0.860
2.5-V GTL
–
–
0
25
2.630
1.260
30
0.860
3.3-V GTL+
–
–
0
25
3.600
1.650
30
1.120
2.5-V GTL+
–
–
0
25
2.630
1.650
30
1.120
3.3-V SSTL-3 Class II
–
–
25
25
3.600
1.750
30
1.750
3.3-V SSTL-3 Class I
–
–
25
50
3.600
1.750
30
1.750
2.5-V SSTL-2 Class II
–
–
25
25
2.630
1.390
30
1.390
2.5-V SSTL-2 Class I
–
–
25
50
2.630
1.390
30
1.390
1.8-V SSTL-18 Class II
–
–
25
25
1.950
1.040
30
1.040
1.8-V SSTL-18 Class I
–
–
25
50
1.950
1.040
30
1.040
1.5-V HSTL Class II
–
–
0
25
1.600
0.800
20
0.900
1.5-V HSTL Class I
–
–
0
50
1.600
0.800
20
0.900
1.8-V HSTL Class II
–
–
0
25
1.950
0.900
20
1.000
1.8-V HSTL Class I
–
–
0
50
1.950
0.900
20
1.000
3.3-V PCI (4)
–/25
25/–
0
–
3.600
1.950
10
1.026/2.214
3.3-V PCI-X 1.0 (4)
–/25
25/–
0
–
3.600
1.950
10
1.026/2.214
3.3-V Compact PCI (4)
–/25
25/–
0
–
3.600
3.600
10
1.026/2.214
3.3-V AGP 1× (4)
–/25
25/–
0
–
3.600
3.600
10
1.026/2.214
4–64
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–102. Reporting Methodology For Minimum Timing For Single-Ended Output Pins (Part 2 of 2)
Notes (1), (2), (3)
Measurement
Point
Loading and Termination
I/O Standard
3.3-V CTT
RUP
RDN
RS
RT
Ω
Ω
Ω
Ω
VCCIO
(V)
VTT
(V)
CL
(pF)
VMEAS
–
–
25
50
3.600
1.650
30
1.650
Notes to Table 4–102:
(1)
(2)
(3)
(4)
Input measurement point at internal node is 0.5 × VCCINT.
Output measuring point for data is VMEAS. When two values are given, the first is the measurement point on the
rising edge and the other is for the falling edge.
Input stimulus edge rate is 0 to VCCINT in 0.5 ns (internal signal) from the driver preceding the I/O buffer.
The first value is for the output rising edge and the second value is for the output falling edge. The hyphen (-)
indicates infinite resistance or disconnection.
Figure 4–8 shows the measurement setup for output disable and output
enable timing. The TCHZ stands for clock to high Z time delay and is the
same as TXZ. The TCLZ stands for clock to low Z (driving) time delay and
is the same as TZX.
Figure 4–8. Measurement Setup for TXZ and TZX
CLK
T CHZ
200mV
OUT
VT =1.5V
R =50Ω
200mV
T CLZ
200mV
C TOTAL=10pF
OUT
200mV
Altera Corporation
January 2006
4–65
Stratix Device Handbook, Volume 1
Timing Model
External I/O Delay Parameters
External I/O delay timing parameters for I/O standard input and output
adders and programmable input and output delays are specified by
speed grade independent of device density. All of the timing parameters
in this section apply to both flip-chip and wire-bond packages.
Tables 4–103 and 4–104 show the input adder delays associated with
column and row I/O pins. If an I/O standard is selected other than 3.3-V
LVTTL or LVCMOS, add the selected delay to the external tINSU and
tINSUPLL I/O parameters shown in Tables 4–54 through 4–96.
Table 4–103. Stratix I/O Standard Column Pin Input Delay Adders
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Parameter
Unit
Min
LVCMOS
Max
0
Min
Max
0
Min
Max
0
Min
Max
0
ps
3.3-V LVTTL
0
0
0
0
ps
2.5-V LVTTL
19
19
22
26
ps
1.8-V LVTTL
221
232
266
313
ps
1.5-V LVTTL
352
369
425
500
ps
GTL
–45
–48
–55
–64
ps
GTL+
–75
–79
–91
–107
ps
3.3-V PCI
0
0
0
0
ps
3.3-V PCI-X 1.0
0
0
0
0
ps
Compact PCI
0
0
0
0
ps
AGP 1×
0
0
0
0
ps
AGP 2×
0
0
0
0
ps
CTT
120
126
144
170
ps
SSTL-3 Class I
–162
–171
–196
–231
ps
SSTL-3 Class II
–162
–171
–196
–231
ps
SSTL-2 Class I
–202
–213
–244
–287
ps
SSTL-2 Class II
–202
–213
–244
–287
ps
SSTL-18 Class I
78
81
94
110
ps
SSTL-18 Class II
78
81
94
110
ps
1.5-V HSTL Class I
–76
–80
–92
–108
ps
1.5-V HSTL Class II
–76
–80
–92
–108
ps
1.8-V HSTL Class I
–52
–55
–63
–74
ps
1.8-V HSTL Class II
–52
–55
–63
–74
ps
4–66
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–104. Stratix I/O Standard Row Pin Input Delay Adders
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Parameter
Unit
Min
Max
Min
Max
Min
Max
Min
Max
LVCMOS
0
0
0
0
ps
3.3-V LVTTL
0
0
0
0
ps
2.5-V LVTTL
21
22
25
29
ps
1.8-V LVTTL
181
190
218
257
ps
1.5-V LVTTL
300
315
362
426
ps
GTL+
–152
–160
–184
–216
ps
CTT
–168
–177
–203
–239
ps
SSTL-3 Class I
–193
–203
–234
–275
ps
SSTL-3 Class II
–193
–203
–234
–275
ps
SSTL-2 Class I
–262
–276
–317
–373
ps
SSTL-2 Class II
–262
–276
–317
–373
ps
SSTL-18 Class I
–105
–111
–127
–150
ps
SSTL-18 Class II
0
0
0
0
ps
1.5-V HSTL Class I
–151
–159
–183
–215
ps
1.8-V HSTL Class I
–126
–133
–153
–179
ps
LVDS
–149
–157
–180
–212
ps
LVPECL
–149
–157
–180
–212
ps
3.3-V PCML
–65
–69
–79
–93
ps
HyperTransport
77
–81
–93
–110
ps
Altera Corporation
January 2006
4–67
Stratix Device Handbook, Volume 1
Timing Model
Tables 4–105 through 4–108 show the output adder delays associated
with column and row I/O pins for both fast and slow slew rates. If an I/O
standard is selected other than 3.3-V LVTTL 4mA or LVCMOS 2 mA with
a fast slew rate, add the selected delay to the external tOUTCO, tOUTCOPLL,
tXZ, tZX, tXZPLL, and tZXPLL I/O parameters shown in Table 4–55 on
page 4–36 through Table 4–96 on page 4–56.
Table 4–105. Stratix I/O Standard Output Delay Adders for Fast Slew Rate on Column Pins (Part 1 of 2)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Parameter
Unit
Min
LVCMOS
3.3-V LVTTL
2.5-V LVTTL
1.8-V LVTTL
1.5-V LVTTL
Max
Min
Max
Min
Max
Min
Max
2 mA
1,895
1,990
1,990
1,990
ps
4 mA
956
1,004
1,004
1,004
ps
8 mA
189
198
198
198
ps
12 mA
0
0
0
0
ps
24 mA
–157
–165
–165
–165
ps
4 mA
1,895
1,990
1,990
1,990
ps
8 mA
1,347
1,414
1,414
1,414
ps
12 mA
636
668
668
668
ps
16 mA
561
589
589
589
ps
24 mA
0
0
0
0
ps
2 mA
2,517
2,643
2,643
2,643
ps
8 mA
834
875
875
875
ps
12 mA
504
529
529
529
ps
16 mA
194
203
203
203
ps
2 mA
1,304
1,369
1,369
1,369
ps
8 mA
960
1,008
1,008
1,008
ps
12 mA
960
1,008
1,008
1,008
ps
2 mA
6,680
7,014
7,014
7,014
ps
4 mA
3,275
3,439
3,439
3,439
ps
8 mA
1,589
1,668
1,668
1,668
ps
16
17
17
17
ps
GTL
GTL+
9
9
9
9
ps
3.3-V PCI
50
52
52
52
ps
3.3-V PCI-X 1.0
50
52
52
52
ps
Compact PCI
50
52
52
52
ps
AGP 1×
50
52
52
52
ps
AGP 2×
1,895
1,990
1,990
1,990
ps
4–68
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–105. Stratix I/O Standard Output Delay Adders for Fast Slew Rate on Column Pins (Part 2 of 2)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Parameter
Unit
Min
Max
Min
Max
Min
Max
Min
Max
CTT
973
1,021
1,021
1,021
ps
SSTL-3 Class I
719
755
755
755
ps
SSTL-3 Class II
146
153
153
153
ps
SSTL-2 Class I
678
712
712
712
ps
SSTL-2 Class II
223
234
234
234
ps
SSTL-18 Class I
1,032
1,083
1,083
1,083
ps
SSTL-18 Class II
447
469
469
469
ps
1.5-V HSTL Class I
660
693
693
693
ps
1.5-V HSTL Class II
537
564
564
564
ps
1.8-V HSTL Class I
304
319
319
319
ps
1.8-V HSTL Class II
231
242
242
242
ps
Table 4–106. Stratix I/O Standard Output Delay Adders for Fast Slew Rate on Row Pins
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
(Part 1 of 2)
-8 Speed Grade
Parameter
Unit
Min
LVCMOS
3.3-V LVTTL
2.5-V LVTTL
1.8-V LVTTL
Altera Corporation
January 2006
Max
Min
Max
1,594
Min
Max
1,594
Min
Max
2 mA
1,518
1,594
ps
4 mA
746
783
783
783
ps
8 mA
96
100
100
100
ps
12 mA
0
0
0
0
ps
4 mA
1,518
1,594
1,594
1,594
ps
8 mA
1,038
1,090
1,090
1,090
ps
12 mA
521
547
547
547
ps
16 mA
414
434
434
434
ps
24 mA
0
0
0
0
ps
2 mA
2,032
2,133
2,133
2,133
ps
8 mA
699
734
734
734
ps
12 mA
374
392
392
392
ps
16 mA
165
173
173
173
ps
2 mA
3,714
3,899
3,899
3,899
ps
8 mA
1,055
1,107
1,107
1,107
ps
12 mA
830
871
871
871
ps
4–69
Stratix Device Handbook, Volume 1
Timing Model
Table 4–106. Stratix I/O Standard Output Delay Adders for Fast Slew Rate on Row Pins
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
(Part 2 of 2)
-8 Speed Grade
Parameter
Unit
Min
1.5-V LVTTL
Max
Min
Max
Min
Max
Min
Max
2 mA
5,460
5,733
5,733
5,733
ps
4 mA
2,690
2,824
2,824
2,824
ps
8 mA
1,398
1,468
1,468
1,468
ps
GTL+
6
6
6
6
ps
CTT
845
887
887
887
ps
SSTL-3 Class I
638
670
670
670
ps
SSTL-3 Class II
144
151
151
151
ps
SSTL-2 Class I
604
634
634
634
ps
SSTL-2 Class II
211
221
221
221
ps
SSTL-18 Class I
955
1,002
1,002
1,002
ps
1.5-V HSTL Class I
733
769
769
769
ps
1.8-V HSTL Class I
372
390
390
390
ps
LVDS
–196
–206
–206
–206
ps
LVPECL
–148
–156
–156
–156
ps
PCML
–147
–155
–155
–155
ps
HyperTransport
technology
–93
–98
–98
–98
ps
Note to Table 4–103 through 4–106:
(1)
These parameters are only available on row I/O pins.
Table 4–107. Stratix I/O Standard Output Delay Adders for Slow Slew Rate on Column Pins (Part 1 of 2)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Parameter
Unit
Min
LVCMOS
Max
Min
Max
Min
Max
Min
Max
2 mA
1,822
1,913
1,913
1,913
ps
4 mA
684
718
718
718
ps
8 mA
233
245
245
245
ps
12 mA
1
1
1
1
ps
24 mA
–608
–638
–638
–638
ps
4–70
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–107. Stratix I/O Standard Output Delay Adders for Slow Slew Rate on Column Pins (Part 2 of 2)
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
Parameter
Unit
Min
3.3-V LVTTL
Max
Min
Max
Min
Max
Min
Max
4 mA
1,822
1,913
1,913
1,913
ps
8 mA
1,586
1,665
1,665
1,665
ps
12 mA
686
720
720
720
ps
16 mA
630
662
662
662
ps
24 mA
0
0
0
0
ps
2 mA
2,925
3,071
3,071
3,071
ps
2.5-V LVTTL
8 mA
1,496
1,571
1,571
1,571
ps
12 mA
937
984
984
984
ps
16 mA
1,003
1,053
1,053
1,053
ps
2 mA
7,101
7,456
7,456
7,456
ps
1.8-V LVTTL
8 mA
3,620
3,801
3,801
3,801
ps
12 mA
3,109
3,265
3,265
3,265
ps
1.5-V LVTTL
2 mA
10,941
11,488
11,488
11,488
ps
4 mA
7,431
7,803
7,803
7,803
ps
8 mA
5,990
6,290
6,290
6,290
ps
GTL
–959
–1,007
–1,007
–1,007
ps
GTL+
–438
–460
–460
–460
ps
3.3-V PCI
660
693
693
693
ps
3.3-V PCI-X 1.0
660
693
693
693
ps
Compact PCI
660
693
693
693
ps
AGP 1×
660
693
693
693
ps
AGP 2×
288
303
303
303
ps
CTT
631
663
663
663
ps
SSTL-3 Class I
301
316
316
316
ps
SSTL-3 Class II
–359
–377
–377
–377
ps
SSTL-2 Class I
523
549
549
549
ps
SSTL-2 Class II
–49
–51
–51
–51
ps
SSTL-18 Class I
2,315
2,431
2,431
2,431
ps
SSTL-18 Class II
723
759
759
759
ps
1.5-V HSTL Class I
1,687
1,771
1,771
1,771
ps
1.5-V HSTL Class II
1,095
1,150
1,150
1,150
ps
1.8-V HSTL Class I
599
629
678
744
ps
1.8-V HSTL Class II
87
102
102
102
ps
Altera Corporation
January 2006
4–71
Stratix Device Handbook, Volume 1
Timing Model
Table 4–108. Stratix I/O Standard Output Delay Adders for Slow Slew Rate on Row Pins
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
-8 Speed Grade
I/O Standard
Unit
Min
LVCMOS
3.3-V LVTTL
2.5-V LVTTL
1.8-V LVTTL
Max
Min
Max
Min
Max
Min
Max
2 mA
1,571
1,650
1,650
1,650
ps
4 mA
594
624
624
624
ps
8 mA
208
218
218
218
ps
12 mA
0
0
0
0
ps
4 mA
1,571
1,650
1,650
1,650
ps
8 mA
1,393
1,463
1,463
1,463
ps
12 mA
596
626
626
626
ps
16 mA
562
590
590
590
ps
2 mA
2,562
2,690
2,690
2,690
ps
8 mA
1,343
1,410
1,410
1,410
ps
12 mA
864
907
907
907
ps
16 mA
945
992
992
992
ps
2 mA
6,306
6,621
6,621
6,621
ps
8 mA
3,369
3,538
3,538
3,538
ps
12 mA
2,932
3,079
3,079
3,079
ps
2 mA
9,759
10,247
10,247
10,247
ps
4 mA
6,830
7,172
7,172
7,172
ps
8 mA
5,699
5,984
5,984
5,984
ps
GTL+
–333
–350
–350
–350
ps
CTT
591
621
621
621
ps
1.5-V LVTTL
SSTL-3 Class I
267
280
280
280
ps
SSTL-3 Class II
–346
–363
–363
–363
ps
SSTL-2 Class I
481
505
505
505
ps
SSTL-2 Class II
–58
–61
–61
–61
ps
SSTL-18 Class I
2,207
2,317
2,317
2,317
ps
1.5-V HSTL Class I
1,966
2,064
2,064‘
2,064
ps
1.8-V HSTL Class I
1,208
1,268
1,460
1,720
ps
4–72
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Tables 4–109 and 4–110 show the adder delays for the column and row
IOE programmable delays. These delays are controlled with the
Quartus II software logic options listed in the Parameter column.
Table 4–109. Stratix IOE Programmable Delays on Column Pins Note (1)
-5 Speed Grade -6 Speed Grade -7 Speed Grade -8 Speed Grade
Parameter
Setting
Unit
Min
Decrease input delay
to internal cells
Max
Min
Max
Min
Max
Min
Max
Off
3,970
4,367
5,022
5,908
ps
Small
3,390
3,729
4,288
5,045
ps
Medium
2,810
3,091
3,554
4,181
ps
224
235
270
318
ps
Large
On
224
235
270
318
ps
Decrease input delay
to input register
Off
3,900
4,290
4,933
5,804
ps
On
0
0
0
0
ps
Decrease input delay
to output register
Off
1,240
1,364
1,568
1,845
ps
On
0
0
0
0
ps
Increase delay to
output pin
Off
0
0
0
0
ps
On
397
417
417
417
ps
Increase delay to
output enable pin
Off
0
0
0
0
ps
On
338
372
427
503
ps
Increase output clock
enable delay
Off
0
0
0
0
ps
Increase input clock
enable delay
Increase output
enable clock enable
delay
Increase tZX delay to
output pin
Altera Corporation
January 2006
Small
540
594
683
804
ps
Large
1,016
1,118
1,285
1,512
ps
On
1,016
1,118
1,285
1,512
ps
Off
0
0
0
0
ps
Small
540
594
683
804
ps
Large
1,016
1,118
1,285
1,512
ps
On
1,016
1,118
1,285
1,512
ps
Off
0
0
0
0
ps
Small
540
594
683
804
ps
Large
1,016
1,118
1,285
1,512
ps
On
1,016
1,118
1,285
1,512
ps
Off
0
0
0
0
ps
On
2,199
2,309
2,309
2,309
ps
4–73
Stratix Device Handbook, Volume 1
Timing Model
Table 4–110. Stratix IOE Programmable Delays on Row Pins Note (1)
-5 Speed Grade -6 Speed Grade -7 Speed Grade -8 Speed Grade
Parameter
Setting
Unit
Min
Decrease input delay
to internal cells
Max
Min
Max
Min
Max
Min
Max
Off
3,970
4,367
5,022
5,908
ps
Small
3,390
3,729
4,288
5,045
ps
Medium
2,810
3,091
3,554
4,181
ps
173
181
208
245
ps
Large
On
173
181
208
245
ps
Decrease input delay
to input register
Off
3,900
4,290
4,933
5,804
ps
On
0
0
0
0
ps
Decrease input delay
to output register
Off
1,240
1,364
1,568
1,845
ps
On
0
0
0
0
ps
Increase delay to
output pin
Off
0
0
0
0
ps
On
397
417
417
417
ps
Increase delay to
output enable pin
Off
0
0
0
0
ps
On
348
383
441
518
ps
Increase output clock
enable delay
Off
0
0
0
0
ps
Small
180
198
227
267
ps
Large
260
286
328
386
ps
On
260
286
328
386
ps
Off
0
0
0
0
ps
Small
180
198
227
267
ps
Large
260
286
328
386
ps
On
260
286
328
386
ps
Off
0
0
0
0
ps
Increase input clock
enable delay
Increase output
enable clock enable
delay
Increase tZX delay to
output pin
Small
540
594
683
804
ps
Large
1,016
1,118
1,285
1,512
ps
On
1,016
1,118
1,285
1,512
ps
Off
0
0
0
0
ps
On
1,993
2,092
2,092
2,092
ps
Note to Table 4–109 and Table 4–110:
(1)
The delay chain delays vary for different device densities. These timing values only apply to EP1S30 and EP1S40
devices. Reference the timing information reported by the Quartus II software for other devices.
4–74
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
The scaling factors for column output pin timing in Tables 4–111 to 4–113
are shown in units of time per pF unit of capacitance (ps/pF). Add this
delay to the tCO or combinatorial timing path for output or bidirectional
pins in addition to the I/O adder delays shown in Tables 4–103 through
4–108 and the IOE programmable delays in Tables 4–109 and 4–110.
Table 4–111. Output Delay Adder for Loading on LVTTL/LVCMOS Output Buffers Note (1)
Conditions
Parameter
Drive Strength
Output Pin Adder Delay (ps/pF)
Value
3.3-V LVTTL
2.5-V LVTTL
1.8-V LVTTL
1.5-V LVTTL
LVCMOS
24mA
15
–
–
-
8
16mA
25
18
–
–
–
12mA
30
25
25
–
15
8mA
50
35
40
35
20
4mA
60
–
–
80
30
2mA
–
75
120
160
60
Note to Table 4–111:
(1)
The timing information in this table is preliminary.
Table 4–112. Output Delay Adder for Loading on SSTL/HSTL Output Buffers
Note (1)
Output Pin Adder Delay (ps/pF)
Conditions
Class I
Class II
SSTL-3
SSTL-2
SSTL-1.8
1.5-V HSTL
25
25
25
25
25
20
25
20
Note to Table 4–112:
(1)
The timing information in this table is preliminary.
Table 4–113. Output Delay Adder for Loading on GTL+/GTL/CTT/PCI Output Buffers
Conditions
Note (1)
Output Pin Adder Delay (ps/pF)
Parameter
Value
GTL+
GTL
CTT
PCI
AGP
VCCIO Voltage
Level
3.3V
18
18
25
20
20
2.5V
15
18
-
-
-
Note to Table 4–113:
(1)
The timing information in this table is preliminary.
Altera Corporation
January 2006
4–75
Stratix Device Handbook, Volume 1
Timing Model
Maximum Input & Output Clock Rates
Tables 4–114 through 4–119 show the maximum input clock rate for
column and row pins in Stratix devices.
Table 4–114. Stratix Maximum Input Clock Rate for CLK[7..4] & CLK[15..12]
Pins in Flip-Chip Packages (Part 1 of 2)
I/O Standard
-5 Speed -6 Speed -7 Speed -8 Speed
Grade
Grade
Grade
Grade
Unit
LVTTL
422
422
390
390
MHz
2.5 V
422
422
390
390
MHz
1.8 V
422
422
390
390
MHz
1.5 V
422
422
390
390
MHz
LVCMOS
422
422
390
390
MHz
GTL
300
250
200
200
MHz
GTL+
300
250
200
200
MHz
SSTL-3 Class I
400
350
300
300
MHz
SSTL-3 Class II
400
350
300
300
MHz
SSTL-2 Class I
400
350
300
300
MHz
SSTL-2 Class II
400
350
300
300
MHz
SSTL-18 Class I
400
350
300
300
MHz
SSTL-18 Class II
400
350
300
300
MHz
1.5-V HSTL Class I
400
350
300
300
MHz
1.5-V HSTL Class II
400
350
300
300
MHz
1.8-V HSTL Class I
400
350
300
300
MHz
1.8-V HSTL Class II
400
350
300
300
MHz
3.3-V PCI
422
422
390
390
MHz
3.3-V PCI-X 1.0
422
422
390
390
MHz
Compact PCI
422
422
390
390
MHz
AGP 1×
422
422
390
390
MHz
AGP 2×
422
422
390
390
MHz
CTT
300
250
200
200
MHz
Differential 1.5-V HSTL
C1
400
350
300
300
MHz
LVPECL (1)
645
645
622
622
MHz
PCML (1)
300
275
275
275
MHz
4–76
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–114. Stratix Maximum Input Clock Rate for CLK[7..4] & CLK[15..12]
Pins in Flip-Chip Packages (Part 2 of 2)
I/O Standard
-5 Speed -6 Speed -7 Speed -8 Speed
Grade
Grade
Grade
Grade
Unit
LVDS (1)
645
645
622
622
MHz
HyperTransport
technology (1)
500
500
450
450
MHz
Table 4–115. Stratix Maximum Input Clock Rate for CLK[0, 2, 9, 11] Pins &
FPLL[10..7]CLK Pins in Flip-Chip Packages
I/O Standard
Altera Corporation
January 2006
-5 Speed -6 Speed -7 Speed -8 Speed
Grade
Grade
Grade
Grade
Unit
LVTTL
422
422
390
390
MHz
2.5 V
422
422
390
390
MHz
1.8 V
422
422
390
390
MHz
1.5 V
422
422
390
390
MHz
LVCMOS
422
422
390
390
MHz
GTL+
300
250
200
200
MHz
SSTL-3 Class I
400
350
300
300
MHz
SSTL-3 Class II
400
350
300
300
MHz
SSTL-2 Class I
400
350
300
300
MHz
SSTL-2 Class II
400
350
300
300
MHz
SSTL-18 Class I
400
350
300
300
MHz
SSTL-18 Class II
400
350
300
300
MHz
1.5-V HSTL Class I
400
350
300
300
MHz
1.8-V HSTL Class I
400
350
300
300
MHz
CTT
300
250
200
200
MHz
Differential 1.5-V HSTL
C1
400
350
300
300
MHz
LVPECL (1)
717
717
640
640
MHz
PCML (1)
400
375
350
350
MHz
LVDS (1)
717
717
640
640
MHz
HyperTransport
technology (1)
717
717
640
640
MHz
4–77
Stratix Device Handbook, Volume 1
Timing Model
Table 4–116. Stratix Maximum Input Clock Rate for CLK[1, 3, 8, 10] Pins in
Flip-Chip Packages
I/O Standard
-5 Speed -6 Speed -7 Speed -8 Speed
Grade
Grade
Grade
Grade
Unit
LVTTL
422
422
390
390
MHz
2.5 V
422
422
390
390
MHz
1.8 V
422
422
390
390
MHz
1.5 V
422
422
390
390
MHz
LVCMOS
422
422
390
390
MHz
GTL+
300
250
200
200
MHz
SSTL-3 Class I
400
350
300
300
MHz
SSTL-3 Class II
400
350
300
300
MHz
SSTL-2 Class I
400
350
300
300
MHz
SSTL-2 Class II
400
350
300
300
MHz
SSTL-18 Class I
400
350
300
300
MHz
SSTL-18 Class II
400
350
300
300
MHz
1.5-V HSTL Class I
400
350
300
300
MHz
1.8-V HSTL Class I
400
350
300
300
MHz
CTT
300
250
200
200
MHz
Differential 1.5-V HSTL
C1
400
350
300
300
MHz
LVPECL (1)
645
645
640
640
MHz
PCML (1)
300
275
275
275
MHz
LVDS (1)
645
645
640
640
MHz
HyperTransport
technology (1)
500
500
450
450
MHz
Table 4–117. Stratix Maximum Input Clock Rate for CLK[7..4] & CLK[15..12]
Pins in Wire-Bond Packages (Part 1 of 2)
I/O Standard
-6 Speed -7 Speed -8 Speed
Grade
Grade
Grade
Unit
LVTTL
422
390
390
MHz
2.5 V
422
390
390
MHz
1.8 V
422
390
390
MHz
1.5 V
422
390
390
MHz
LVCMOS
422
390
390
MHz
GTL
250
200
200
MHz
4–78
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–117. Stratix Maximum Input Clock Rate for CLK[7..4] & CLK[15..12]
Pins in Wire-Bond Packages (Part 2 of 2)
I/O Standard
-6 Speed -7 Speed -8 Speed
Grade
Grade
Grade
Unit
GTL+
250
200
200
MHz
SSTL-3 Class I
300
250
250
MHz
SSTL-3 Class II
300
250
250
MHz
SSTL-2 Class I
300
250
250
MHz
SSTL-2 Class II
300
250
250
MHz
SSTL-18 Class I
300
250
250
MHz
SSTL-18 Class II
300
250
250
MHz
1.5-V HSTL Class I
300
180
180
MHz
1.5-V HSTL Class II
300
180
180
MHz
1.8-V HSTL Class I
300
180
180
MHz
1.8-V HSTL Class II
300
180
180
MHz
3.3-V PCI
422
390
390
MHz
3.3-V PCI-X 1.0
422
390
390
MHz
Compact PCI
422
390
390
MHz
AGP 1×
422
390
390
MHz
AGP 2×
422
390
390
MHz
CTT
250
180
180
MHz
Differential 1.5-V HSTL
C1
300
180
180
MHz
LVPECL (1)
422
400
400
MHz
PCML (1)
215
200
200
MHz
LVDS (1)
422
400
400
MHz
HyperTransport
technology (1)
422
400
400
MHz
Table 4–118. Stratix Maximum Input Clock Rate for CLK[0, 2, 9, 11] Pins &
FPLL[10..7]CLK Pins in Wire-Bond Packages (Part 1 of 2)
I/O Standard
Altera Corporation
January 2006
-6 Speed -7 Speed -8 Speed
Grade
Grade
Grade
Unit
LVTTL
422
390
390
MHz
2.5 V
422
390
390
MHz
1.8 V
422
390
390
MHz
1.5 V
422
390
390
MHz
4–79
Stratix Device Handbook, Volume 1
Timing Model
Table 4–118. Stratix Maximum Input Clock Rate for CLK[0, 2, 9, 11] Pins &
FPLL[10..7]CLK Pins in Wire-Bond Packages (Part 2 of 2)
I/O Standard
-6 Speed -7 Speed -8 Speed
Grade
Grade
Grade
Unit
LVCMOS
422
390
390
MHz
GTL+
250
200
200
MHz
SSTL-3 Class I
350
300
300
MHz
SSTL-3 Class II
350
300
300
MHz
SSTL-2 Class I
350
300
300
MHz
SSTL-2 Class II
350
300
300
MHz
SSTL-18 Class I
350
300
300
MHz
SSTL-18 Class II
350
300
300
MHz
1.5-V HSTL Class I
350
300
300
MHz
1.8-V HSTL Class I
350
300
300
MHz
CTT
250
200
200
MHz
Differential 1.5-V HSTL
C1
350
300
300
MHz
LVPECL (1)
717
640
640
MHz
PCML (1)
375
350
350
MHz
LVDS (1)
717
640
640
MHz
HyperTransport
technology (1)
717
640
640
MHz
Table 4–119. Stratix Maximum Input Clock Rate for CLK[1, 3, 8, 10] Pins in
Wire-Bond Packages (Part 1 of 2)
I/O Standard
-6 Speed -7 Speed -8 Speed
Grade
Grade
Grade
Unit
LVTTL
422
390
390
MHz
2.5 V
422
390
390
MHz
1.8 V
422
390
390
MHz
1.5 V
422
390
390
MHz
LVCMOS
422
390
390
MHz
GTL+
250
200
200
MHz
SSTL-3 Class I
350
300
300
MHz
SSTL-3 Class II
350
300
300
MHz
SSTL-2 Class I
350
300
300
MHz
SSTL-2 Class II
350
300
300
MHz
4–80
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–119. Stratix Maximum Input Clock Rate for CLK[1, 3, 8, 10] Pins in
Wire-Bond Packages (Part 2 of 2)
I/O Standard
-6 Speed -7 Speed -8 Speed
Grade
Grade
Grade
Unit
SSTL-18 Class I
350
300
300
MHz
SSTL-18 Class II
350
300
300
MHz
1.5-V HSTL Class I
350
300
300
MHz
1.8-V HSTL Class I
350
300
300
MHz
CTT
250
200
200
MHz
Differential 1.5-V HSTL
C1
350
300
300
MHz
LVPECL (1)
645
622
622
MHz
PCML (1)
275
275
275
MHz
LVDS (1)
645
622
622
MHz
HyperTransport
technology (1)
500
450
450
MHz
Note to Tables 4–114 through 4–119:
(1)
These parameters are only available on row I/O pins.
Tables 4–120 through 4–123 show the maximum output clock rate for
column and row pins in Stratix devices.
Table 4–120. Stratix Maximum Output Clock Rate for PLL[5, 6, 11, 12] Pins
in Flip-Chip Packages (Part 1 of 2)
I/O Standard
Altera Corporation
January 2006
-5 Speed -6 Speed -7 Speed -8 Speed
Grade
Grade
Grade
Grade
Unit
LVTTL
350
300
250
250
MHz
2.5 V
350
300
300
300
MHz
1.8 V
250
250
250
250
MHz
1.5 V
225
200
200
200
MHz
LVCMOS
350
300
250
250
MHz
GTL
200
167
125
125
MHz
GTL+
200
167
125
125
MHz
SSTL-3 Class I
200
167
167
133
MHz
SSTL-3 Class II
200
167
167
133
MHz
SSTL-2 Class I (3)
200
200
167
167
MHz
SSTL-2 Class I (4)
200
200
167
167
MHz
SSTL-2 Class I (5)
150
134
134
134
MHz
4–81
Stratix Device Handbook, Volume 1
Timing Model
Table 4–120. Stratix Maximum Output Clock Rate for PLL[5, 6, 11, 12] Pins
in Flip-Chip Packages (Part 2 of 2)
I/O Standard
-5 Speed -6 Speed -7 Speed -8 Speed
Grade
Grade
Grade
Grade
Unit
SSTL-2 Class II (3)
200
200
167
167
MHz
SSTL-2 Class II (4)
200
200
167
167
MHz
SSTL-2 Class II (5)
150
134
134
134
MHz
SSTL-18 Class I
150
133
133
133
MHz
SSTL-18 Class II
150
133
133
133
MHz
1.5-V HSTL Class I
250
225
200
200
MHz
1.5-V HSTL Class II
225
200
200
200
MHz
1.8-V HSTL Class I
250
225
200
200
MHz
1.8-V HSTL Class II
225
200
200
200
MHz
3.3-V PCI
350
300
250
250
MHz
3.3-V PCI-X 1.0
350
300
250
250
MHz
Compact PCI
350
300
250
250
MHz
AGP 1×
350
300
250
250
MHz
AGP 2×
350
300
250
250
MHz
CTT
200
200
200
200
MHz
Differential 1.5-V HSTL
C1
225
200
200
200
MHz
Differential 1.8-V HSTL
Class I
250
225
200
200
MHz
Differential 1.8-V HSTL
Class II
225
200
200
200
MHz
Differential SSTL-2 (6)
200
200
167
167
MHz
LVPECL (2)
500
500
500
500
MHz
PCML (2)
350
350
350
350
MHz
LVDS (2)
500
500
500
500
MHz
HyperTransport
technology (2)
350
350
350
350
MHz
4–82
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–121. Stratix Maximum Output Clock Rate (Using I/O Pins) for PLL[1,
2, 3, 4] Pins in Flip-Chip Packages
I/O Standard
Altera Corporation
January 2006
-5 Speed -6 Speed -7 Speed -8 Speed
Grade
Grade
Grade
Grade
Unit
LVTTL
400
350
300
300
MHz
2.5 V
400
350
300
300
MHz
1.8 V
400
350
300
300
MHz
1.5 V
350
300
300
300
MHz
LVCMOS
400
350
300
300
MHz
GTL
200
167
125
125
MHz
GTL+
200
167
125
125
MHz
SSTL-3 Class I
167
150
133
133
MHz
SSTL-3 Class II
167
150
133
133
MHz
SSTL-2 Class I
150
133
133
133
MHz
SSTL-2 Class II
150
133
133
133
MHz
SSTL-18 Class I
150
133
133
133
MHz
SSTL-18 Class II
150
133
133
133
MHz
1.5-V HSTL Class I
250
225
200
200
MHz
1.5-V HSTL Class II
225
225
200
200
MHz
1.8-V HSTL Class I
250
225
200
200
MHz
1.8-V HSTL Class II
225
225
200
200
MHz
3.3-V PCI
250
225
200
200
MHz
3.3-V PCI-X 1.0
225
225
200
200
MHz
Compact PCI
400
350
300
300
MHz
AGP 1×
400
350
300
300
MHz
AGP 2×
400
350
300
300
MHz
CTT
300
250
200
200
MHz
LVPECL (2)
717
717
500
500
MHz
PCML (2)
420
420
420
420
MHz
LVDS (2)
717
717
500
500
MHz
HyperTransport
technology (2)
420
420
420
420
MHz
4–83
Stratix Device Handbook, Volume 1
Timing Model
Table 4–122. Stratix Maximum Output Clock Rate for PLL[5, 6, 11, 12] Pins
in Wire-Bond Packages (Part 1 of 2)
I/O Standard
-6 Speed -7 Speed -8 Speed
Grade
Grade
Grade
Unit
LVTTL
175
150
150
MHz
2.5 V
175
150
150
MHz
1.8 V
175
150
150
MHz
1.5 V
175
150
150
MHz
LVCMOS
175
150
150
MHz
GTL
125
100
100
MHz
GTL+
125
100
100
MHz
SSTL-3 Class I
110
90
90
MHz
SSTL-3 Class II
133
125
125
MHz
SSTL-2 Class I
166
133
133
MHz
SSTL-2 Class II
133
100
100
MHz
SSTL-18 Class I
110
100
100
MHz
SSTL-18 Class II
110
100
100
MHz
1.5-V HSTL Class I
167
167
167
MHz
1.5-V HSTL Class II
167
133
133
MHz
1.8-V HSTL Class I
167
167
167
MHz
1.8-V HSTL Class II
167
133
133
MHz
3.3-V PCI
167
167
167
MHz
3.3-V PCI-X 1.0
167
133
133
MHz
Compact PCI
175
150
150
MHz
AGP 1×
175
150
150
MHz
AGP 2×
175
150
150
MHz
CTT
125
100
100
MHz
Differential 1.5-V HSTL
C1
167
133
133
MHz
Differential 1.8-V HSTL
Class I
167
167
167
MHz
Differential 1.8-V HSTL
Class II
167
133
133
MHz
Differential SSTL-2 (1)
110
100
100
MHz
LVPECL (2)
311
275
275
MHz
PCML (2)
250
200
200
MHz
4–84
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–122. Stratix Maximum Output Clock Rate for PLL[5, 6, 11, 12] Pins
in Wire-Bond Packages (Part 2 of 2)
I/O Standard
-6 Speed -7 Speed -8 Speed
Grade
Grade
Grade
Unit
LVDS (2)
311
275
275
MHz
HyperTransport
technology (2)
311
275
275
MHz
Table 4–123. Stratix Maximum Output Clock Rate (Using I/O Pins) for PLL[1,
2, 3, 4] Pins in Wire-Bond Packages (Part 1 of 2)
I/O Standard
Altera Corporation
January 2006
-6 Speed -7 Speed -8 Speed
Grade
Grade
Grade
Unit
LVTTL
200
175
175
MHz
2.5 V
200
175
175
MHz
1.8 V
200
175
175
MHz
1.5 V
200
175
175
MHz
LVCMOS
200
175
175
MHz
GTL
125
100
100
MHz
GTL+
125
100
100
MHz
SSTL-3 Class I
110
90
90
MHz
SSTL-3 Class II
150
133
133
MHz
SSTL-2 Class I
90
80
80
MHz
SSTL-2 Class II
110
100
100
MHz
SSTL-18 Class I
110
100
100
MHz
SSTL-18 Class II
110
100
100
MHz
1.5-V HSTL Class I
225
200
200
MHz
1.5-V HSTL Class II
200
167
167
MHz
1.8-V HSTL Class I
225
200
200
MHz
1.8-V HSTL Class II
200
167
167
MHz
3.3-V PCI
200
175
175
MHz
3.3-V PCI-X 1.0
200
175
175
MHz
Compact PCI
200
175
175
MHz
AGP 1×
200
175
175
MHz
AGP 2×
200
175
175
MHz
CTT
125
100
100
MHz
LVPECL (2)
311
270
270
MHz
PCML (2)
400
311
311
MHz
4–85
Stratix Device Handbook, Volume 1
Timing Model
Table 4–123. Stratix Maximum Output Clock Rate (Using I/O Pins) for PLL[1,
2, 3, 4] Pins in Wire-Bond Packages (Part 2 of 2)
I/O Standard
-6 Speed -7 Speed -8 Speed
Grade
Grade
Grade
Unit
LVDS (2)
400
311
311
MHz
HyperTransport
technology (2)
420
400
400
MHz
Notes to Tables 4–120 through 4–123:
(1)
(2)
(3)
Differential SSTL-2 outputs are only available on column clock pins.
These parameters are only available on row I/O pins.
SSTL-2 in maximum drive strength condition. See Table 4–101 on page 4–62 for
more information on exact loading conditions for each I/O standard.
(4)
(5)
(6)
SSTL-2 in minimum drive strength with 10pF output load condition.
SSTL-2 in minimum drive strength with > 10pF output load condition.
Differential SSTL-2 outputs are only supported on column clock pins.
4–86
Stratix Device Handbook, Volume 1
≤
Altera Corporation
January 2006
DC & Switching Characteristics
High-Speed I/O
Specification
Table 4–124 provides high-speed timing specifications definitions.
Table 4–124. High-Speed Timing Specifications & Terminology
High-Speed Timing Specification
Terminology
tC
High-speed receiver/transmitter input and output clock period.
fHSCLK
High-speed receiver/transmitter input and output clock frequency.
tRISE
Low-to-high transmission time.
tFALL
High-to-low transmission time.
Timing unit interval (TUI)
The timing budget allowed for skew, propagation delays, and data
sampling window. (TUI = 1/(Receiver Input Clock Frequency ×
Multiplication Factor) = tC/w).
fHSDR
Maximum LVDS data transfer rate (fHSDR = 1/TUI).
Channel-to-channel skew (TCCS)
The timing difference between the fastest and slowest output edges,
including tCO variation and clock skew. The clock is included in the TCCS
measurement.
Sampling window (SW)
The period of time during which the data must be valid to be captured
correctly. The setup and hold times determine the ideal strobe position
within the sampling window.
SW = tSW (max) – tSW (min).
Input jitter (peak-to-peak)
Peak-to-peak input jitter on high-speed PLLs.
Output jitter (peak-to-peak)
Peak-to-peak output jitter on high-speed PLLs.
tDUTY
Duty cycle on high-speed transmitter output clock.
tLOCK
Lock time for high-speed transmitter and receiver PLLs.
J
Deserialization factor (width of internal data bus).
W
PLL multiplication factor.
Altera Corporation
January 2006
4–87
Stratix Device Handbook, Volume 1
Table 4–125. High-Speed I/O Specifications for Flip-Chip Packages (Part 1 of 4) Notes (1), (2)
-5 Speed Grade
Symbol
fHSDR Device
operation
(LVDS,
LVPECL,
HyperTransport
technology)
-7 Speed Grade
-8 Speed Grade
Unit
Min
fHSCLK (Clock
frequency)
(LVDS,
LVPECL,
HyperTransport
technology)
fHSCLK = fHSDR /
W
-6 Speed Grade
Conditions
Typ
Max
Min
Typ
Max
Min
Typ
Max
Min
Typ
Max
W = 4 to 30
(Serdes used)
10
210
10
210
10
156
10
115.5
MHz
W = 2 (Serdes
bypass)
50
231
50
231
50
231
50
231
MHz
W = 2 (Serdes
used)
150
420
150
420
150
312
150
231
MHz
W = 1 (Serdes
bypass)
100
462
100
462
100
462
100
462
MHz
W = 1 (Serdes
used)
300
717
300
717
300
624
300
462
MHz
J = 10
300
840
300
840
300
640
300
462
Mbps
J=8
300
840
300
840
300
640
300
462
Mbps
J=7
300
840
300
840
300
640
300
462
Mbps
J=4
300
840
300
840
300
640
300
462
Mbps
J=2
100
462
100
462
100
640
100
462
Mbps
J = 1 (LVDS
and LVPECL
only)
100
462
100
462
100
640
100
462
Mbps
High-Speed I/O Specification
4–88
Stratix Device Handbook, Volume 1
Tables 4–125 and 4–126 show the high-speed I/O timing for Stratix devices.
Altera Corporation
January 2006
-5 Speed Grade
Symbol
fHSDR Device
operation
(PCML)
4–89
Stratix Device Handbook, Volume 1
TCCS
-7 Speed Grade
-8 Speed Grade
Unit
Min
fHSCLK (Clock
frequency)
(PCML)
fHSCLK = fHSDR /
W
-6 Speed Grade
Conditions
Typ
Max
Min
Typ
Max
Min
Typ
Max
Min
Typ
Max
W = 4 to 30
(Serdes used)
10
100
10
100
10
77.75
10
77.75
MHz
W = 2 (Serdes
bypass)
50
200
50
200
50
150
50
150
MHz
W = 2 (Serdes
used)
150
200
150
200
150
155.5
150
155.5
MHz
W = 1 (Serdes
bypass)
100
250
100
250
100
200
100
200
MHz
W = 1 (Serdes
used)
300
400
300
400
300
311
300
311
MHz
J = 10
300
400
300
400
300
311
300
311
Mbps
J=8
300
400
300
400
300
311
300
311
Mbps
J=7
300
400
300
400
300
311
300
311
Mbps
J=4
300
400
300
400
300
311
300
311
Mbps
J=2
100
400
100
400
100
300
100
300
Mbps
J=1
100
250
100
250
100
200
100
200
Mbps
300
ps
All
200
200
300
High-Speed I/O Specification
Altera Corporation
January 2006
Table 4–125. High-Speed I/O Specifications for Flip-Chip Packages (Part 2 of 4) Notes (1), (2)
-5 Speed Grade
Symbol
-7 Speed Grade
-8 Speed Grade
Unit
Min
SW
-6 Speed Grade
Conditions
Typ
Max
Min
Typ
Max
Min
Typ
Max
Min
Typ
Max
PCML (J = 4, 7,
8, 10)
750
750
800
800
ps
PCML (J = 2)
900
900
1,200
1,200
ps
PCML (J = 1)
1,500
1,500
1,700
1,700
ps
LVDS and
LVPECL (J = 1)
500
500
550
550
ps
LVDS,
LVPECL,
HyperTransport
technology
(J = 2 through
10)
440
440
500
500
ps
Altera Corporation
January 2006
Input jitter
tolerance
(peak-to-peak)
All
250
250
250
250
ps
Output jitter
(peak-to-peak)
All
160
160
200
200
ps
Output tRISE
LVDS
80
110
120
80
110
120
80
110
120
80
110
120
ps
HyperTransport
technology
110
170
200
110
170
200
120
170
200
120
170
200
ps
LVPECL
90
130
150
90
130
150
100
135
150
100
135
150
ps
PCML
80
110
135
80
110
135
80
110
135
80
110
135
ps
LVDS
80
110
120
80
110
120
80
110
120
80
110
120
ps
HyperTransport
technology
110
170
200
110
170
200
110
170
200
110
170
200
ps
LVPECL
90
130
160
90
130
160
100
135
160
100
135
160
ps
PCML
105
140
175
105
140
175
110
145
175
110
145
175
ps
Output tFALL
High-Speed I/O Specification
4–90
Stratix Device Handbook, Volume 1
Table 4–125. High-Speed I/O Specifications for Flip-Chip Packages (Part 3 of 4) Notes (1), (2)
-5 Speed Grade
Symbol
tDUTY
LVDS (J = 2
through 10)
LVDS (J =1)
and LVPECL,
PCML,
HyperTransport
technology
tLOCK
-7 Speed Grade
-8 Speed Grade
Unit
Min
Typ
Max
Min
Typ
Max
Min
Typ
Max
Min
Typ
Max
47.5
50
52.5
47.5
50
52.5
47.5
50
52.5
47.5
50
52.5
%
45
50
55
45
50
55
45
50
55
45
50
55
%
100
μs
All
Notes to Table 4–125:
(1)
(2)
-6 Speed Grade
Conditions
When J = 4, 7, 8, and 10, the SERDES block is used.
When J = 2 or J = 1, the SERDES is bypassed.
100
100
100
High-Speed I/O Specification
Altera Corporation
January 2006
Table 4–125. High-Speed I/O Specifications for Flip-Chip Packages (Part 4 of 4) Notes (1), (2)
4–91
Stratix Device Handbook, Volume 1
-6 Speed Grade
Symbol
fHSCLK (Clock
frequency)
(LVDS,LVPECL,
HyperTransport
technology)
fHSCLK = fHSDR / W
W = 4 to 30 (Serdes used)
fHSDR Device operation,
(LVDS,LVPECL,
HyperTransport
technology)
Device operation,
fH S D R
(PCML)
Altera Corporation
January 2006
TCCS
-8 Speed Grade
Unit
Min
fH S C L K (Clock
frequency)
(PCML)
fHSCLK = fHSDR / W
-7 Speed Grade
Conditions
10
Typ
Max
Min
156
10
Typ
Max
Min
115.5
10
Typ
Max
115.5
MHz
W = 2 (Serdes bypass)
50
231
50
231
50
231
MHz
W = 2 (Serdes used)
150
312
150
231
150
231
MHz
W = 1 (Serdes bypass)
100
311
100
270
100
270
MHz
W = 1 (Serdes used)
300
624
300
462
300
462
MHz
J = 10
300
624
300
462
300
462
Mbps
J=8
300
624
300
462
300
462
Mbps
J=7
300
624
300
462
300
462
Mbps
J=4
300
624
300
462
300
462
Mbps
J=2
100
462
100
462
100
462
Mbps
J = 1 (LVDS and LVPECL
only)
100
311
100
270
100
270
Mbps
W = 4 to 30 (Serdes used)
10
77.75
W = 2 (Serdes bypass)
50
150
W = 2 (Serdes used)
150
155.5
MHz
50
77.5
50
77.5
MHz
MHz
W = 1 (Serdes bypass)
100
200
W = 1 (Serdes used)
300
311
MHz
J = 10
300
311
Mbps
J=8
300
311
Mbps
J=7
300
311
Mbps
J=4
300
311
Mbps
J=2
100
300
100
155
100
155
Mbps
J=1
100
200
100
155
100
155
Mbps
400
ps
All
400
100
155
400
100
155
MHz
High-Speed I/O Specification
4–92
Stratix Device Handbook, Volume 1
Table 4–126. High-Speed I/O Specifications for Wire-Bond Packages (Part 1 of 2)
-6 Speed Grade
Symbol
-8 Speed Grade
Unit
Min
SW
-7 Speed Grade
Conditions
PCML (J = 4, 7, 8, 10) only
Typ
Max
Min
Typ
Max
Min
Typ
Max
800
800
800
ps
PCML (J = 2) only
1,200
1,200
1,200
ps
PCML (J = 1) only
1,700
1,700
1,700
ps
LVDS and LVPECL (J = 1)
only
550
550
550
ps
LVDS, LVPECL,
HyperTransport technology
(J = 2 through 10) only
500
500
500
ps
Input jitter tolerance
(peak-to-peak)
All
250
250
250
ps
Output jitter (peak-topeak)
All
200
200
200
ps
Output tR I S E
LVDS
120
ps
4–93
Stratix Device Handbook, Volume 1
Output tFA L L
tD U T Y
110
120
80
110
120
80
110
HyperTransport technology
120
170
200
120
170
200
120
170
200
ps
LVPECL
100
135
150
100
135
150
100
135
150
ps
PCML
80
110
135
80
110
135
80
110
135
ps
LVDS
80
110
120
80
110
120
80
110
120
ps
HyperTransport
110
170
200
110
170
200
110
170
200
ps
LVPECL
100
135
160
100
135
160
100
135
160
ps
PCML
110
145
175
110
145
175
110
145
175
ps
LVDS (J = 2 through10) only
47.5
50
52.5
47.5
50
52.5
47.5
50
52.5
%
45
50
55
45
50
55
45
50
55
%
100
μs
LVDS (J =1) and LVPECL,
PCML, HyperTransport
technology
tL O C K
80
All
100
100
High-Speed I/O Specification
Altera Corporation
January 2006
Table 4–126. High-Speed I/O Specifications for Wire-Bond Packages (Part 2 of 2)
PLL Specifications
PLL
Specifications
Tables 4–127 through 4–129 describe the Stratix device enhanced PLL
specifications.
Table 4–127. Enhanced PLL Specifications for -5 Speed Grades (Part 1 of 2)
Symbol
Parameter
Min
Typ
Max
Unit
3
(1), (2)
684
MHz
3
420
MHz
fIN
Input clock frequency
fINPFD
Input frequency to PFD
fINDUTY
Input clock duty cycle
40
60
%
fEINDUTY
External feedback clock input duty
cycle
40
60
%
tINJITTER
Input clock period jitter
±200 (3)
ps
tEINJITTER
External feedback clock period jitter
±200 (3)
ps
tFCOMP
External feedback clock
compensation time (4)
6
ns
fOUT
Output frequency for internal global
or regional clock
0.3
500
MHz
fOUT_EXT
Output frequency for external clock
(3)
0.3
526
MHz
tOUTDUTY
Duty cycle for external clock output
(when set to 50%)
45
55
%
tJITTER
Period jitter for external clock output
(6)
±100 ps for >200-MHz outclk
±20 mUI for <200-MHz outclk
ps or
mUI
tCONFIG5,6
Time required to reconfigure the
scan chains for PLLs 5 and 6
289/fSCANCLK
tCONFIG11,12
Time required to reconfigure the
scan chains for PLLs 11 and 12
193/fSCANCLK
tSCANCLK
scanclk frequency (5)
22
MHz
tDLOCK
Time required to lock dynamically
(after switchover or reconfiguring
any non-post-scale
counters/delays) (7)
100
μs
tLOCK
Time required to lock from end of
device configuration
10
400
μs
fVCO
PLL internal VCO operating range
300
800 (8)
MHz
tLSKEW
Clock skew between two external
clock outputs driven by the same
counter
4–94
Stratix Device Handbook, Volume 1
±50
ps
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–127. Enhanced PLL Specifications for -5 Speed Grades (Part 2 of 2)
Symbol
Parameter
Min
Typ
Max
±75
Unit
ps
tSKEW
Clock skew between two external
clock outputs driven by the different
counters with the same settings
fSS
Spread spectrum modulation
frequency
30
% spread
Percentage spread for spread
spectrum frequency (10)
0.4
tARESET
Minimum pulse width on areset
signal
10
ns
tA R E S E T _ R E C O N
FIG
Minimum pulse width on the
areset signal when using PLL
reconfiguration. Reset the PLL after
scandataout goes high.
500
ns
0.5
Table 4–128. Enhanced PLL Specifications for -6 Speed Grades
Symbol
Parameter
fIN
Input clock frequency
Min
Typ
150
kHz
0.6
%
(Part 1 of 2)
Max
Unit
3
(1), (2)
650
MHz
fINPFD
Input frequency to PFD
3
420
MHz
fINDUTY
Input clock duty cycle
40
60
%
fEINDUTY
External feedback clock input duty
cycle
40
60
%
tINJITTER
Input clock period jitter
±200 (3)
ps
tEINJITTER
External feedback clock period jitter
±200 (3)
ps
tFCOMP
External feedback clock compensation
time (4)
6
ns
fOUT
Output frequency for internal global or
regional clock
0.3
450
MHz
fOUT_EXT
Output frequency for external clock (3)
0.3
500
MHz
tOUTDUTY
Duty cycle for external clock output
(when set to 50%)
45
55
%
tJITTER
Period jitter for external clock output
(6)
±100 ps for >200-MHz outclk
±20 mUI for <200-MHz outclk
ps or
mUI
tCONFIG5,6
Time required to reconfigure the scan
chains for PLLs 5 and 6
289/fSCANCLK
tCONFIG11,12 Time required to reconfigure the scan
chains for PLLs 11 and 12
193/fSCANCLK
Altera Corporation
January 2006
4–95
Stratix Device Handbook, Volume 1
PLL Specifications
Table 4–128. Enhanced PLL Specifications for -6 Speed Grades
Symbol
Parameter
Min
(Part 2 of 2)
Typ
Max
Unit
22
MHz
tSCANCLK
scanclk frequency (5)
tDLOCK
Time required to lock dynamically
(after switchover or reconfiguring any
non-post-scale counters/delays) (7)
(11)
(9)
100
μs
tLOCK
Time required to lock from end of
device configuration (11)
10
400
μs
fVCO
PLL internal VCO operating range
300
800 (8)
MHz
tLSKEW
Clock skew between two external
clock outputs driven by the same
counter
±50
ps
tSKEW
Clock skew between two external
clock outputs driven by the different
counters with the same settings
±75
ps
fSS
Spread spectrum modulation
frequency
30
% spread
Percentage spread for spread
spectrum frequency (10)
0.4
tARESET
Minimum pulse width on areset
signal
10
0.5
150
kHz
0.6
%
ns
Table 4–129. Enhanced PLL Specifications for -7 Speed Grade (Part 1 of 2)
Symbol
Parameter
Max
Unit
3
(1), (2)
565
MHz
Input frequency to PFD
3
420
MHz
fINDUTY
Input clock duty cycle
40
60
%
fEINDUTY
External feedback clock input duty
cycle
40
60
%
tINJITTER
Input clock period jitter
±200 (3)
ps
tEINJITTER
External feedback clock period jitter
±200 (3)
ps
tFCOMP
External feedback clock
compensation time (4)
6
ns
fOUT
Output frequency for internal global
or regional clock
0.3
420
MHz
fOUT_EXT
Output frequency for external clock
(3)
0.3
434
MHz
fIN
Input clock frequency
fINPFD
4–96
Stratix Device Handbook, Volume 1
Min
Typ
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–129. Enhanced PLL Specifications for -7 Speed Grade (Part 2 of 2)
Symbol
Parameter
Min
Typ
Max
Unit
55
%
±100 ps for >200-MHz outclk
±20 mUI for <200-MHz outclk
ps or
mUI
tOUTDUTY
Duty cycle for external clock output
(when set to 50%)
45
tJITTER
Period jitter for external clock output
(6)
tCONFIG5,6
Time required to reconfigure the
scan chains for PLLs 5 and 6
289/fSCANCLK
tCONFIG11,12
Time required to reconfigure the
scan chains for PLLs 11 and 12
193/fSCANCLK
tSCANCLK
scanclk frequency (5)
tDLOCK
Time required to lock dynamically
(after switchover or reconfiguring any
non-post-scale counters/delays) (7)
(11)
tLOCK
22
MHz
(9)
100
μs
Time required to lock from end of
device configuration (11)
10
400
μs
fVCO
PLL internal VCO operating range
300
600 (8)
MHz
tLSKEW
Clock skew between two external
clock outputs driven by the same
counter
±50
ps
tSKEW
Clock skew between two external
clock outputs driven by the different
counters with the same settings
±75
ps
fSS
Spread spectrum modulation
frequency
30
150
kHz
% spread
Percentage spread for spread
spectrum frequency (10)
0.5
0.6
%
tARESET
Minimum pulse width on areset
signal
10
ns
Table 4–130. Enhanced PLL Specifications for -8 Speed Grade (Part 1 of 3)
Symbol
Parameter
Max
Unit
3
(1), (2)
480
MHz
Input frequency to PFD
3
420
MHz
fINDUTY
Input clock duty cycle
40
60
%
fEINDUTY
External feedback clock input duty
cycle
40
60
%
tINJITTER
Input clock period jitter
±200 (3)
ps
fIN
Input clock frequency
fINPFD
Altera Corporation
January 2006
Min
Typ
4–97
Stratix Device Handbook, Volume 1
PLL Specifications
Table 4–130. Enhanced PLL Specifications for -8 Speed Grade (Part 2 of 3)
Symbol
Parameter
Min
Typ
Max
Unit
±200 (3)
ps
6
ns
tEINJITTER
External feedback clock period jitter
tFCOMP
External feedback clock
compensation time (4)
fOUT
Output frequency for internal global
or regional clock
0.3
357
MHz
fOUT_EXT
Output frequency for external clock
(3)
0.3
369
MHz
tOUTDUTY
Duty cycle for external clock output
(when set to 50%)
45
55
%
tJITTER
Period jitter for external clock output
(6)
±100 ps for >200-MHz outclk
±20 mUI for <200-MHz outclk
ps or
mUI
tCONFIG5,6
Time required to reconfigure the
scan chains for PLLs 5 and 6
289/fSCANCLK
tCONFIG11,12
Time required to reconfigure the
scan chains for PLLs 11 and 12
193/fSCANCLK
tSCANCLK
scanclk frequency (5)
tDLOCK
Time required to lock dynamically
(after switchover or reconfiguring
any non-post-scale counters/delays)
(7) (11)
tLOCK
fVCO
22
MHz
(9)
100
μs
Time required to lock from end of
device configuration (11)
10
400
μs
PLL internal VCO operating range
300
600 (8)
MHz
4–98
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–130. Enhanced PLL Specifications for -8 Speed Grade (Part 3 of 3)
Symbol
Parameter
Min
Typ
Max
Unit
tLSKEW
Clock skew between two external
clock outputs driven by the same
counter
±50
ps
tSKEW
Clock skew between two external
clock outputs driven by the different
counters with the same settings
±75
ps
fSS
Spread spectrum modulation
frequency
30
150
kHz
% spread
Percentage spread for spread
spectrum frequency (10)
0.5
0.6
%
tARESET
Minimum pulse width on areset
signal
10
ns
Notes to Tables 4–127 through 4–130:
(1)
(2)
The minimum input clock frequency to the PFD (fIN/N) must be at least 3 MHz for Stratix device enhanced PLLs.
Use this equation (fOUT = fI N * ml(n × post-scale counter)) in conjunction with the specified fI N P F D and fV C O ranges
to determine the allowed PLL settings.
(3) See “Maximum Input & Output Clock Rates” on page 4–76.
(4) tFCOMP can also equal 50% of the input clock period multiplied by the pre-scale divider n (whichever is less).
(5) This parameter is timing analyzed by the Quartus II software because the scanclk and scandata ports can be
driven by the logic array.
(6) Actual jitter performance may vary based on the system configuration.
(7) Total required time to reconfigure and lock is equal to tDLOCK + tCONFIG. If only post-scale counters and delays are
changed, then tDLOCK is equal to 0.
(8) When using the spread-spectrum feature, the minimum VCO frequency is 500 MHz. The maximum VCO
frequency is determined by the speed grade selected.
(9) Lock time is a function of PLL configuration and may be significantly faster depending on bandwidth settings or
feedback counter change increment.
(10) Exact, user-controllable value depends on the PLL settings.
(11) The LOCK circuit on Stratix PLLs does not work for industrial devices below -20C unless the PFD frequency > 200
MHz. See the Stratix FPGA Errata Sheet for more information on the PLL.
Altera Corporation
January 2006
4–99
Stratix Device Handbook, Volume 1
PLL Specifications
Tables 4–131 through 4–133 describe the Stratix device fast PLL
specifications.
Table 4–131. Fast PLL Specifications for -5 & -6 Speed Grade Devices
Symbol
Parameter
Min
Max
Unit
fIN
CLKIN frequency (1), (2), (3)
10
717
MHz
fINPFD
Input frequency to PFD
10
500
MHz
fOUT
Output frequency for internal global or
regional clock (3)
9.375
420
MHz
fOUT_DIFFIO
Output frequency for external clock
driven out on a differential I/O data
channel (2)
(5)
(5)
fVCO
VCO operating frequency
300
1,000
MHz
tINDUTY
CLKIN duty cycle
40
60
%
±200
ps
55
%
(5)
ps
tINJITTER
Period jitter for CLKIN pin
tDUTY
Duty cycle for DFFIO 1× CLKOUT pin (6)
tJITTER
Period jitter for DIFFIO clock out (6)
tLOCK
Time required for PLL to acquire lock
10
100
μs
m
Multiplication factors for m counter (6)
1
32
Integer
l0, l1, g0
Multiplication factors for l0, l1, and g0
counter (7), (8)
1
32
Integer
tARESET
Minimum pulse width on areset
signal
10
45
ns
Table 4–132. Fast PLL Specifications for -7 Speed Grades (Part 1 of 2)
Symbol
Parameter
Min
Max
Unit
fIN
CLKIN frequency (1), (3)
10
640
MHz
fINPFD
Input frequency to PFD
10
500
MHz
fOUT
Output frequency for internal global or
regional clock (4)
9.375
420
MHz
fOUT_DIFFIO
Output frequency for external clock
driven out on a differential I/O data
channel
(5)
(5)
MHz
fVCO
VCO operating frequency
300
700
MHz
tINDUTY
CLKIN duty cycle
40
60
%
tINJITTER
Period jitter for CLKIN pin
±200
ps
tDUTY
Duty cycle for DFFIO 1× CLKOUT pin (6)
55
%
4–100
Stratix Device Handbook, Volume 1
45
Altera Corporation
January 2006
DC & Switching Characteristics
Table 4–132. Fast PLL Specifications for -7 Speed Grades (Part 2 of 2)
Symbol
Parameter
tJITTER
Period jitter for DIFFIO clock out (6)
tLOCK
Time required for PLL to acquire lock
Min
10
Max
Unit
(5)
ps
100
μs
m
Multiplication factors for m counter (7)
1
32
Integer
l0, l1, g0
Multiplication factors for l0, l1, and g0
counter (7), (8)
1
32
Integer
tARESET
Minimum pulse width on areset
signal
10
ns
Table 4–133. Fast PLL Specifications for -8 Speed Grades (Part 1 of 2)
Symbol
Parameter
fIN
CLKIN frequency (1), (3)
fINPFD
Input frequency to PFD
fOUT
Output frequency for internal global or
regional clock (4)
fOUT_DIFFIO
Min
Max
Unit
10
460
MHz
10
500
MHz
9.375
420
MHz
Output frequency for external clock
driven out on a differential I/O data
channel
(5)
(5)
MHz
fVCO
VCO operating frequency
300
700
MHz
tINDUTY
CLKIN duty cycle
40
60
%
tINJITTER
Period jitter for CLKIN pin
±200
ps
tDUTY
Duty cycle for DFFIO 1× CLKOUT pin (6)
45
55
%
tJITTER
Period jitter for DIFFIO clock out (6)
(5)
ps
tLOCK
Time required for PLL to acquire lock
10
100
μs
m
Multiplication factors for m counter (7)
1
32
Integer
l0, l1, g0
Multiplication factors for l0, l1, and g0
counter (7), (8)
1
32
Integer
Altera Corporation
January 2006
4–101
Stratix Device Handbook, Volume 1
DLL Specifications
Table 4–133. Fast PLL Specifications for -8 Speed Grades (Part 2 of 2)
Symbol
tARESET
Parameter
Min
Minimum pulse width on areset
signal
Max
Unit
10
ns
Notes to Tables 4–131 through 4–133:
(1)
(2)
See “Maximum Input & Output Clock Rates” on page 4–76.
PLLs 7, 8, 9, and 10 in the EP1S80 device support up to 717-MHz input and output.
(3)
Use this equation (fO U T = fI N * ml(n × post-scale counter)) in conjunction with the specified fI N P F D and fV C O
ranges to determine the allowed PLL settings.
When using the SERDES, high-speed differential I/O mode supports a maximum output frequency of 210 MHz
to the global or regional clocks (that is, the maximum data rate 840 Mbps divided by the smallest SERDES J factor
of 4).
(4)
(5)
(6)
(7)
(8)
Refer to the section “High-Speed I/O Specification” on page 4–87 for more information.
This parameter is for high-speed differential I/O mode only.
These counters have a maximum of 32 if programmed for 50/50 duty cycle. Otherwise, they have a maximum
of 16.
High-speed differential I/O mode supports W = 1 to 16 and J = 4, 7, 8, or 10.
DLL
Specifications
Table 4–134 reports the jitter for the DLL in the DQS phase shift reference
circuit.
Table 4–134. DLL Jitter for DQS Phase Shift Reference Circuit
Frequency (MHz)
f
DLL Jitter (ps)
197 to 200
± 100
160 to 196
± 300
100 to 159
± 500
For more information on DLL jitter, see the DDR SRAM section in the
Stratix Architecture chapter of the Stratix Device Handbook, Volume 1.
Table 4–135 lists the Stratix DLL low frequency limit for full phase shift
across all PVT conditions. The Stratix DLL can be used below these
frequencies, but it will not achieve the full phase shift requested across all
4–102
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
DC & Switching Characteristics
process and operating conditions. Run the timing analyzer in the
Quartus II software at the fast and slow operating conditions to see the
phase shift range that is achieved below these frequencies.
Table 4–135. Stratix DLL Low Frequency Limit for Full Phase Shift
Altera Corporation
January 2006
Phase Shift
Minimum Frequency for
Full Phase Shift
Unit
72°
119
MHz
90°
149
MHz
4–103
Stratix Device Handbook, Volume 1
DLL Specifications
4–104
Stratix Device Handbook, Volume 1
Altera Corporation
January 2006
5. Reference & Ordering
Information
S51005-2.1
Software
Stratix® devices are supported by the Altera® Quartus® II design
software, which provides a comprehensive environment for system-on-aprogrammable-chip (SOPC) design. The Quartus II software includes
HDL and schematic design entry, compilation and logic synthesis, full
simulation and advanced timing analysis, SignalTap® II logic analyzer,
and device configuration. See the Design Software Selector Guide for more
details on the Quartus II software features.
The Quartus II software supports the Windows XP/2000/NT/98, Sun
Solaris, Linux Red Hat v7.1 and HP-UX operating systems. It also
supports seamless integration with industry-leading EDA tools through
the NativeLink® interface.
Device Pin-Outs
Stratix device pin-outs can be found on the Altera web site
(www.altera.com).
Ordering
Information
Figure 5–1 describes the ordering codes for Stratix devices. For more
information on a specific package, see the Package Information for Stratix
Devices chapter.
Altera Corporation
September 2004
5–1
Ordering Information
Figure 5–1. Stratix Device Packaging Ordering Information
EP1S
80
F
1508
C
Family Signature
7
ES
Optional Suffix
EP1S: Stratix
Indicates specific device options or
shipment method.
ES: Engineering sample
Device Type
10
20
25
30
40
60
80
Speed Grade
5, 6, or 7, with 5 being the fastest
Operating Temperature
C: Commercial temperature (tJ = 0˚ C to 85˚ C)
I: Industrial temperature (tJ = -40˚ C to 100˚ C)
Package Type
B: Ball-grid array (BGA)
F: FineLine BGA
5–2
Stratix Device Handbook, Volume 1
Pin Count
Number of pins for a particular BGA or FineLine BGA package
Altera Corporation
September 2004
Index
A
Accumulator 2–63
Adder/Output Blocks 2–61
Adder/Subtractor
2–63
Accumulator
2–63
AGP 1x Specifications 4–13
AGP 2x Specifications 4–13
Architecture 2–1
36 x 36 Multiply Mode 2–66
addnsub Signal 2–8
Block Diagram 2–2
Bus Hold 2–121
Byte Alignment 2–140
Carry-Select Chain 2–11
Clear & Preset Logic Control 2–13
Combined Resources 2–78
Dedicated Circuitry 2–137
Device Resources 2–3
Device Routing Scheme 2–20
Digital Signal Processing Block 2–52
Direct Link Connection 2–5
Dynamic Arithmetic Mode 2–10
in LE 2–11
Four-Multipliers
Adder Mode 2–68
Functional Description 2–1
LAB
Interconnects 2–4
Logic Array Blocks 2–3
Structure 2–4
LE Operating Modes 2–8
Logic Elements 2–6
Modes of Operation 2–64
Multiplier Size & Configurations per DSP
block 2–70
Multiply-Accumulator Mode 2–67
MultiTrack Interconnect 2–14
Normal Mode 2–9
in LE 2–9
Altera Corporation
Open-Drain Output 2–120
Power Sequencing & Hot Socketing 2–140
Programmable Drive Strength 2–119
Programmable Pull-Up Resistor 2–122
Simple Multiplier Mode 2–64
Single-Port Mode 2–51
Slew-Rate Control 2–120
Two-Multipliers
Adder Mode 2–67
Adder Mode Implementing Complex
Multiply 2–68
C
Class I Specifications 4–11, 4–12
Class II Specifications 4–11, 4–12, 4–13
Clocks
Clock Feedback 2–96
Clock Multiplication & Division 2–88, 2–101
Clock Switchover
2–88
Delay 2–97
EP1S10, EP1S20 & EP1S25
Device I/O Clock Groups
2–80
EP1S25, EP1S20 & EP1S10 Device Fast Clock
Pin Connections to Fast Regional
Clocks 2–77
EP1S30 Device Fast Regional Clock Pin Connections to Fast Regional Clocks 2–78
EP1S30, EP1S40, EP1S60, EP1S80
Device I/O Clock Groups
2–81
External Clock
Inputs 2–102
Outputs 2–92, 2–103
Outputs for Enhanced PLLs 11 & 12 2–95
Outputs for PLLs 5 & 6 2–93
Fast Regional Clock External I/O Timing
Parameters 4–34
Fast Regional Clock Network 2–76
Index–1
Stratix Device Handbook, Volume 1
Global & Hierarchical Clocking 2–73
Global & Regional Clock Connections
from Side Pins & Fast PLL Outputs 2–85
from Top Clock Pins & Enhanced PLL
Outputs 2–86
Global Clock External I/O Timing
Parameters 4–35
Global Clock Network 2–74
Global Clocking 2–75
Independent Clock Mode 2–44
Input/Output
Clock Mode
2–46
Simple Dual-Port Mode 2–48
True Dual-Port Mode 2–47
Maximum Input & Output Clock Rates 4–76
Maximum Input Clock Rate
for CLK
(0, 2, 9, 11) Pins in
Flip-Chip
Packages
Wire-Bond
Packages
4–77
4–79
(1, 3, 8, 10) Pins in
Flip-Chip
Packages
Wire-Bond
Packages
4–78
(5, 6, 11, 12) Pins in
Flip-Chip
Index–2
4–84
Phase & Delay Shifting 2–96
Phase Delay 2–96
PLL Clock Networks 2–73
Read/Write Clock Mode
2–49
in Simple Dual-Port Mode 2–50
Regional Clock 2–75
External I/O Timing Parameters 4–34
Regional Clock Bus 2–79
Regional Clock Network 2–75
Spread-Spectrum Clocking 2–98
Configuration 3–5
32-Bit IDCODE 3–3
and Testing 3–1
Data Sources for Configuration 3–7
Local Update Mode 3–12
Local Update Transition Diagram 3–12
Operating Modes 3–5
Partial Reconfiguration 3–7
Remote Update 3–8
Remote Update Transition Diagram 3–11
Schemes 3–7
SignalTap II Embedded Logic Analyzer 3–5
Stratix FPGAs with JRunner 3–7
Control Signals 2–104
D
4–76
4–78
Maximum Output Clock Rate
for PLL
(1, 2, 3, 4) Pins in
Flip-Chip
Packages
Wire-Bond
Packages
4–81
4–80
(7..4) & CLK(15..12) Pins in
Flip-Chip
Packages
Wire-Bond
Packages
Packages
Wire-Bond
Packages
4–83
4–85
DC Switching
Absolute Maximum Ratings 4–1
Bus Hold Parameters 4–16
Capacitance 4–17
DC & Switching Characteristics 4–1
External Timing Parameters 4–33
Operating Conditions 4–1
Performance 4–20
Power Consumption 4–17
Recommended Operating Conditions 4–1
DDR
Double-Data Rate I/O Pins 2–111
Device Features
EP1S10, EP1S20, EP1S25, EP1S30, 1–3
EP1S40, EP1S60, EP1S80, 1–3
Altera Corporation
Stratix Device Handbook, Volume 1
Differential HSTL Specifications 4–15
DSP
Block Diagram
Configuration
for 18 x 18-Bit 2–55
for 9 x 9-Bit 2–56
Block Interconnect Interface 2–71
Block Interface 2–70
Block Signal Sources & Destinations 2–73
Blocks
Arranged in Columns 2–53
in Stratix Devices 2–54
Input Register Modes 2–60
Input Registers 2–58
Multiplier
2–60
Block 2–57
Signed Representation 2–60
Sub-Block 2–57
Sub-Blocks Using Input Shift Register
Connections 2–59
Pipeline/Post Multiply Register 2–61
E
EP1S10 Devices
Column Pin
Fast Regional Clock External I/O
Parameters 4–36
Global Clock External I/O
Parameters 4–37
Regional Clock External I/O
Parameters 4–36
Row Pin
Fast Regional Clock External I/O
Parameters 4–37
Global Clock External I/O
Parameters 4–38
Regional Clock External I/O
Parameters 4–38
EP1S20 Devices
Column Pin
Fast Regional Clock External I/O
Parameters 4–39
Global Clock External I/O
Parameters 4–40
Regional Clock External I/O
Altera Corporation
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Parameters 4–39
Row Pin
Fast Regional Clock External I/O
Parameters 4–40
Global Clock External I/O
Parameters 4–41
Regional Clock External I/O
Parameters 4–41
EP1S25 Devices
Column Pin
Fast Regional Clock External I/O
Parameters 4–42
Global Clock External I/O
Parameters 4–43
Regional Clock External I/O
Parameters 4–42
Row Pin
Fast Regional Clock External I/O
Parameters 4–43
Global Clock External I/O
Parameters 4–44
Regional Clock External I/O
Parameters 4–44
EP1S30 Devices
Column Pin
Fast Regional Clock External I/O
Parameters 4–45
Global Clock External I/O
Parameters 4–45
Regional Clock External I/O
Parameters 4–45
Row Pin
Fast Regional Clock External I/O
Parameters 4–46
Global Clock External I/O
Parameters 4–47
Regional Clock External I/O
Parameters 4–47
EP1S40 Devices
Column Pin
Fast Regional Clock External I/O
Parameters 4–48
Global Clock External I/O
Parameters 4–49
Regional Clock External I/O
Parameters 4–48
Row Pin
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Index–3
Stratix Device Handbook, Volume 1
Fast Regional Clock External I/O
Parameters 4–49
Global Clock External I/O
Parameters 4–50
Regional Clock External I/O
Parameters 4–50
EP1S60 Devices
Column Pin
Fast Regional Clock External I/O
Parameters 4–51
Global Clock External I/O
Parameters 4–52
Regional Clock External I/O
Parameters 4–51
M-RAM
Interface Locations 2–38
Row Pin
Fast Regional Clock External I/O
Parameters 4–52
Global Clock External I/O
Parameters 4–53
Regional Clock External I/O
Parameters 4–53
EP1S80 Devices
Column Pin
Fast Regional Clock External I/O
Parameters 4–54
Global Clock External I/O
Parameters 4–55
Regional Clock External I/O
Parameters 4–54
Global Clock External I/O
Parameters 4–56
Row Pin
Fast Regional Clock External I/O
Parameters 4–55
Regional Clock External I/O
Parameters 4–56
H
HSTL
Class I Specifications 4–14, 4–15
Class II Specifications 4–14, 4–15
Index–4
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Timing
Timing
I
I/O
Standards
1.5-V 4–14, 4–15
I/O Specifications 4–4
1.8-V
I/O Specifications 4–4
2.5-V
I/O Specifications 4–3
3.3-V 4–13
LVDS I/O Specifications 4–6
PCI Specifications 4–9
PCML Specifications 4–8
Advanced I/O Standard Support 2–122
Column I/O Block Connection to the
Interconnect 2–107
Column Pin
Input Delay Adders 4–66
Control Signal Selection per IOE 2–109
CTT I/O Specifications 4–16
Differential
LVDS
Input
On-Chip
Termination 2–128
External I/O Delay Parameters 4–66
GTL+ I/O Specifications 4–10
High-Speed
Differential
I/O
Support 2–130
HyperTransport
Technology
Specifications 4–9
I/O Banks 2–125
I/O Structure 2–104
I/O Support by Bank 2–126
IOE Structure 2–105
LVCMOS Specifications 4–3
LVDS Performance on Fast PLL
Input 2–103
LVPECL Specifications 4–8
LVTTL Specifications 4–3
MultiVolt I/O Interface 2–129
MultiVolt I/O Support 2–130
Output Delay Adders for Fast Slew Rate
on Column Pins 4–68
Output Delay Adders for Fast Slew Rate
on Row Pins 4–69
Output Delay Adders for Slow Slew Rate
on Column Pins 4–70
Package Options & I/O Pin Counts 1–4
Receiver Input Waveforms for Differential
Altera Corporation
Stratix Device Handbook, Volume 1
I/O Standards 4–5
I/O Block Connection to the
Interconnect 2–106
Row Pin
Input Delay Adders 4–67
Signal Path through the I/O Block 2–108
SSTL-18 4–11
SSTL-2 4–12
SSTL-3 4–12, 4–13
Stratix IOE in Bidirectional I/O
Configuration 2–110
Supported I/O Standards 2–123
Transmitter Output Waveforms for Differential I/O Standards 4–6
Interconnect
C4 Connections 2–18
DSP Block Interface to Interconnect 2–72
Left-Facing
M-RAM
to
Interconnect
Interface 2–40
LUT
Chain
Register
Chain
Interconnects 2–17
M-RAM
Column
Unit
Interface
to
Interconnect 2–42
Row Unit Interface to Interconnect 2–41
R4 Connections 2–15
IOE
Internal Timing Microparameters 4–29
Row
J
JTAG
Boundary-Scan
Register Length 3–3
Support 3–1
Stratix JTAG
Instructions 3–2
Waveforms 3–4
L
LAB
Control Signals 2–5
Wide Control Signals 2–6
LUT
Chain & Register Chain 2–8
Altera Corporation
M
Memory Architecture
Byte Enable for M4K
RAM Block
2–32
Byte Enable for M-RAM
Block
2–35
External RAM Interfacing 2–115
M4K
Block Internal Timing
Microparameter
Descriptions 4–24
Microparameters 4–31
RAM Block
2–30
Configurations (Simple DualPort) 2–31
Configurations
(True
DualPort) 2–31
Control Signals 2–33
LAB Row Interface 2–33
M512
Block Internal Timing
Microparameter
Descriptions 4–24
Microparameters 4–30
RAM Block
Architecture 2–27
Configurations (Simple Dual-Port
RAM) 2–27
Control Signals 2–29
LAB Row Interface 2–30
Memory Block Size 2–26
Memory Modes 2–21
M-RAM
Block
2–34
Configurations (Simple DualPort) 2–34
Configurations
(True
DualPort) 2–35
Block Control Signals 2–37
Block Internal Timing
Microparameter
Descriptions 4–25
Combined Byte Selection for x144
Index–5
Stratix Device Handbook, Volume 1
Mode 2–36
&
Column
Interface
Unit
Signals 2–43
Parity Bit Support 2–24
Shift Register
Memory Configuration 2–26
Support 2–25
Simple Dual-Port & Single-Port Memory
Configurations 2–23
Stratix
IOE
in
DDR
Input
I/O
Configuration 2–112
Stratix IOE in DDR Output I/O
Configuration 2–114
TriMatrix Memory 2–21
True
Dual-Port
Memory
Configuration 2–22
Port I/O Standards 2–102
I/O Standards Supported for Enhanced PLL
Pins 2–94
Lock Detect & Programmable Gated
Locked 2–98
PLL Locations 2–84
Programmable Bandwidth 2–91
Programmable Delay Chain 2–111
Programmable Duty Cycle 2–98
Reconfiguration 2–90
Row
O
Ordering Information 5–1
Device Pin-Outs 5–1
Packaging Ordering Information 5–2
Reference & Ordering Information 5–1
Output Registers 2–64
Output Selection Multiplexer 2–64
P
Packaging
BGA Package Sizes 1–4
Device Speed Grades 1–5
FineLine BGA Package Sizes 1–5
PCI-X 1.0 Specifications 4–10
Phase Shifting 2–103
PLL
Advanced Clear & Enable Control 2–98
Dynamically Programmable Counters & Delays in Stratix Device Enhanced
PLLs 2–91
Enhanced
Fast PLLs 2–81
Fast PLL 2–100
Channel Layout EP1S10, EP1S20 or
EP1S25 Devices 2–138
Channel Layout EP1S30 to EP1S80
Devices 2–139
Index–6
T
Testing
Temperature Sensing Diode 3–13
Electrical Characteristics 3–14
External 3–14
Temperature vs. Temperature-Sensing Diode
Voltage 3–15
Timing
DSP
Block Internal Timing
Microparameter
Descriptions 4–23
Microparameters 4–29
Dual-Port RAM Timing Microparameter
Waveform 4–27
External Timing in Stratix Devices 4–33
High-Speed I/O Timing 4–87
High-Speed
Timing
Specifications
&
Terminology 4–87
Internal Parameters 4–22
IOE Internal Timing Microparameter
Descriptions 4–22
LE Internal Timing Microparameters 4–28
Logic Elements Internal Timing Microparameter Descriptions 4–22
Model 4–19
PLL Timing 4–94
Preliminary & Final 4–19
Stratix Device Timing Model Status 4–19
Stratix JTAG
Timing Parameters & Values 3–4
TriMatrix Memory
TriMatrix Memory Features 2–21
Altera Corporation
Stratix Device Handbook, Volume 2
101 Innovation Drive
San Jose, CA 95134
(408) 544-7000
http://www.altera.com
S5V2-3.5
Copyright © 2006 Altera Corporation. All rights reserved. Altera, The Programmable Solutions Company, the stylized Altera logo, specific device designations, and all other words and logos that are identified as trademarks and/or service marks are, unless noted otherwise, the trademarks and
service marks of Altera Corporation in the U.S. and other countries. All other product or service names are the property of their respective holders. Altera products are protected under numerous U.S. and foreign patents and pending applications, maskwork rights, and copyrights. Altera warrants
performance of its semiconductor products to current specifications in accordance with Altera's standard warranty, but reserves the right to make
changes to any products and services at any time without notice. Altera assumes no responsibility or liability arising out of the application or use of any information, product, or service described herein except as
expressly agreed to in writing by Altera Corporation. Altera customers are advised to obtain the latest version of device specifications before relying on any published information and before placing orders for
products or services.
ii
Altera Corporation
Contents
Chapter Revision Dates ......................................................................... xiii
About This Handbook ............................................................................. xv
How to Find Information ...................................................................................................................... xv
How to Contact Altera ........................................................................................................................... xv
Typographic Conventions .................................................................................................................... xvi
Section I. Clock Management
Revision History ....................................................................................................................... Section I–1
Chapter 1. General-Purpose PLLs in Stratix & Stratix GX Devices
Introduction ............................................................................................................................................ 1–1
Enhanced PLLs ....................................................................................................................................... 1–5
Clock Multiplication & Division .................................................................................................... 1–9
External Clock Outputs ................................................................................................................. 1–10
Clock Feedback ............................................................................................................................... 1–14
Phase Shifting ................................................................................................................................. 1–14
Lock Detect ...................................................................................................................................... 1–15
Programmable Duty Cycle ........................................................................................................... 1–16
General Advanced Clear & Enable Control ............................................................................... 1–16
Programmable Bandwidth ............................................................................................................ 1–18
Clock Switchover ............................................................................................................................ 1–25
Spread-Spectrum Clocking ........................................................................................................... 1–25
PLL Reconfiguration ...................................................................................................................... 1–30
Enhanced PLL Pins ........................................................................................................................ 1–30
Fast PLLs ............................................................................................................................................... 1–31
Clock Multiplication & Division .................................................................................................. 1–34
External Clock Outputs ................................................................................................................. 1–34
Phase Shifting ................................................................................................................................. 1–35
Programmable Duty Cycle ........................................................................................................... 1–36
Control Signals ................................................................................................................................ 1–36
Pins ................................................................................................................................................... 1–37
Clocking ................................................................................................................................................ 1–39
Global & Hierarchical Clocking ................................................................................................... 1–39
Clock Input Connections ............................................................................................................... 1–41
Clock Output Connections ............................................................................................................ 1–43
Board Layout ........................................................................................................................................ 1–50
VCCA & GNDA ............................................................................................................................. 1–50
Altera Corporation
iii
Contents
Stratix Device Handbook, Volume 2
VCCG & GNDG ..............................................................................................................................
External Clock Output Power ......................................................................................................
Guidelines ........................................................................................................................................
Conclusion ............................................................................................................................................
1–52
1–53
1–56
1–56
Section II. Memory
Revision History ..................................................................................................................... Section II–1
Chapter 2. TriMatrix Embedded Memory Blocks in Stratix & Stratix GX Devices
Introduction ............................................................................................................................................ 2–1
TriMatrix Memory ................................................................................................................................. 2–1
Clear Signals ...................................................................................................................................... 2–3
Parity Bit Support ............................................................................................................................. 2–3
Byte Enable Support ........................................................................................................................ 2–4
Using TriMatrix Memory ..................................................................................................................... 2–7
Implementing Single-Port Mode .................................................................................................... 2–7
Implementing Simple Dual-Port Mode ......................................................................................... 2–8
Implementing True Dual-Port Mode .......................................................................................... 2–11
Implementing Shift-Register Mode ............................................................................................. 2–14
Implementing ROM Mode ............................................................................................................ 2–15
Implementing FIFO Buffers .......................................................................................................... 2–16
Clock Modes ......................................................................................................................................... 2–16
Independent Clock Mode .............................................................................................................. 2–16
Input/Output Clock Mode ........................................................................................................... 2–18
Read/Write Clock Mode ............................................................................................................... 2–21
Single-Port Mode ............................................................................................................................ 2–23
Designing With TriMatrix Memory .................................................................................................. 2–23
Selecting TriMatrix Memory Blocks ............................................................................................ 2–24
Pipeline & Flow-Through Modes ................................................................................................ 2–24
Power-up Conditions & Memory Initialization ......................................................................... 2–25
Read-During-Write Operation at the Same Address ..................................................................... 2–25
Same-Port Read-During-Write Mode .......................................................................................... 2–25
Mixed-Port Read-During-Write Mode ........................................................................................ 2–26
Conclusion ............................................................................................................................................ 2–27
Chapter 3. External Memory Interfaces in Stratix & Stratix GX Devices
Introduction ............................................................................................................................................ 3–1
External Memory Standards ................................................................................................................ 3–1
DDR SDRAM .................................................................................................................................... 3–1
RLDRAM II ....................................................................................................................................... 3–4
QDR & QDRII SRAM ...................................................................................................................... 3–6
ZBT SRAM ......................................................................................................................................... 3–8
DDR Memory Support Overview ..................................................................................................... 3–10
DDR Memory Interface Pins ......................................................................................................... 3–11
DQS Phase-Shift Circuitry ............................................................................................................ 3–15
iv
Altera Corporation
Contents
Contents
DDR Registers ................................................................................................................................. 3–20
PLL ................................................................................................................................................... 3–27
Conclusion ............................................................................................................................................ 3–27
Section III. I/O Standards
Revision History .................................................................................................................... Section III–1
Chapter 4. Selectable I/O Standards in Stratix & Stratix GX Devices
Introduction ............................................................................................................................................ 4–1
Stratix & Stratix GX I/O Standards .................................................................................................... 4–1
3.3-V Low Voltage Transistor-Transistor Logic (LVTTL) - EIA/JEDEC Standard JESD8-B . 4–2
3.3-V LVCMOS - EIA/JEDEC Standard JESD8-B ........................................................................ 4–3
2.5-V LVTTL Normal Voltage Range - EIA/JEDEC Standard EIA/JESD8-5 .......................... 4–3
2.5-V LVCMOS Normal Voltage Range - EIA/JEDEC Standard EIA/JESD8-5 ..................... 4–3
1.8-V LVTTL Normal Voltage Range - EIA/JEDEC Standard EIA/JESD8-7 .......................... 4–4
1.8-V LVCMOS Normal Voltage Range - EIA/JEDEC Standard EIA/JESD8-7 ..................... 4–4
1.5-V LVCMOS Normal Voltage Range - EIA/JEDEC Standard JESD8-11 ............................ 4–4
1.5-V HSTL Class I & II - EIA/JEDEC Standard EIA/JESD8-6 ................................................. 4–5
1.5-V Differential HSTL - EIA/JEDEC Standard EIA/JESD8-6 ................................................ 4–6
3.3-V PCI Local Bus - PCI Special Interest Group PCI Local Bus Specification Rev. 2.3 ....... 4–6
3.3-V PCI-X 1.0 Local Bus - PCI-SIG PCI-X Local Bus Specification Revision 1.0a ................ 4–7
3.3-V Compact PCI Bus - PCI SIG PCI Local Bus Specification Revision 2.3 .......................... 4–7
3.3-V 1× AGP - Intel Corporation Accelerated Graphics Port Interface Specification 2.0 ..... 4–7
3.3-V 2× AGP - Intel Corporation Accelerated Graphics Port Interface Specification 2.0 ..... 4–8
GTL - EIA/JEDEC Standard EIA/JESD8-3 .................................................................................. 4–8
GTL+ .................................................................................................................................................. 4–8
CTT - EIA/JEDEC Standard JESD8-4 ............................................................................................ 4–9
SSTL-3 Class I & II - EIA/JEDEC Standard JESD8-8 .................................................................. 4–9
SSTL-2 Class I & II - EIA/JEDEC Standard JESD8-9A ............................................................. 4–10
SSTL-18 Class I & II - EIA/JEDEC Preliminary Standard JC42.3 ............................................ 4–11
Differential SSTL-2 - EIA/JEDEC Standard JESD8-9A ............................................................. 4–11
LVDS - ANSI/TIA/EIA Standard ANSI/TIA/EIA-644 .......................................................... 4–12
LVPECL ........................................................................................................................................... 4–13
Pseudo Current Mode Logic (PCML) ......................................................................................... 4–13
HyperTransport Technology - HyperTransport Consortium ................................................. 4–14
High-Speed Interfaces ......................................................................................................................... 4–15
OIF-SPI4.2 ........................................................................................................................................ 4–15
OIF-SFI4.1 ........................................................................................................................................ 4–15
10 Gigabit Ethernet Sixteen Bit Interface (XSBI) - IEEE Draft Standard P802.3ae/D2.0 ...... 4–16
RapidIO Interconnect Specification Revision 1.1 ....................................................................... 4–16
HyperTransport Technology - HyperTransport Consortium ................................................. 4–17
UTOPIA Level 4 – ATM Forum Technical Committee Standard AF-PHY-0144.001 ........... 4–17
Stratix & Stratix GX I/O Banks .......................................................................................................... 4–17
Non-Voltage-Referenced Standards ............................................................................................ 4–24
Voltage-Referenced Standards ..................................................................................................... 4–24
Altera Corporation
v
Contents
Stratix Device Handbook, Volume 2
Mixing Voltage Referenced & Non-Voltage Referenced Standards ....................................... 4–25
Drive Strength ...................................................................................................................................... 4–26
Standard Current Drive Strength ................................................................................................. 4–26
Programmable Current Drive Strength ...................................................................................... 4–27
Hot Socketing ....................................................................................................................................... 4–27
DC Hot Socketing Specification ................................................................................................... 4–28
AC Hot Socketing Specification ................................................................................................... 4–28
I/O Termination .................................................................................................................................. 4–28
Voltage-Referenced I/O Standards ............................................................................................. 4–28
Differential I/O Standards ............................................................................................................ 4–29
Differential Termination (RD) ...................................................................................................... 4–29
Transceiver Termination ............................................................................................................... 4–30
I/O Pad Placement Guidelines .......................................................................................................... 4–30
Differential Pad Placement Guidelines ....................................................................................... 4–30
VREF Pad Placement Guidelines ................................................................................................. 4–31
Output Enable Group Logic Option in Quartus II .................................................................... 4–34
Toggle Rate Logic Option in Quartus II ...................................................................................... 4–35
DC Guidelines ................................................................................................................................. 4–35
Power Source of Various I/O Standards ......................................................................................... 4–38
Quartus II Software Support .............................................................................................................. 4–38
Compiler Settings ........................................................................................................................... 4–38
Device & Pin Options .................................................................................................................... 4–39
Assign Pins ...................................................................................................................................... 4–39
Programmable Drive Strength Settings ...................................................................................... 4–40
I/O Banks in the Floorplan View ................................................................................................. 4–40
Auto Placement & Verification of Selectable I/O Standards ................................................... 4–41
Conclusion ............................................................................................................................................ 4–42
More Information ................................................................................................................................ 4–42
References ............................................................................................................................................. 4–42
Chapter 5. High-Speed Differential I/O Interfaces in Stratix Devices
Introduction ............................................................................................................................................ 5–1
Stratix I/O Banks ................................................................................................................................... 5–1
Stratix Differential I/O Standards ................................................................................................. 5–2
Stratix Differential I/O Pin Location ............................................................................................. 5–5
Principles of SERDES Operation ......................................................................................................... 5–6
Stratix Differential I/O Receiver Operation ................................................................................. 5–7
Stratix Differential I/O Transmitter Operation ........................................................................... 5–9
Transmitter Clock Output ............................................................................................................. 5–10
Divided-Down Transmitter Clock Output ................................................................................. 5–10
Center-Aligned Transmitter Clock Output ................................................................................ 5–11
SDR Transmitter Clock Output .................................................................................................... 5–12
Using SERDES to Implement DDR ................................................................................................... 5–13
Using SERDES to Implement SDR .................................................................................................... 5–14
Differential I/O Interface & Fast PLLs ............................................................................................. 5–16
Clock Input & Fast PLL Output Relationship ............................................................................ 5–18
Fast PLL Specifications .................................................................................................................. 5–20
vi
Altera Corporation
Contents
Contents
High-Speed Phase Adjust ............................................................................................................. 5–21
Counter Circuitry ........................................................................................................................... 5–22
Fast PLL SERDES Channel Support ............................................................................................ 5–23
Advanced Clear & Enable Control .............................................................................................. 5–25
Receiver Data Realignment ................................................................................................................ 5–25
Data Realignment Principles of Operation ................................................................................. 5–25
Generating the TXLOADEN Signal ............................................................................................. 5–27
Realignment Implementation ....................................................................................................... 5–28
Source-Synchronous Timing Budget ................................................................................................ 5–30
Differential Data Orientation ........................................................................................................ 5–30
Differential I/O Bit Position ......................................................................................................... 5–31
Timing Definition ........................................................................................................................... 5–32
Input Timing Waveform ............................................................................................................... 5–39
Output Timing ................................................................................................................................ 5–40
Receiver Skew Margin ................................................................................................................... 5–40
Switching Characteristics .............................................................................................................. 5–42
Timing Analysis .............................................................................................................................. 5–42
SERDES Bypass DDR Differential Signaling ................................................................................... 5–42
SERDES Bypass DDR Differential Interface Review ................................................................. 5–42
SERDES Clock Domains ................................................................................................................ 5–42
SERDES Bypass DDR Differential Signaling Receiver Operation .......................................... 5–43
SERDES Bypass DDR Differential Signaling Transmitter Operation ..................................... 5–44
High-Speed Interface Pin Locations ................................................................................................. 5–45
Differential I/O Termination ............................................................................................................. 5–46
RD Differential Termination .......................................................................................................... 5–46
HyperTransport & LVPECL Differential Termination ............................................................. 5–47
PCML Differential Termination ................................................................................................... 5–47
Differential HSTL Termination .................................................................................................... 5–48
Differential SSTL-2 Termination .................................................................................................. 5–49
Board Design Consideration .............................................................................................................. 5–50
Software Support ................................................................................................................................. 5–51
Differential Pins in Stratix ............................................................................................................. 5–51
Fast PLLs .......................................................................................................................................... 5–52
LVDS Receiver Block ..................................................................................................................... 5–60
LVDS Transmitter Module ........................................................................................................... 5–65
SERDES Bypass Mode ................................................................................................................... 5–70
Summary ............................................................................................................................................... 5–75
Section IV. Digital Signal Processing (DSP)
Revision History .................................................................................................................... Section IV–1
Chapter 6. DSP Blocks in Stratix & Stratix GX Devices
Introduction ............................................................................................................................................ 6–1
DSP Block Overview ............................................................................................................................. 6–2
Architecture ............................................................................................................................................ 6–5
Altera Corporation
vii
Contents
Stratix Device Handbook, Volume 2
Multiplier Block ................................................................................................................................ 6–5
Adder/Output Block ....................................................................................................................... 6–9
Routing Structure & Control Signals ........................................................................................... 6–12
Operational Modes .............................................................................................................................. 6–18
Simple Multiplier Mode ................................................................................................................ 6–18
Multiply Accumulator Mode ........................................................................................................ 6–22
Two-Multiplier Adder Mode ........................................................................................................ 6–23
Four-Multiplier Adder Mode ....................................................................................................... 6–24
Software Support ................................................................................................................................. 6–28
Conclusion ............................................................................................................................................ 6–28
Chapter 7. Implementing High Performance DSP Functions
in Stratix & Stratix GX Devices
Introduction ............................................................................................................................................ 7–1
Stratix & Stratix GX DSP Block Overview ......................................................................................... 7–1
TriMatrix Memory Overview .............................................................................................................. 7–4
DSP Function Overview ....................................................................................................................... 7–5
Finite Impulse Response (FIR) Filters ................................................................................................. 7–5
FIR Filter Background ...................................................................................................................... 7–6
Basic FIR Filter .................................................................................................................................. 7–7
Time-Domain Multiplexed FIR Filters ........................................................................................ 7–13
Polyphase FIR Interpolation Filters ............................................................................................. 7–17
Polyphase FIR Decimation Filters ................................................................................................ 7–24
Complex FIR Filter ......................................................................................................................... 7–31
Infinite Impulse Response (IIR) Filters ............................................................................................. 7–34
IIR Filter Background .................................................................................................................... 7–34
Basic IIR Filters ............................................................................................................................... 7–36
Butterworth IIR Filters ................................................................................................................... 7–39
Matrix Manipulation ........................................................................................................................... 7–45
Background on Matrix Manipulation .......................................................................................... 7–45
Two-Dimensional Filtering & Video Imaging ........................................................................... 7–46
Discrete Cosine Transform (DCT) ..................................................................................................... 7–52
DCT Background ............................................................................................................................ 7–52
2-D DCT Algorithm ....................................................................................................................... 7–53
Arithmetic Functions ........................................................................................................................... 7–59
Background ..................................................................................................................................... 7–59
Arithmetic Function Implementation ......................................................................................... 7–60
Arithmetic Function Implementation Results ............................................................................ 7–62
Arithmetic Function Design Example ......................................................................................... 7–62
Conclusion ............................................................................................................................................ 7–62
References ............................................................................................................................................. 7–63
Section V. IP & Design Considerations
Revision History ..................................................................................................................... Section V–1
viii
Altera Corporation
Contents
Contents
Chapter 8. Implementing 10-Gigabit Ethernet Using Stratix & Stratix GX Devices
Introduction ............................................................................................................................................ 8–1
Related Links ..................................................................................................................................... 8–1
10-Gigabit Ethernet ................................................................................................................................ 8–1
Interfaces ................................................................................................................................................. 8–5
XSBI .................................................................................................................................................... 8–5
XGMII ............................................................................................................................................... 8–13
XAUI ................................................................................................................................................. 8–19
I/O Characteristics for XSBI, XGMII & XAUI ................................................................................. 8–21
Software Implementation .............................................................................................................. 8–22
AC/DC Specifications ................................................................................................................... 8–22
10-Gigabit Ethernet MAC Core .................................................................................................... 8–24
Conclusion ....................................................................................................................................... 8–25
Chapter 9. Implementing SFI-4 in Stratix & Stratix GX Devices
Introduction ............................................................................................................................................ 9–1
System Topology .............................................................................................................................. 9–3
Interface Implementation in Stratix & Stratix GX Devices ......................................................... 9–5
AC Timing Specifications .............................................................................................................. 9–10
Electrical Specifications ................................................................................................................. 9–12
Software Implementation .............................................................................................................. 9–13
Conclusion ....................................................................................................................................... 9–13
Chapter 10. Transitioning APEX Designs to Stratix & Stratix GX Devices
Introduction .......................................................................................................................................... 10–1
General Architecture ........................................................................................................................... 10–1
Logic Elements ................................................................................................................................ 10–2
MultiTrack Interconnect ................................................................................................................ 10–3
DirectDrive Technology ................................................................................................................ 10–4
Architectural Element Names ...................................................................................................... 10–5
TriMatrix Memory ............................................................................................................................... 10–8
Same-Port Read-During-Write Mode ........................................................................................ 10–10
Mixed-Port Read-During-Write Mode ...................................................................................... 10–11
Memory Megafunctions .............................................................................................................. 10–12
FIFO Conditions ........................................................................................................................... 10–13
Design Migration Mode in Quartus II Software ...................................................................... 10–13
DSP Block ............................................................................................................................................ 10–16
DSP Block Megafunctions ........................................................................................................... 10–16
PLLs & Clock Networks ................................................................................................................... 10–18
Clock Networks ............................................................................................................................ 10–18
PLLs ................................................................................................................................................ 10–19
I/O Structure ...................................................................................................................................... 10–25
External RAM Interfacing ........................................................................................................... 10–25
I/O Standard Support ................................................................................................................. 10–26
High-Speed Differential I/O Standards .................................................................................... 10–26
altlvds Megafunction ................................................................................................................... 10–29
Configuration ..................................................................................................................................... 10–30
Altera Corporation
ix
Contents
Stratix Device Handbook, Volume 2
Configuration Speed & Schemes ................................................................................................
Remote Update Configuration ...................................................................................................
JTAG Instruction Support ...........................................................................................................
Conclusion ..........................................................................................................................................
10–30
10–31
10–31
10–32
Section VI. System Configuration & Upgrades
Revision History .................................................................................................................... Section VI–2
Chapter 11. Configuring Stratix & Stratix GX Devices
Introduction .......................................................................................................................................... 11–1
Device Configuration Overview ....................................................................................................... 11–2
MSEL[2..0] Pins ............................................................................................................................... 11–3
VCCSEL Pins ...................................................................................................................................... 11–3
PORSEL Pins ................................................................................................................................... 11–5
nIO_PULLUP Pins ......................................................................................................................... 11–5
TDO & nCEO Pins .......................................................................................................................... 11–6
Configuration File Size ....................................................................................................................... 11–6
Altera Configuration Devices ............................................................................................................ 11–7
Configuration Schemes ....................................................................................................................... 11–7
PS Configuration ............................................................................................................................ 11–7
FPP Configuration ........................................................................................................................ 11–21
PPA Configuration ....................................................................................................................... 11–30
JTAG Programming & Configuration ....................................................................................... 11–36
JTAG Programming & Configuration of Multiple Devices ................................................... 11–39
Configuration with JRunner Software Driver .......................................................................... 11–41
Jam STAPL Programming & Test Language ............................................................................ 11–42
Configuring Using the MicroBlaster Driver .................................................................................. 11–51
Device Configuration Pins ............................................................................................................... 11–51
Chapter 12. Remote System Configuration with Stratix & Stratix GX Devices
Introduction .......................................................................................................................................... 12–1
Remote Configuration Operation ...................................................................................................... 12–1
Remote System Configuration Modes ........................................................................................ 12–3
Remote System Configuration Components .............................................................................. 12–5
Quartus II Software Support ............................................................................................................ 12–12
altremote_update Megafunction ................................................................................................ 12–14
Remote Update WYSIWYG ATOM ........................................................................................... 12–17
Using Enhanced Configuration Devices ........................................................................................ 12–19
Local Update Programming File Generation ........................................................................... 12–21
Remote Update Programming File Generation ....................................................................... 12–32
Combining MAX Devices & Flash Memory .................................................................................. 12–42
Using an External Processor ............................................................................................................ 12–43
Conclusion .......................................................................................................................................... 12–44
x
Altera Corporation
Contents
Contents
Section VII. PCB Layout Guidelines
Revision History .................................................................................................................. Section VII–1
Chapter 13. Package Information for Stratix Devices
Introduction .......................................................................................................................................... 13–1
Device & Package Cross Reference ................................................................................................... 13–1
Thermal Resistance .............................................................................................................................. 13–2
Package Outlines ................................................................................................................................. 13–3
484-Pin FineLine BGA - Flip Chip ............................................................................................... 13–4
672-Pin FineLine BGA - Flip Chip ............................................................................................... 13–6
780-Pin FineLine BGA - Flip Chip ............................................................................................... 13–8
956-Pin Ball Grid Array (BGA) - Flip Chip ............................................................................... 13–10
1,020-Pin FineLine BGA - Flip Chip .......................................................................................... 13–12
1,508-Pin FineLine BGA - Flip Chip .......................................................................................... 13–14
Chapter 14. Designing with 1.5-V Devices
Introduction .......................................................................................................................................... 14–1
Power Sequencing & Hot Socketing ................................................................................................. 14–1
Using MultiVolt I/O Pins ................................................................................................................... 14–2
Voltage Regulators .............................................................................................................................. 14–3
Linear Voltage Regulators ............................................................................................................. 14–5
Switching Voltage Regulators ...................................................................................................... 14–7
Maximum Output Current ........................................................................................................... 14–8
Selecting Voltage Regulators ........................................................................................................ 14–9
Voltage Divider Network ............................................................................................................ 14–10
1.5-V Regulator Circuits .............................................................................................................. 14–10
1.5-V Regulator Application Examples .......................................................................................... 14–19
Synchronous Switching Regulator Example ............................................................................ 14–20
Board Layout ...................................................................................................................................... 14–21
Split-Plane Method ....................................................................................................................... 14–23
Conclusion .......................................................................................................................................... 14–23
References ........................................................................................................................................... 14–24
Altera Corporation
xi
Contents
xii
Stratix Device Handbook, Volume 2
Altera Corporation
Chapter Revision Dates
The chapters in this book, Stratix Device Handbook, Volume 2, were revised on the following dates.
Where chapters or groups of chapters are available separately, part numbers are listed.
Chapter 1. General-Purpose PLLs in Stratix & Stratix GX Devices
Revised:
July 2005
Part number: S52001-3.2
Chapter 2. TriMatrix Embedded Memory Blocks in Stratix & Stratix GX Devices
Revised:
July 2005
Part number: S52003-3.3
Chapter 3. External Memory Interfaces in Stratix & Stratix GX Devices
Revised:
June 2006
Part number: SII52003-3.3
Chapter 4.
Selectable I/O Standards in Stratix & Stratix GX Devices
Revised:
June 2006
Part number: S52004-3.4
Chapter 5. High-Speed Differential I/O Interfaces in Stratix Devices
Revised:
July 2005
Part number: S52005-3.2
Chapter 6. DSP Blocks in Stratix & Stratix GX Devices
Revised:
July 2005
Part number: S52006-2.2
Chapter 7. Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
Revised:
September 2004
Part number: S52007-1.1
Chapter 8. Implementing 10-Gigabit Ethernet Using Stratix & Stratix GX Devices
Revised:
July 2005
Part number: S52010-2.0
Chapter 9. Implementing SFI-4 in Stratix & Stratix GX Devices
Revised:
July 2005
Part number: S52011-2.0
Altera Corporation
xiii
Chapter Revision Dates
Stratix Device Handbook, Volume 2
Chapter 10. Transitioning APEX Designs to Stratix & Stratix GX Devices
Revised:
July 2005
Part number: S52012-3.0
Chapter 11. Configuring Stratix & Stratix GX Devices
Revised:
July 2005
Part number: S52013-3.2
Chapter 12. Remote System Configuration with Stratix & Stratix GX Devices
Revised:
September 2004
Part number: S52015-3.1
Chapter 13. Package Information for Stratix Devices
Revised:
July 2005
Part number: S53008-3.0
Chapter 14. Designing with 1.5-V Devices
Revised:
January 2005
Part number: C51012-1.1
xiv
Altera Corporation
About This Handbook
This handbook provides comprehensive information about the Altera®
Stratix® family of devices.
How to Find
Information
You can find more information in the following ways:
■
■
■
■
How to Contact
Altera
Information Type
Technical support
Product literature
The Adobe Acrobat Find feature, which searches the text of a PDF
document. Click the binoculars toolbar icon to open the Find dialog
box.
Acrobat bookmarks, which serve as an additional table of contents in
PDF documents.
Thumbnail icons, which provide miniature previews of each page
and provide a link to the pages.
Numerous links, shown in green text, which allow you to jump to
related information.
For the most up-to-date information about Altera products, go to the
Altera world-wide web site at www.altera.com. For technical support on
this product, go to www.altera.com/mysupport. For additional
information about Altera products, consult the sources shown below.
USA & Canada
All Other Locations
www.altera.com/mysupport/
www.altera.com/mysupport/
(800) 800-EPLD (3753)
(7:00 a.m. to 5:00 p.m. Pacific Time)
+1 408-544-8767
7:00 a.m. to 5:00 p.m. (GMT -8:00)
Pacific Time
www.altera.com
www.altera.com
Altera literature services
[email protected]
[email protected]
Non-technical customer
service
(800) 767-3753
+ 1 408-544-7000
7:00 a.m. to 5:00 p.m. (GMT -8:00)
Pacific Time
FTP site
ftp.altera.com
ftp.altera.com
Altera Corporation
xv
Typographic Conventions
Typographic
Conventions
Visual Cue
Stratix Device Handbook, Volume 2
This document uses the typographic conventions shown below.
Meaning
Bold Type with Initial
Capital Letters
Command names, dialog box titles, checkbox options, and dialog box options are
shown in bold, initial capital letters. Example: Save As dialog box.
bold type
External timing parameters, directory names, project names, disk drive names,
filenames, filename extensions, and software utility names are shown in bold
type. Examples: fMAX, \qdesigns directory, d: drive, chiptrip.gdf file.
Italic Type with Initial Capital
Letters
Document titles are shown in italic type with initial capital letters. Example: AN 75:
High-Speed Board Designs.
Italic type
Internal timing parameters and variables are shown in italic type.
Examples: tPIA, n + 1.
Variable names are enclosed in angle brackets (< >) and shown in italic type.
Example: <file name>, <project name>.pof file.
Initial Capital Letters
Keyboard keys and menu names are shown with initial capital letters. Examples:
Delete key, the Options menu.
“Subheading Title”
References to sections within a document and titles of on-line help topics are
shown in quotation marks. Example: “Typographic Conventions.”
Courier type
Signal and port names are shown in lowercase Courier type. Examples: data1,
tdi, input. Active-low signals are denoted by suffix n, e.g., resetn.
Anything that must be typed exactly as it appears is shown in Courier type. For
example: c:\qdesigns\tutorial\chiptrip.gdf. Also, sections of an
actual file, such as a Report File, references to parts of files (e.g., the AHDL
keyword SUBDESIGN), as well as logic function names (e.g., TRI) are shown in
Courier.
1., 2., 3., and
a., b., c., etc.
Numbered steps are used in a list of items when the sequence of the items is
important, such as the steps listed in a procedure.
■
Bullets are used in a list of items when the sequence of the items is not important.
●
•
v
The checkmark indicates a procedure that consists of one step only.
1
The hand points to information that requires special attention.
r
The angled arrow indicates you should press the Enter key.
f
The feet direct you to more information on a particular topic.
xvi
Altera Corporation
Section I. Clock
Management
This section provides information on the different types of phase-lock
loops (PLLs). The feature-rich, enhanced PLLs assist you in managing
clocks internally and also have the ability to drive off-chip to control
system-level clock networks. The fast PLLs offer general-purpose clock
management with multiplication and phase shifting as well as
high-speed outputs to manage the high-speed differential I/O interfaces.
This chapter contains detailed information on the features, the
interconnections to the core and off-chip, and the specifications for both
types of PLLs.
This section contains the following:
■
Revision History
Chapter
Date/Version
1
July 2005, v3.2
Chapter 1, General-Purpose PLLs in Stratix & Stratix GX Devices
The table below shows the revision history for Chapter 1.
Changes Made
●
●
●
●
●
●
September 2004, v3.1
●
●
●
April 2004, v3.0
●
●
●
●
●
●
●
●
●
●
Altera Corporation
Removed information regarding delay shift (time delay elements).
Updated Table 1–8.
Updated “Clock Switchover” section.
Updated Figure 1–22.
Updated “Control Signals” section.
Updated Table 1–16.
Updated Note 1 in Table 1–17 on page 1–32.
Updated Note 1 in Table 1–21 on page 1–48.
Updated Table 1–12 on page 1–34.
Changed PCI-X to PCI-X 1.0 throughout volume.
Note 3 added to columns 11 and 12 in Table 1–1.
Deleted “Stratix GX Clock Input Sources for Enhanced and Fast PLLs”
table.
Deleted “Stratix GX Global and Regional Clock Output Line Sharing for
Enhanced and Fast PLLS” table.
Deleted “Stratix GX CLK and FPLLCLK Input Pin Connections to Global
& Regional Clock Networks” table.
Changed CLK checkmarks in Table 1–14.
Updated notes to Table 1–3. and Figure 1–3.
Added Table 1–7.
Clock Switchover section has been moved to AN 313.
Changed RCLK values in Figures 1–20 and 1–22.
Section I–1
Clock Management
Stratix Device Handbook, Volume 2
Chapter
Date/Version
Changes Made
1
November 2003, v2.2
●
Updated the “Lock Detect” section.
October 2003, v2.1
●
Updated the “VCCG & GNDG” section.
Updated Figure 1–14.
●
July 2003, v2.0
●
●
●
●
●
●
●
●
●
Section I–2
Updated clock multiplication and division, spread spectrum, and Notes 1
and 8 in Table 1-3.
Updated inclk[1..0] port name in Table 1-4.
Updated ranges for EPLL post-scale and pre-scale dividers on page 1-9
Added 1.8V HSTL support for EPLL in Table 1-6 and 1-13.
New requirement to assert are set signal each PLL when it has to reacquire lock on either a new clock after loss of lock (page 1-16)
Corrected input port extswitch to clkswitch throughout this
chapter.
Updated clkloss description in Table 1-9.
Updated text on jitter for spread spectrum on page 1-38.
Removed PLL specifications. See Chapter 4 of Volume 1.
Altera Corporation
1. General-Purpose PLLs in
Stratix & Stratix GX Devices
S52001-3.2
Introduction
Stratix® and Stratix GX devices have highly versatile phase-locked loops
(PLLs) that provide robust clock management and synthesis for on-chip
clock management, external system clock management, and high-speed
I/O interfaces. There are two types of PLLs in each Stratix and Stratix GX
device: enhanced PLLs and fast PLLs. Each device has up to four
enhanced PLLs, which are feature-rich, general-purpose PLLs supporting
advanced capabilities such as external feedback, clock switchover, phase
and delay control, PLL reconfiguration, spread spectrum clocking, and
programmable bandwidth. There are also up to eight fast PLLs per
device, which offer general-purpose clock management with
multiplication and phase shifting as well as high-speed outputs to
manage the high-speed differential I/O interfaces.
The Altera® Quartus® II software enables the PLLs and their features
without requiring any external devices.
Tables 1–1 and 1–2 show PLL availability for Stratix and Stratix GX
devices, respectively.
Table 1–1. Stratix Device PLL Availability
Fast PLLs
Enhanced PLLs
Device
1
2
3
4
EP1S10
v
v
v
EP1S20
v
v
EP1S25
v
EP1S30
7
8
9
10
5(1)
6(1)
11(2)
12(2)
v
v
v
v
v
v
v
v
v
v
v
v
v
v
v
v
v (3)
v (3)
v (3)
v (3)
v
v
EP1S40
v
v
v
v
v (3)
v (3)
v (3)
v (3)
v
v
v (3)
v (3)
EP1S60
v
v
v
v
v
v
v
v
v
v
v
v
EP1S80
v
v
v
v
v
v
v
v
v
v
v
v
Notes to Table 1–1:
(1)
(2)
PLLs 5 and 6 each have eight single-ended outputs or four differential outputs.
PLLs 11 and 12 each have one single-ended output.
(3)
EP1S30 and EP1S40 devices do not support these PLLs in the 780-pin FineLine BGA® package.
Altera Corporation
July 2005
1–1
Introduction
Table 1–2. Stratix GX Device PLL Availability
Fast PLLs
Enhanced PLLs
Device
1
2
EP1S10C
v
EP1S10D
5
6
v
v
v
v
v
v
v
EP1S25C
v
v
v
v
EP1S25D
v
v
v
v
EP1S25F
v
v
v
v
EP1S40D
v
v
v
v
v
EP1S40G
v
v
v
v
v
1–2
Stratix Device Handbook, Volume 2
7
8
11
12
v
v
v
v
v
v
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
Table 1–3 shows the enhanced and fast PLL features in Stratix and
Stratix GX devices.
Table 1–3. Stratix & Stratix GX PLL Features
Feature
Enhanced PLL
Fast PLL
Clock multiplication and division
m/(n × post-scale counter) (1)
m/(post-scale counter) (2)
Down to 156.25-ps increments (3), (4)
Down to 125-ps increments (3), (4)
Phase shift
Clock switchover
v
PLL reconfiguration
v
Programmable bandwidth
v
Spread spectrum clocking
v
Programmable duty cycle
v
v
Number of internal clock outputs
6
3 (5)
Number of external clock outputs
Four differential/eight singled-ended
or one single-ended (6)
(7)
Number of feedback clock inputs
2 (8)
Notes to Table 1–3:
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
For enhanced PLLs, m, n, range from 1 to 512 and post-scale counters g, l, e range from 1 to 1024 with 50% duty
cycle. With a non-50% duty cycle the post-scale counters g, l, e range from 1 to 512.
For fast PLLs, m, n, and post-scale counters range from 1 to 32.
The smallest phase shift is determined by the voltage controlled oscillator (VCO) period divided by 8.
For degree increments, Stratix and Stratix GX devices can shift all output frequencies in increments of at least 45° .
Smaller degree increments are possible depending on the frequency and divide parameters.
PLLs 7, 8, 9, and 10 have two output ports per PLL. PLLs 1, 2, 3, and 4 have three output ports per PLL. On Stratix
GX devices, PLLs 3, 4, 9, and 10 are not available for general-purpose use.
Every Stratix and Stratix GX device has two enhanced PLLs (PLLs 5 and 6) with either eight single-ended outputs
or four differential outputs each. Two additional enhanced PLLs (PLLs 11 and 12) in EP1S80, EP1S60, EP1S40 (PLL
11 and 12 not supported for F780 package), and EP1SGX40 devices each have one single-ended output.
Fast PLLs can drive to any I/O pin as an external clock. For high-speed differential I/O pins, the device uses a data
channel to generate txclkout.
Every Stratix and Stratix GX device has two enhanced PLLs with one single-ended or differential external feedback
input per PLL.
Altera Corporation
July 2005
1–3
Stratix Device Handbook, Volume 2
Introduction
Figure 1–1 shows a top-level diagram of the Stratix device and PLL
floorplan. Figure 1–2 shows a top-level diagram of the Stratix GX device
and PLL floorplan. See “Clocking” on page 1–39 for more detail on PLL
connections to global and regional clocks.
Figure 1–1. Stratix PLL Locations
CLK12-15
5
11
FPLL7CLK
7
10
FPLL10CLK
CLK0-3
1
2
4
3
CLK8-11
8
9
FPLL9CLK
PLLs
FPLL8CLK
6
12
CLK4-7
1–4
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
Figure 1–2. Stratix GX PLL Locations
CLK12-15
5
LVDSCLK0
11
7
HSSI
CLK0-3
1
2
PLLs
LVDSCLK1
HSSI
8
6
12
CLK4-7
Enhanced PLLs
Altera Corporation
July 2005
Stratix and Stratix GX devices contain up to four enhanced PLLs with
advanced clock management features. Figure 1–3 shows a diagram of the
enhanced PLL.
1–5
Stratix Device Handbook, Volume 2
Enhanced PLLs
Figure 1–3. Stratix & Stratix GX Enhanced PLL
Programmable
Time Delay on
Each PLL Port
Post-Scale
Counters
VCO Phase Selection
Selectable at Each
PLL Output Port
From Adjacent PLL (4)
÷l0
Δt
Regional
Clocks
Clock
Switch-Over
Circuitry
÷l1
Δt
Spread
Spectrum
Phase Frequency
Detector (PFD)
INCLK0
4
Δtn
÷n
Charge
Pump
8
Loop
Filter
VCO
INCLK1
(1)
FBIN
Δtm
÷m
÷g0
Δt
÷g1
Δt
÷g2
Δt
÷g3
Δt
÷e0
Δt
÷e1
Δt
÷e2
Δt
÷e3
Δt
Global
Clocks
I/O Buffers (2)
to I/O or general
routing
Lock Detect
& Filter
VCO Phase Selection
Affecting All Outputs
4
I/O Buffers (3)
Notes to Figure 1–3:
(1)
(2)
(3)
(4)
External feedback is available in PLLs 5 and 6.
This single-ended external output is available from the g0 counter for PLLs 11 and 12.
These four counters and external outputs are available in PLLs 5 and 6.
This connection is only available on EP1SGX40 Stratix GX devices and EP1S40 and larger Stratix devices. For
example, PLLs 5 and 11 are adjacent and PLLs 6 and 12 are adjacent. The EP1S40 device in the F780 package does
not support PLLs 11 and 12.
1–6
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
Figure 1–4 shows all the possible ports of the enhanced PLLs.
Figure 1–4. Enhanced PLL Signals
(1)
pllenable
(2)
inclk0
(2)
inclk1
areset
clkswitch
scanclk
Physical Pin
clk[5..0]
Signal Driven by Internal Logic
Signal Driven to Internal Logic
Internal Clock Signal
locked
clkloss
clkbad[1..0]
active_clock
scandata
scanaclr
clkena[5..0]
Only PLLs
11 and 12
extclk4
scandataout
pfdena
(2)
fbin
Only PLLs
5 and 6
pll_out0p
pll_out0n
extclkena[3..0]
pll_out1p
pll_out1n
pll_out2p
(3)
pll_out2n
(3)
pll_out3p
(3)
pll_out3n
(3)
Notes to Figure 1–4:
(1)
(2)
(3)
This input pin is shared by all enhanced and fast PLLs.
These are either single-ended or differential pins.
EP1S10, EP1S20, and EP1S25 devices in 672-pin ball grid array (BGA) and 484- and 672-pin FineLine BGA packages
only have two pairs of external clocks (i.e., pll_out0p, pll_out0n, pll_out1p, and pll_out1n).
Altera Corporation
July 2005
1–7
Stratix Device Handbook, Volume 2
Enhanced PLLs
Tables 1–4 and 1–5 describe all the enhanced PLL ports.
Table 1–4. Enhanced PLL Input Signals
Port
Description
Source
Destination
inclk[1..0]
Primary and secondary reference clock inputs to
PLL
Pin
×n counter
fbin
External feedback input to the PLL (PLLs 5 and 6
only)
Pin
Phase frequency
detector (PFD)
pllena
Enable pin for enabling or disabling all or a set of
PLLs⎯active high
Pin
General PLL
control signal
clkswitch
Switchover signal used to initiate external clock
switchover control⎯this signal switches the clock
on the rising edge of clkswitch
Logic array
PLL switchover
circuit
areset
Signal used to reset the PLL which resynchronizes all the counter outputs⎯active high
Logic array
General PLL
control signal
clkena[5..0]
Enable clock driving regional or global
clock⎯active high
Logic array
Clock output
extclkena[3..0]
Enable clock driving external clock (PLLs 5 and 6
only)⎯active high
Logic array
Clock output
pfdena
Enables the outputs from the phase frequency
detector⎯active high
Logic array
PFD
scanclk
Serial clock signal for the real-time PLL control
feature
Logic array
Reconfiguration
circuit
scandata
Serial input data stream for the real-time PLL
control feature
Logic array
Reconfiguration
circuit
scanaclr
Serial shift register reset clearing all registers in
the serial shift chain⎯active high
Logic array
Reconfiguration
circuit
1–8
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
Table 1–5. Enhanced PLL Output Signals
Port
Description
Source
Destination
clk[5..0]
PLL outputs driving regional or global clock
PLL counter Internal Clock
pll_out[3..0]p/n
pll_out[3..0] are PLL outputs driving the four PLL counter Pin(s)
differential or eight single-ended external clock
output pins for PLLs 5 or 6. p or n are the positive
(p) and negative (n) pins for differential pins.
extclk4
PLL output driving external clock output pin from
PLLs 11 and 12
clkloss
Signal indicating the switchover circuit detected a PLL
switchover condition
switchover
circuit
Logic array
clkbad[1..0]
Signals indicating which reference clock is no
longer toggling. clkbad1 indicates inclk1
status, clkbad0 indicates inclk0 status
PLL
switchover
circuit
Logic array
locked
Lock output from lock detect circuit⎯active high
PLL lock
detect
Logic array
activeclock
Signal to indicate which clock (1 = inclk0 or
0 = inclk1) is driving the PLL.
PLL clock
multiplexer
Logic array
scandataout
Output of the last shift register in the scan chain
PLL scan
chain
Logic array
PLL g0
counter
Pin
Clock Multiplication & Division
Each Stratix and Stratix GX device enhanced PLL provides clock
synthesis for PLL output ports using m/(n × post-scale counter) scaling
factors. The input clock is divided by a pre-scale counter, n, and is then
multiplied by the m feedback factor. The control loop drives the VCO to
match fIN × (m/n). Each output port has a unique post-scale counter that
divides down the high-frequency VCO.
For multiple PLL outputs with different frequencies, the VCO is set to the
least common multiple of the output frequencies that meets its frequency
specifications. Then, the post-scale counters scale down the output
frequency for each output port. For example, if output frequencies
required from one PLL are 33 and 66 MHz, then the Quartus II software
sets the VCO to 330 MHz (the least common multiple of 33 and 66 MHz
within the VCO range).
There is one pre-scale counter, n, and one multiply counter, m, per PLL,
with a range of 1 to 512 on each. There are two post-scale counters (l) for
regional clock output ports, four counters (g) for global clock output
ports, and up to four counters (e) for external clock outputs, all ranging
from 1 to 1024 with a 50% duty cycle setting. The post-scale counters
Altera Corporation
July 2005
1–9
Stratix Device Handbook, Volume 2
Enhanced PLLs
range from 1 to 512 with any non-50% duty cycle setting. The Quartus II
software automatically chooses the appropriate scaling factors according
to the input frequency, multiplication, and division values entered into
the altpll MegaWizard Plug-In Manager.
External Clock Outputs
Enhanced PLLs 5 and 6 each support up to eight single-ended clock
outputs (or four differential pairs). See Figure 1–5.
1–10
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
Figure 1–5. External Clock Outputs for PLLs 5 & 6
From IOE (1), (2)
pll_out0p (3), (4)
(3)
e0 Counter
From IOE (1)
From IOE (1)
pll_out0n (3), (4)
pll_out1p (3), (4)
e1 Counter
4
From IOE (1)
From IOE (1)
pll_out1n (3), (4)
pll_out2p (3), (4)
e2 Counter
pll_out2n (3), (4)
From IOE (1)
From IOE (1)
pll_out3p (3), (4)
e3 Counter
From IOE (1)
pll_out3n (3), (4)
Notes to Figure 1–5:
(1)
(2)
(3)
(4)
LE: logic element.
The design can use each external clock output pin as a general-purpose output pin from the logic array. These pins
are multiplexed with IOE outputs.
Two single-ended outputs are possible per output counter⎯either two outputs of the same frequency and phase or
one shifted 180° .
EP1S10, EP1S20, and EP1S25 devices in 672-pin ball grid array (BGA) and 484- and 672-pin FineLine BGA packages
only have two pairs of external clocks (i.e., pll_out0p, pll_out0n, pll_out1p, and pll_out1n).
Any of the four external output counters can drive the single-ended or
differential clock outputs for PLLs 5 and 6. This means one counter or
frequency can drive all output pins available from PLL 5 or PLL 6. Each
Altera Corporation
July 2005
1–11
Stratix Device Handbook, Volume 2
Enhanced PLLs
pair of output pins (four pins total) has dedicated VCC and GND pins to
reduce the output clock’s overall jitter by providing improved isolation
from switching I/O pins.
For PLLs 5 and 6, each pin of a single-ended output pair can either be in
phase or 180° out of phase. The Quartus II software transfers the NOT
gate in the design into the IOE to implement 180° phase with respect to
the other pin in the pair. The clock output pin pairs support the same I/O
standards as standard output pins (in the top and bottom banks) as well
as LVDS, LVPECL, PCML, HyperTransportTM technology, differential
HSTL, and differential SSTL. Table 1–6 shows which I/O standards the
enhanced PLL clock pins support. When in single-ended or differential
mode, one power pin supports two differential or four single-ended pins.
Both outputs use the same standards in single-ended mode to maintain
performance. You can also use the external clock output pins as user
output pins if external enhanced PLL clocking is not needed.
The enhanced PLL can also drive out to any regular I/O pin through the
global or regional clock network. The jitter on the output clock is not
guaranteed for this case.
Table 1–6. I/O Standards Supported for Enhanced PLL Pins (Part 1 of 2)
Input
Output
I/O Standard
INCLK
FBIN
PLLENABLE
EXTCLK
LVTTL
v
v
v
v
LVCMOS
v
v
v
v
2.5 V
v
v
v
1.8 V
v
v
v
1.5 V
v
v
v
3.3-V PCI
v
v
v
3.3-V PCI-X 1.0
v
v
v
LVPECL
v
v
v
PCML
v
v
v
LVDS
v
v
v
HyperTransport technology
v
v
v
Differential HSTL
v
v
v
Differential SSTL
3.3-V GTL
1–12
Stratix Device Handbook, Volume 2
v
v
v
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
Table 1–6. I/O Standards Supported for Enhanced PLL Pins (Part 2 of 2)
Input
Output
I/O Standard
INCLK
FBIN
PLLENABLE
EXTCLK
3.3-V GTL+
v
v
v
1.5-V HSTL Class I
v
v
v
1.5-V HSTL Class II
v
v
v
1.8-V HSTL Class I
v
v
v
1.8-V HSTL Class II
v
v
v
SSTL-18 Class I
v
v
v
SSTL-18 Class II
v
v
v
SSTL-2 Class I
v
v
v
SSTL-2 Class II
v
v
v
SSTL-3 Class I
v
v
v
SSTL-3 Class II
v
v
v
AGP (1× and 2×)
v
v
v
CTT
v
v
v
Enhanced PLLs 11 and 12 support one single-ended output each (see
Figure 1–6). These outputs do not have their own VCC and GND signals.
Therefore, to minimize jitter, do not place switching I/O pins next to this
output pin.
Figure 1–6. External Clock Outputs for Enhanced PLLs 11 & 12
g0
Counter
CLK13n, I/O, PLL11_OUT
or CLK6n, I/O, PLL12_OUT (1)
From Internal
Logic or IOE
Note to Figure 1–6:
(1)
For PLL11, this pin is CLK13n; for PLL 12 this pin is CLK6n.
Altera Corporation
July 2005
1–13
Stratix Device Handbook, Volume 2
Enhanced PLLs
Stratix and Stratix GX devices can drive any enhanced PLL driven
through the global clock or regional clock network to any general I/O pin
as an external output clock. The jitter on the output clock is not
guaranteed for these cases.
Clock Feedback
The following three feedback modes in Stratix and Stratix GX device
enhanced PLLs allow multiplication and/or phase shifting:
■
■
■
■
Zero delay buffer: The external clock output pin is phase-aligned
with the clock input pin for zero delay. Altera recommends using the
same I/O standard on the input clock and the output clocks for
optimum performance.
External feedback: The external feedback input pin, FBIN, is phasealigned with the clock input, CLK, pin. Aligning these clocks allows
you to remove clock delay and skew between devices. This mode is
only possible for PLLs 5 and 6. PLLs 5 and 6 each support feedback
for one of the dedicated external outputs, either one single-ended or
one differential pair. In this mode, one encounter feeds back to the
PLL FBIN input, becoming part of the feedback loop.
Normal mode: If an internal clock is used in this mode, it is phasealigned to the input clock pin. The external clock output pin has a
phase delay relative to the clock input pin if connected in this mode.
No compensation: In this mode, the PLL does not compensate for
any clock networks or external clock outputs.
Table 1–7 shows which modes are supported by which PLL type.
Table 1–7. Clock Feedback Mode Availability
Mode Available in
Clock Feedback Mode
Enhanced PLLs
Fast PLLs
Yes
Yes
Normal Mode
Yes
Yes
Zero delay buffer mode
Yes
No
External feedback mode
Yes
No
No compensation mode
Phase Shifting
Stratix and Stratix GX device enhanced PLLs provide advanced
programmable phase shifting. You set these parameters in the Quartus II
software.
1–14
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
Phase Delay
The Quartus II software automatically sets the phase taps and counter
settings according to the phase shift entry. You enter a desired phase shift
and the Quartus II software automatically sets the closest setting
achievable. This type of phase shift is not reconfigurable during system
operation. For phase shifting, enter a phase shift (in degrees or time units)
for each PLL clock output port or for all outputs together in one shift.
You can select phase-shifting values in time units with a resolution of
156.25 to 416.66 ps. This resolution is a function of frequency input and
the multiplication and division factors (that is, it is a function of the VCO
period), with the finest step being equal to an eighth (× 0.125) of the VCO
period. Each clock output counter can choose a different phase of the
VCO period from up to eight taps for individual fine-step selection. Also,
each clock output counter can use a unique initial count setting to achieve
individual coarse-shift selection in steps of one VCO period. The
combination of coarse and fine shifts allows phase shifting for the entire
input clock period.
The equation to determine the precision of the phase shifting in degrees
is: 45° ÷ post-scale counter value. Therefore, the maximum step size is
45° , and smaller steps are possible depending on the multiplication and
division ratio necessary on the output counter port.
This type of phase shift provides the highest precision since it is the least
sensitive to process, supply, and temperature variation.
Lock Detect
The lock output indicates that there is a stable clock output signal in
phase with the reference clock. Without any additional circuitry, the lock
signal may toggle as the PLL begins tracking the reference clock. You may
need to gate the lock signal for use as a system control. The lock signal
from the locked port can drive the logic array or an output pin.
Whenever the PLL loses lock for any reason (be it excessive inclk jitter,
clock switchover, PLL reconfiguration, power supply noise, etc.), the PLL
must be reset with the areset signal to guarantee correct phase
relationship between the PLL output clocks. If the phase relationship
between the input clock versus output clock, and between different
output clocks from the PLL is not important in your design, the PLL need
not be reset.
f
Altera Corporation
July 2005
See the Stratix FPGA Errata Sheet for more information on implementing
the gated lock signal in your design.
1–15
Stratix Device Handbook, Volume 2
Enhanced PLLs
Programmable Duty Cycle
The programmable duty cycle allows enhanced PLLs to generate clock
outputs with a variable duty cycle. This feature is supported on each
enhanced PLL post-scale counter (g0..g3, l0..l3, e0..e3). The duty cycle
setting is achieved by a low and high time count setting for the post-scale
counters. The Quartus II software uses the frequency input and the
required multiply or divide rate to determine the duty cycle choices. The
precision of the duty cycle is determined by the post-scale counter value
chosen on an output. The precision is defined by 50% divided by the postscale counter value. The closest value to 100% is not achievable for a given
counter value. For example, if the g0 counter is 10, then steps of 5% are
possible for duty cycle choices between 5 to 90%.
If the device uses external feedback, you must set the duty cycle for the
counter driving off the device to 50%.
General Advanced Clear & Enable Control
There are several control signals for clearing and enabling PLLs and PLL
outputs. You can use these signals to control PLL resynchronization and
gate PLL output clocks for low-power applications.
The pllenable pin is a dedicated pin that enables/disables PLLs. When
the pllenable pin is low, the clock output ports are driven by GND and
all the PLLs go out of lock. When the pllenable pin goes high again, the
PLLs relock and resynchronize to the input clocks. You can choose which
PLLs are controlled by the pllenable signal by connecting the
pllenable input port of the altpll megafunction to the common
pllenable input pin.
The areset signals are reset/resynchronization inputs for each PLL. The
areset signal should be asserted every time the PLL loses lock to
guarantee correct phase relationship between the PLL output clocks.
Users should include the areset signal in designs if any of the following
conditions are true:
■
■
PLL reconfiguration or clock switchover enables in the design
Phase relationships between output clocks need to be maintained
after a loss of lock condition
The device input pins or logic elements (LEs) can drive these input
signals. When driven high, the PLL counters reset, clearing the PLL
output and placing the PLL out of lock. The VCO sets back to its nominal
setting (~700 MHz). When driven low again, the PLL resynchronizes to
its input as it relocks. If the target VCO frequency is below this nominal
1–16
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
frequency, then the output frequency starts at a higher value than desired
as the PLL locks. If the system cannot tolerate this, the clkena signal can
disable the output clocks until the PLL locks.
The pfdena signals control the phase frequency detector (PFD) output
with a programmable gate. If you disable the PFD, the VCO operates at
its last set value of control voltage and frequency with some long-term
drift to a lower frequency. The system continues running when the PLL
goes out of lock or the input clock is disabled. By maintaining the last
locked frequency, the system has time to store its current settings before
shutting down. You can either use your own control signal or a clkloss
status signal to trigger pdfena.
The clkena signals control the enhanced PLL regional and global
outputs. Each regional and global output port has its own clkena signal.
The clkena signals synchronously disable or enable the clock at the PLL
output port by gating the outputs of the g and l counters. The clkena
signals are registered on the falling edge of the counter output clock to
enable or disable the clock without glitches.
Figure 1–7 shows the waveform example for a PLL clock port enable. The
PLL can remain locked independent of the clkena signals since the looprelated counters are not affected. This feature is useful for applications
that require a low power or sleep mode. Upon re-enabling, the PLL does
not need a resynchronization or relock period. The clkena signal can
also disable clock outputs if the system is not tolerant to frequency
overshoot during resynchronization.
The extclkena signals work in the same way as the clkena signals, but
they control the external clock output counters (e0, e1, e2, and e3). Upon
re-enabling, the PLL does not need a resynchronization or relock period
unless the PLL is using external feedback mode. In order to lock in
external feedback mode, the external output must drive the board trace
back to the FBIN pin.
Altera Corporation
July 2005
1–17
Stratix Device Handbook, Volume 2
Enhanced PLLs
Figure 1–7. extclkena Signals
COUNTER
OUTPUT
CLKENA
CLKOUT
Programmable Bandwidth
Enhanced PLLs provide advanced control of the PLL bandwidth using
the programmable characteristics of the PLL loop, including loop filter
and charge pump.
Background
The PLL bandwidth is the measure of the PLLs ability to track the input
clock and jitter. It is determined by the −3-dB frequency of the closed-loop
gain in the PLL or approximately the unity gain point for open loop PLL
response. As Figure 1–8 shows, these points correspond to approximately
the same frequency.
1–18
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
Figure 1–8. Open- & Closed-Loop Response Bode Plots
Open-Loop Reponse Bode Plot
Increasing the PLL's
bandwidth in effect pushes
the open loop response out.
0 dB
Gain
Frequency
Closed-Loop Reponse Bode Plot
Gain
Frequency
A high-bandwidth PLL provides a fast lock time and tracks jitter on the
reference clock source, passing it through to the PLL output. A lowbandwidth PLL filters out reference clock jitter, but increases lock time.
Stratix device enhanced PLLs allow you to control the bandwidth over a
Altera Corporation
July 2005
1–19
Stratix Device Handbook, Volume 2
Enhanced PLLs
finite range to customize the PLL characteristics for a particular
application. Applications that require clock switchover (such as TDMA,
frequency hopping wireless, and redundant clocking) can benefit from
the programmable bandwidth feature of the Stratix and Stratix GX PLLs.
The bandwidth and stability of such a system is determined by a number
of factors including the charge pump current, the loop filter resistor
value, the high-frequency capacitor value (in the loop filter), and the mcounter value. You can use the Quartus II software to control these factors
and to set the bandwidth to the desired value within a given range.
You can set the bandwidth to the appropriate value to balance the need
for jitter filtering and lock time. Figures 1–9 and 1–10 show the output of
a low- and high-bandwidth PLL, respectively, as it locks onto the input
clock.
Figure 1–9. Low-Bandwidth PLL Lock Time
160
155
Lock Time = 8 μs
150
145
Frequency (MHz)
140
135
130
125
120
0
5
10
15
Time (μs)
1–20
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
Figure 1–10. High-Bandwidth PLL Lock Time
160
155
Lock Time = 4 μs
150
145
Frequency (MHz)
140
135
130
125
120
0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
Time (μs)
A high-bandwidth PLL may benefit a system with two cascaded PLLs. If
the first PLL uses spread spectrum (as user-induced jitter), the second
PLL needs a high bandwidth so it can track the jitter that is feeding it. A
low-bandwidth PLL may, in this case, lose lock due to the spread
spectrum-induced jitter on the input clock.
A low-bandwidth PLL may benefit a system using clock switchover.
When the clock switchover happens, the PLL input temporarily stops. A
low-bandwidth PLL would react more slowly to changes to its input
clock and take longer to drift to a lower frequency (caused by the input
stopping) than a high-bandwidth PLL. Figures 1–11 and 1–12
demonstrate this property.
The two plots show the effects of clock switchover with a low- or highbandwidth PLL. When the clock switchover happens, the output of the
low-bandwidth PLL (see Figure 1–11) drifts to lower frequency much
slower than the high-bandwidth PLL output (see Figures 1–12).
Altera Corporation
July 2005
1–21
Stratix Device Handbook, Volume 2
Enhanced PLLs
Figure 1–11. Effect of Low Bandwidth on Clock Switchover
164
162
160
158
Frequency (MHz)
Input Clock Stops
Re-lock
156
Initial Lock
154
152
Switchover
150
0
5
10
15
20
25
30
35
40
Time (μs)
1–22
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
Figure 1–12. Effect of High Bandwidth on Clock Switchover
160
Input Clock Stops
Re-lock
155
Initial Lock
150
145
Frequency (MHz)
140
135
Switchover
130
125
0
2
4
6
8
10
12
14
16
18
20
Time (μs)
Implementation
Traditionally, external components such as the VCO or loop filter control
a PLL’s bandwidth. Most loop filters are made up of passive components,
such as resistors and capacitors, which take up unnecessary board space
and increase cost. With Stratix and Stratix GX device enhanced PLLs, all
the components are contained within the device to increase performance
and decrease cost.
Stratix and Stratix GX device enhanced PLLs implement programmable
bandwidth by giving you control of the charge pump current and loop
filter resistor (R) and high-frequency capacitor (Ch) values (see
Table 1–8). The Stratix and Stratix GX device enhanced PLL bandwidth
ranges from approximately 150 kHz to 2 MHz.
Altera Corporation
July 2005
1–23
Stratix Device Handbook, Volume 2
Enhanced PLLs
The charge pump current directly affects the PLL bandwidth. The higher
the charge pump current, the higher the PLL bandwidth. You can choose
from a fixed set of values for the charge pump current. Figure 1–13 shows
the loop filter and the components that you can set via the Quartus II
software.
Figure 1–13. Loop Filter Programmable Components
IUP
PFD
R
Ch
IDN
C
Software Support
The Quartus II software provides two levels of programmable bandwidth
control. The first level allows you to enter a value for the desired
bandwidth directly into the Quartus II software using the MegaWizard®
Plug-In Manager. Alternatively, you can set the bandwidth parameter in
the altpll function to the desired bandwidth. The Quartus II software
then chooses each individual bandwidth parameter to achieve the
desired setting. If designs cannot achieve the desired bandwidth setting,
the Quartus II software selects the closest achievable value. For preset
low, medium, and high bandwidth settings, the Quartus II software sets
the bandwidth as follows:
■
■
■
Low bandwidth is set at 150 KHz
Medium bandwidth is set at 800 KHz
High bandwidth is set at 2 Mhz
If you choose Auto bandwidth, the Quartus II software chooses the PLL
settings and you can get a bandwidth setting outside the 150-Khz to
2-Mhz range.
1–24
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
An advanced level of control is also possible for precise control of the
loop filter parameters. This level allows you to specifically select the
charge pump current, loop filter resistor value, and loop filter (high
frequency) capacitor value. These parameters are: charge_pump_current,
loop_filter_r, and loop_filter_c. Each parameter supports the specific
range of values listed in Table 1–8.
Table 1–8. Advanced Loop Filter Parameters
Parameter
f
Values
Resistor values (kΩ)
1, 2, 3, 4, 7, 8, 9, 10
High-frequency capacitance
values (pF)
5, 10, 15, 20
Charge pump current settings
(μA)
10, 15, 20, 24, 30, 35, 40, 45, 50, 55, 60, 65,
70, 75, 80, 85, 90, 100, 112, 135, 148, 164,
212
For more information on PLL software support in the Quartus II
software, see the altpll Megafunction User Guide.
Clock Switchover
f
For more information on implementing clock switchover, see
AN 313: Implementing Clock Switchover in Stratix & Stratix GX Devices.
Spread-Spectrum Clocking
Digital clocks are generally square waves with short rise times and a 50%
duty cycle. These high-speed digital clocks concentrate a significant
amount of energy in a narrow bandwidth at the target frequency and at
the higher frequency harmonics. This results in high energy peaks and
increased electromagnetic interference (EMI). The radiated noise from
the energy peaks travels in free air and, if not minimized, can lead to
corrupted data and intermittent system errors, which can jeopardize
system reliability.
Background
Traditional methods for limiting EMI include shielding, filtering, and
multi-layer printed circuit boards (PCBs). However, these methods
significantly increase the overall system cost and sometimes are not
enough to meet EMI compliance. Spread-spectrum technology provides
a simple and effective technique for reducing EMI emissions without
additional cost and the trouble of re-designing a board.
Altera Corporation
July 2005
1–25
Stratix Device Handbook, Volume 2
Enhanced PLLs
Spread-spectrum technology modulates the target frequency over a small
range. For example, if a 100-MHz signal has a 0.5% down-spread
modulation, then the frequency is swept from 99.5 to 100 MHz.
Figure 1–14 gives a graphical representation of the energy present in a
spread-spectrum signal as opposed to a non-spread-spectrum signal. It is
apparent that instead of concentrating the energy at the target frequency,
the energy is re-distributed across a wider band of frequencies, which
reduces peak energy.
Not only is there a reduction in the fundamental peak EMI components,
but there is also a reduction in EMI of the higher order harmonics. Since
some regulations focus on peak EMI emissions, rather than average EMI
emissions, spread-spectrum technology is a valuable method of EMI
reduction.
Figure 1–14. Spread-Spectrum Signal Energy versus Non-Spread-Spectrum Signal Energy
Spread-Spectrum Signal
Non-Spread-Spectrum Signal
Δ = ~5 dB
Amplitude
(dB)
δ = 0.5%
Frequency
(MHz)
Spread-spectrum technology would benefit a design with high EMI
emissions and/or strict EMI requirements. Device-generated EMI is
dependent on frequency, output voltage swing amplitude, and slew rate.
For example, a design using LVDS already has low EMI emissions
1–26
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
because of the low-voltage swing. The differential LVDS signal also
allows for EMI rejection within the signal. Therefore, this situation may
not require spread-spectrum technology.
Description
Stratix and Stratix GX device enhanced PLLs feature spread-spectrum
technology to reduce the EMI emitted from the device. The enhanced PLL
provides up to a 0.5% down spread (–0.5%) using a triangular, also
known as linear, modulation profile. The modulation frequency is
programmable and ranges from approximately 30 to 150 kHz. The spread
percentage is based on the clock input to the PLL and the m and n settings.
Spread-spectrum technology reduces the peak energy by 2 to 5 dB at the
target frequency. However, this number is dependent on bandwidth and
the m and n counter values and can vary from design to design.
Spread percentage, also known as modulation width, is defined as the
percentage that the design modulates the target frequency. A negative
(–) percentage indicates a down spread, a positive (+) percentage
indicates an up spread, and a (±) indicates a center spread. Modulation
frequency is the frequency of the spreading signal or how fast the signal
sweeps from the minimum to the maximum frequency. Down-spread
modulation shifts the target frequency down by half the spread
percentage, centering the modulated waveforms on a new target
frequency.
The m and n counter values are toggled at the same time between two
fixed values. The loop filter then slowly changes the VCO frequency to
provide the spreading effect, which results in a triangular modulation.
An additional spread-spectrum counter (shown in Figure 1–15) sets the
modulation frequency. Figure 1–15 shows how spread-spectrum
technology is implemented in the Stratix device enhanced PLL.
Altera Corporation
July 2005
1–27
Stratix Device Handbook, Volume 2
Enhanced PLLs
Figure 1–15. Spread-Spectrum Circuit Block Diagram
÷n
refclk
Up
PFD
Down
Spread
Spectrum
Counter
÷m
n count1
n count2
m count2
m count1
Figure 1–16 shows a VCO frequency waveform when toggling between
different counter values. Since the enhanced PLL switches between two
different m and n values, the result is a straight line between two
frequencies, which gives a linear modulation. The magnitude of
modulation is determined by the ratio of two m/n sets. The percent
spread is determined by:
percent spread = (fVCOmax −fVCOmin)/fVCOmax = 1 −[(m2 × n1)/(m1 × n2)]
The maximum and minimum VCO frequency is defined as:
fVCOmax = (m1/n1) × fref
fVCOmin = (m2/n2) × fref
1–28
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
Figure 1–16. VCO Frequency Modulation Waveforms
count2 values
count1 values
VCO Frequency
Software Support
You can enter the desired down-spread percentage and modulation
frequency in the MegaWizard Plug-In Manager through the Quartus II
software. Alternatively, the MegaWizard Plug-In Manager can set the
downspread parameter in the altpll megafunction to the desired
down-spread percentage. Timing analysis ensures the design operates at
the maximum spread frequency and meets all timing requirements.
f
For more information on PLL software support in the Quartus II
software, see the altpll Megafunction User Guide.
Guidelines
If the design cascades PLLs, the source, or upstream PLL should have a
low bandwidth setting, while the destination, or downstream PLL should
have a high bandwidth setting. The upstream PLL must have a low
bandwidth setting because a PLL does not generate jitter higher than its
bandwidth. The downstream PLL must have a high bandwidth setting to
track the jitter. The design must use the spread-spectrum feature in a lowbandwidth PLL and, therefore, the Quartus II software automatically sets
the spread-spectrum PLL’s bandwidth to low.
1
Designs cannot use spread-spectrum PLLs with the
programmable bandwidth feature.
Stratix and Stratix GX devices can accept a spread-spectrum input with
typical modulation frequencies. However, the device cannot
automatically detect that the input is a spread-spectrum signal. Instead,
the input signal looks like deterministic jitter at the input of the
downstream PLL.
Spread spectrum should only have a minor effect on period jitter, but
period jitter increases. Period jitter is the deviation of a clock’s cycle time
from its previous cycle position. Period jitter measures the variation of a
clock’s output transition from its ideal position over consecutive edges.
Altera Corporation
July 2005
1–29
Stratix Device Handbook, Volume 2
Enhanced PLLs
With down-spread modulation, the peak of the modulated waveform is
the actual target frequency. Therefore, the system never exceeds the
maximum clock speed. To maintain reliable communication, the entire
system/subsystem should use the Stratix or Stratix GX device as the clock
source. Communication could fail if the Stratix or Stratix GX logic array
is clocked by the spread-spectrum clock, but the data it receives from
another device is not.
Since spread spectrum affects the m counter values, all spread-spectrum
PLL outputs are affected. Therefore, if only one spread-spectrum signal is
needed, the clock signal should use a separate PLL without other outputs
from that PLL.
No special considerations are needed when using spread spectrum with
the clock switchover feature. This is because the clock switchover feature
does not affect the m and n counter values, which are the counter values
that are switching when using spread spectrum.
PLL Reconfiguration
f
See AN 282: Implementing PLL Reconfiguration in Stratix & Stratix GX
Devices for information on PLL reconfiguration.
Enhanced PLL Pins
Table 1–9 shows the physical pins and their purpose for the Enhanced
PLLs. For inclk port connections to pins see “Clocking” on page 1–39.
Table 1–9. Enhanced PLL Pins (Part 1 of 2)
Pin
Description
CLK4p/n
Single-ended or differential pins that can drive the inclk port for PLL 6.
CLK5p/n
Single-ended or differential pins that can drive the inclk port for PLL 6.
CLK6p/n
Single-ended or differential pins that can drive the inclk port for PLL 12.
CLK7p/n
Single-ended or differential pins that can drive the inclk port for PLL 12.
CLK12p/n
Single-ended or differential pins that can drive the inclk port for PLL 11.
CLK13p/n
Single-ended or differential pins that can drive the inclk port for PLL 11.
CLK14p/n
Single-ended or differential pins that can drive the inclk port for PLL 5.
CLK15p/n
Single-ended or differential pins that can drive the inclk port for PLL 5.
PLL5_FBp/n
Single-ended or differential pins that can drive the fbin port for PLL 5.
PLL6_FBp/n
Single-ended or differential pins that can drive the fbin port for PLL 6.
PLLENABLE
Dedicated input pin that drives the pllena port of all or a set of PLLs. If you do not
use this pin, connect it to ground.
1–30
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
Table 1–9. Enhanced PLL Pins (Part 2 of 2)
Pin
Description
PLL5_OUT[3..0]p/n Single-ended or differential pins driven by extclk[3..0] ports from PLL 5.
PLL6_OUT[3..0]p/n Single-ended or differential pins driven by extclk[3..0] ports from PLL 6.
PLL11_OUT, CLK13n Single-ended output pin driven by clk0 port from PLL 11.
PLL12_OUT, CLK6n
Single-ended output pin driven by clk0 port from PLL 12.
VCCA_PLL5
Analog power for PLL 5. Connect this pin to 1.5 V, even if the PLL is not used.
VCCG_PLL5
Guard ring power for PLL 5. Connect this pin to 1.5 V, even if the PLL is not used.
GNDA_PLL5
Analog ground for PLL 5. You can connect this pin to the GND plane on the board.
GNDG_PLL5
Guard ring ground for PLL 5. You can connect this pin to the GND plane on the board.
VCCA_PLL6
Analog power for PLL 6. Connect this pin to 1.5 V, even if the PLL is not used.
VCCG_PLL6
Guard ring power for PLL 6. Connect this pin to 1.5 V, even if the PLL is not used.
GNDA_PLL6
Analog ground for PLL 6. You can connect this pin to the GND plane on the board.
GNDG_PLL6
Guard ring ground for PLL 6. You can connect this pin to the GND plane on the board.
VCCA_PLL11
Analog power for PLL 11. Connect this pin to 1.5 V, even if the PLL is not used.
VCCG_PLL11
Guard ring power for PLL 11. Connect this pin to 1.5 V, even if the PLL is not used.
GNDA_PLL11
Analog ground for PLL 11. You can connect this pin to the GND plane on the board.
GNDG_PLL11
Guard ring ground for PLL 11.You can connect this pin to the GND plane on the board.
VCCA_PLL12
Analog power for PLL 12. Connect this pin to 1.5 V, even if the PLL is not used.
VCCG_PLL12
Guard ring power for PLL 12. Connect this pin to 1.5 V, even if the PLL is not used.
GNDA_PLL12
Analog ground for PLL 12. You can connect this pin to the GND plane on the board.
GNDG_PLL12
Guard ring ground for PLL 12. You can connect this pin to the GND plane on the board.
VCC_PLL5_OUTA
External clock output VCCIO power for PLL5_OUT0p, PLL5_OUT0n, PLL5_OUT1p,
and PLL5_OUT1n outputs from PLL 5.
VCC_PLL5_OUTB
External clock output VCCIO power for PLL5_OUT2p, PLL5_OUT2n, PLL5_OUT3p,
and PLL5_OUT3n outputs from PLL 5.
VCC_PLL6_OUTA
External clock output VCCIO power for PLL5_OUT0p, PLL5_OUT0n, PLL5_OUT1p,
and PLL5_OUT1n outputs from PLL 6.
VCC_PLL6_OUTB
External clock output VCCIO power for PLL5_OUT2p, PLL5_OUT2n, PLL5_OUT3p,
and PLL5_OUT3n outputs from PLL 6.
Fast PLLs
Altera Corporation
July 2005
Stratix devices contain up to eight fast PLLs and Stratix GX devices
contain up to four fast PLLs. Both device PLLs have high-speed
differential I/O interface ability along with general-purpose features.
Figure 1–17 shows a diagram of the fast PLL. This section discusses the
1–31
Stratix Device Handbook, Volume 2
Fast PLLs
general purpose abilities of the Fast PLL. For information on the highspeed differential I/O interface capabilities, see the High-Speed Differential
I/O Interfaces in Stratix Devices chapter.
Figure 1–17. Stratix & Stratix GX Fast PLL Block Diagram
Post-Scale
Counters
diffioclk1 (2)
÷l0
VCO Phase Selection
Selectable at each PLL
Output Port
Global or
regional clock (1)
Clock
Input
Phase
Frequency
Detector
Global or
regional clock
txload_en (3)
rxload_en (3)
÷l1
Global or
regional clock
diffioclk2 (2)
PFD
Charge
Pump
8
Loop
Filter
VCO
÷g0
Global or
regional clock
÷m
Notes to Figure 1–17:
(1)
(2)
(3)
The global or regional clock input can be driven by an output from another PLL or any dedicated CLK or FCLK pin.
It cannot be driven by internally-generated global signals.
In high-speed differential I/O support mode, this high-speed PLL clock feeds the SERDES. Stratix and Stratix GX
devices only support one rate of data transfer per fast PLL in high-speed differential I/O support mode.
This signal is a high-speed differential I/O support SERDES control signal.
1–32
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
Figure 1–18 shows all possible ports related to fast PLLs.
Figure 1–18. Fast PLL Ports & Physical Destinations
Fast PLL Signals
(1)
pllena
(2)
inclk0
clk[2..0]
locked
areset
pfdena
Physical Pin
Signal Driven by Internal Logic
Signal Driven to Internal Logic
Internal Clock Signal
Notes to Figure 1–18:
(1)
(2)
This input pin is shared by all enhanced and fast PLLs.
This input pin is either single-ended or differential.
Tables 1–10 and 1–11 show the description of all fast PLL ports.
Table 1–10. Fast PLL Input Signals
Name
Description
Source
Destination
inclk1
Reference clock input to PLL
Pin
PFD
pllena
Enable pin for enabling or disabling all or a set of
PLLs – active high
Pin
PLL control signal
areset
Signal used to reset the PLL which resynchronizes all the counter outputs⎯active high
Logic array
PLL control signal
pfdena
Enables the up/down outputs from the phasefrequency detector⎯active high
Logic array
PFD
Table 1–11. Fast PLL Output Signals
Name
Description
Source
Destination
clk[2..0]
PLL outputs driving regional or global clock
PLL counter
Internal clock
locked
Lock output from lock detect circuit⎯active high
PLL lock
detect
Logic array
Altera Corporation
July 2005
1–33
Stratix Device Handbook, Volume 2
Fast PLLs
Clock Multiplication & Division
Stratix and Stratix GX device fast PLLs provide clock synthesis for PLL
output ports using m/(post scaler) scaling factors. The input clock is
multiplied by the m feedback factor. Each output port has a unique post
scale counter to divide down the high-frequency VCO. There is one
multiply counter, m, per fast PLL with a range of 1 to 32. There are three
post-scale counters (g0, l0, and l1) for the regional and global clock output
ports. All post-scale counters range from 1 to 32. If the design uses a
high-speed serial interface, you can set the output counter to 1 to allow
the high-speed VCO frequency to drive the SERDES.
External Clock Outputs
Each fast PLL supports differential or single-ended outputs for sourcesynchronous transmitters or for general-purpose external clocks. There
are no dedicated external clock output pins. The fast PLL global or
regional outputs can drive any I/O pin as an external clock output pin.
The I/O standards supported by any particular bank determines what
standards are possible for an external clock output driven by the fast PLL
in that bank. See the Selectable I/O Standards in Stratix & Stratix GX Devices
chapter in the Stratix Device Handbook, Volume 2 or the Stratix GX Device
Handbook, Volume 2 for output standard support.
Table 1–12 shows the I/O standards supported by fast PLL input pins.
Table 1–12. Fast PLL Port I/O Standards (Part 1 of 2)
Input
I/O Standard
INCLK
PLLENABLE
LVTTL
v
v
LVCMOS
v
v
2.5 V
v
1.8 V
v
1.5 V
v
3.3-V PCI
3.3-V PCI-X 1.0
LVPECL
v
PCML
v
LVDS
v
HyperTransport technology
v
Differential HSTL
v
1–34
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
Table 1–12. Fast PLL Port I/O Standards (Part 2 of 2)
Input
I/O Standard
INCLK
PLLENABLE
Differential SSTL
3.3-V GTL
3.3-V GTL+
v
1.5-V HSTL Class I
v
1.5-V HSTL Class II
1.8-V HSTL Class I
v
1.8-V HSTL Class II
SSTL-18 Class I
v
SSTL-18 Class II
SSTL-2 Class I
v
SSTL-2 Class II
v
SSTL-3 Class I
v
SSTL-3 Class II
v
AGP (1× and 2×)
CTT
v
Phase Shifting
Stratix and Stratix GX device fast PLLs have advanced clock shift ability
to provide programmable phase shift. These parameters are set in the
Quartus II software.
The Quartus II software automatically sets the phase taps and counter
settings according to the phase shift entry. Enter a desired phase shift and
the Quartus II software automatically sets the closest setting achievable.
This type of phase shift is not reconfigurable during system operation.
You can enter a phase shift (in degrees or time units) for each PLL clock
output port or for all outputs together in one shift. You can perform phase
shifting in time units with a resolution range of 125 to 416.66 ps to create
a function of frequency input and the multiplication and division factors
(that is, it is a function of the VCO period), with the finest step being equal
to an eighth (× 0.125) of the VCO period. Each clock output counter can
choose a different phase of the VCO period from up to eight taps for
individual fine-step selection. Also, each clock output counter can use a
unique initial count setting to achieve individual coarse shift selection in
steps of one VCO period. The combination of coarse and grain shifts
allows phase shifting for the entire input clock period.
Altera Corporation
July 2005
1–35
Stratix Device Handbook, Volume 2
Fast PLLs
The equation to determine the precision of phase in degrees is: 45° ÷ postscale counter value. Therefore, the maximum step size is 45° , and smaller
steps are possible depending on the multiplication and division ratio
necessary on the output counter port.
This type of phase shift provides the highest precision since it is the least
sensitive to process, supply, and temperature variation.
Programmable Duty Cycle
The programmable duty cycle allows the fast PLL to generate clock
outputs with a variable duty cycle. This feature is supported on each fast
PLL post-scale counter. g0, l0, and l1 all support programmable duty. You
use a low- and high-time count setting for the post-scale counters to set
the duty cycle.
The Quartus II software uses the frequency input and multiply/divide
rate desired to select the post-scale counter, which determines the
possible choices for each duty cycle. The precision of the duty cycle is
determined by the post-scale counter value chosen on an output. The
precision is defined by 50% divided by the post-scale counter value. The
closest value to 100% is not achievable for a given counter value. For
example, if the g0 counter is 10, then steps of 5% are possible for duty
cycle choices between 5 to 90%.
If the device uses external feedback, you must set the duty cycle for the
counter driving off the device to 50%.
Control Signals
The lock output indicates a stable clock output signal in phase with the
reference clock. Unlike enhanced PLLs, fast PLLs do not have a lock filter
counter.
The pllenable pin is a dedicated pin that enables/disables both PLLs.
When the pllenable pin is low, the clock output ports are driven by
GND and all the PLLs go out of lock. When the pllenable pin goes high
again, the PLLs relock and resynchronize to the input clocks. You can
choose which PLLs are controlled by the pllenable by connecting the
pllenable input port of the altpll megafunction to the common
pllenable input pin.
The areset signals are reset/resynchronization inputs for each fast PLL.
The Stratix and Stratix GX devices can drive these input signals from an
input pin or from LEs. When driven high, the PLL counters reset, clearing
the PLL output and placing the PLL out of lock. The VCO sets back to its
nominal setting (~700 MHz). When driven low again, the PLL
1–36
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
resynchronizes to its input clock as it relocks. If the target VCO frequency
is below this nominal frequency, then the output frequency starts at a
higher value then desired as it locks.
The pfdena signals control the PFD output with a programmable gate. If
you disable the PFD, the VCO operates at its last set value of control
voltage and frequency with some long-term drift to a lower frequency.
The system continues running when the PLL goes out of lock or the input
clock disables. By maintaining the last locked frequency, the system has
time to store its current settings before shutting down.
If the PLL loses lock for any reason (for example, because of excessive
inclk jitter, clock switchover, PLL reconfiguration, or power supply
noise), the PLL must be reset with the areset signal to guarantee correct
phase relationship between the PLL output clocks. If the phase
relationship between the input clock and the output clock and between
different output clocks from the PLL is not important in your design, it is
not necessary to reset the PLL.
Pins
Table 1–13 shows the physical pins and their purpose for the Fast PLLs.
For inclk port connections to pins see “Clocking” on page 1–39.
Table 1–13. Fast PLL Pins (Part 1 of 3)
Pin
Description
CLK0p/n
Single-ended or differential pins that can drive the inclk port for PLL 1 or 7.
CLK1p/n
Single-ended or differential pins that can drive the inclk port for PLL 1.
CLK2p/n
Single-ended or differential pins that can drive the inclk port for PLL 2 or 8.
CLK3p/n
Single-ended or differential pins that can drive the inclk port for PLL 2.
CLK8p/n
Single-ended or differential pins that can drive the inclk port for PLL 3 or 9. (1)
CLK9p/n
Single-ended or differential pins that can drive the inclk port for PLL 3. (1)
CLK10p/n
Single-ended or differential pins that can drive the inclk port for PLL 4 or 10. (1)
CLK11p/n
Single-ended or differential pins that can drive the inclk port for PLL 4. (1)
FPLL7CLKp/n
Single-ended or differential pins that can drive the inclk port for PLL 7.
FPLL8CLKp/n
Single-ended or differential pins that can drive the inclk port for PLL 8.
FPLL9CLKp/n
Single-ended or differential pins that can drive the inclk port for PLL 9. (1)
FPLL10CLKp/n
Single-ended or differential pins that can drive the inclk port for PLL 10. (1)
PLLENABLE
Dedicated input pin that drives the pllena port of all or a set of PLLs. If you do not
use this pin, connect it to ground.
VCCA_PLL1
Analog power for PLL 1. Connect this pin to 1.5 V, even if the PLL is not used.
Altera Corporation
July 2005
1–37
Stratix Device Handbook, Volume 2
Fast PLLs
Table 1–13. Fast PLL Pins (Part 2 of 3)
Pin
Description
VCCG_PLL1
Guard ring power for PLL 1. Connect this pin to 1.5 V, even if the PLL is not used.
GNDA_PLL1
Analog ground for PLL 1. You can connect this pin to the GND plane on the board.
GNDG_PLL1
Guard ring ground for PLL 1. You can connect this pin to the GND plane on the
board.
VCCA_PLL2
Analog power for PLL 2. Connect this pin to 1.5 V, even if the PLL is not used.
VCCG_PLL2
Guard ring power for PLL 2. Connect this pin to1.5 V, even if the PLL is not used.
GNDA_PLL2
Analog ground for PLL 2. You can connect this pin to the GND plane on the board.
GNDG_PLL2
Guard ring ground for PLL 2. You can connect this pin to the GND plane on the
board.
VCCA_PLL3
Analog power for PLL 3. Connect this pin to 1.5 V, even if the PLL is not used. (1)
VCCG_PLL3
Guard ring power for PLL 3. Connect this pin to 1.5 V, even if the PLL is not
used. (1)
GNDA_PLL3
Analog ground for PLL 3. You can connect this pin to the GND plane on the
board. (1)
GNDG_PLL3
Guard ring ground for PLL 3. You can connect this pin to the GND plane on the
board. (1)
VCCA_PLL4
Analog power for PLL 4. Connect this pin to 1.5 V, even if the PLL is not used. (1)
VCCG_PLL4
Guard ring power for PLL 4. Connect this pin to 1.5 V, even if the PLL is not
used. (1)
GNDA_PLL4
Analog ground for PLL 4. You can connect this pin to the GND plane on the
board. (1)
GNDG_PLL4
Guard ring ground for PLL 4. You can connect this pin to the GND plane on the
board. (1)
VCCA_PLL7
Analog power for PLL 7. Connect this pin to 1.5 V, even if the PLL is not used.
VCCG_PLL7
Guard ring power for PLL 7. Connect this pin to 1.5 V, even if the PLL is not used.
GNDA_PLL7
Analog ground for PLL 7. You can connect this pin to the GND plane on the board.
GNDG_PLL7
Guard ring ground for PLL 7. You can connect this pin to the GND plane on the
board.
VCCA_PLL8
Analog power for PLL 8. Connect this pin to 1.5 V, even if the PLL is not used.
VCCG_PLL8
Guard ring power for PLL 8. Connect this pin to 1.5 V, even if the PLL is not used.
GNDA_PLL8
Analog ground for PLL 8. You can connect this pin to the GND plane on the board.
GNDG_PLL8
Guard ring ground for PLL 8. You can connect this pin to the GND plane on the
board.
VCCA_PLL9
Analog power for PLL 9. Connect this pin to 1.5 V, even if the PLL is not used. (1)
VCCG_PLL9
Guard ring power for PLL 9. Connect this pin to 1.5 V, even if the PLL is not
used. (1)
1–38
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
Table 1–13. Fast PLL Pins (Part 3 of 3)
Pin
Description
GNDA_PLL9
Analog ground for PLL 9. You can connect this pin to the GND plane on the
board. (1)
GNDG_PLL9
Guard ring ground for PLL 9. You can connect this pin to the GND plane on the
board. (1)
VCCA_PLL10
Analog power for PLL 10. Connect this pin to 1.5 V, even if the PLL is not
used. (1)
VCCG_PLL10
Guard ring power for PLL 10. Connect this pin to 1.5 V, even if the PLL is not
used. (1)
GNDA_PLL10
Analog ground for PLL 10. Connect this pin to the GND plane on the board. (1)
GNDG_PLL10
Guard ring ground for PLL 10. You can connect this pin to the GND plane on the
board. (1)
Note to Table 1–13:
(1)
PLLs 3, 4, 9, and 10 are not available on Stratix GX devices for general-purpose configuration. These PLLs are part
of the HSSI block. See AN 236: Using Source-Synchronous Signaling with DPA in Stratix GX Devices for more
information.
Clocking
Stratix and Stratix GX devices provide a hierarchical clock structure and
multiple PLLs with advanced features. The large number of clocking
resources in combination with the clock synthesis precision provided by
enhanced and fast PLLs provides a complete clock management solution.
Global & Hierarchical Clocking
Stratix and Stratix GX devices provide 16 dedicated global clock
networks, 16 regional clock networks (4 per device quadrant), and
8 dedicated fast regional clock networks. These clocks are organized into
a hierarchical clock structure that allows for up to 22 clocks per device
region with low skew and delay. This hierarchical clocking scheme
provides up to 48 unique clock domains within Stratix and Stratix GX
devices.
There are 16 dedicated clock pins (CLK[15..0]) on Stratix devices and
12 dedicated clock pins (CLK[11..0]) on Stratix GX devices to drive
either the global or regional clock networks. Four clock pins drive each
side of the Stratix device, as shown in Figures 1–19 and 1–20. On Stratix
GX devices, four clock pins drive the top, left, and bottom sides of the
device. The clocks on the right side of the device are not available for
general-purpose PLLs. Enhanced and fast PLL outputs can also drive the
global and regional clock networks.
Altera Corporation
July 2005
1–39
Stratix Device Handbook, Volume 2
Clocking
Global Clock Network
These clocks drive throughout the entire device, feeding all device
quadrants. All resources within the device—IOEs, LEs, DSP blocks, and
all memory blocks—can use the global clock networks as clock sources.
These resources can also be used for control signals, such as clock enables
and synchronous or asynchronous clears fed from the external pin.
Internal logic can also drive the global clock networks for internally
generated global clocks and asynchronous clears, clock enables, or other
control signals with large fanout. Figure 1–19 shows the 16 dedicated CLK
pins driving global clock networks.
Figure 1–19. Global Clocking
CLK[15..12]
Global Clock [15..0]
CLK[3..0]
Global Clock [15..0]
CLK[11..8]
CLK[7..4]
Regional Clock Network
There are four regional clock networks within each quadrant of the
Stratix or Stratix GX device that are driven by the same dedicated
CLK[15..0] input pins or from PLL outputs. From a top view of the
silicon, RCLK[0..3] are in the top-left quadrant, RCLK[8..11] are in
the top-right quadrant, RCLK[4..7] are in the bottom-left quadrant, and
1–40
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
RCLK[12..15] are in the bottom-right quadrant. The regional clock
networks only pertain to the quadrant they drive into. The regional clock
networks provide the lowest clock delay and skew for logic contained
within a single quadrant. RCLK clock networks cannot be driven by
internal logic. The CLK clock pins symmetrically drive the RCLK networks
within a particular quadrant, as shown in Figure 1–20. See Figures 1–21
and 1–22 for RCLK connections from PLLs and CLK pins.
Figure 1–20. Regional Clocks
RCLK[2..3]
RCLK[11..10]
CLK[15..12]
RCLK[9..8]
RCLK[1..0]
CLK[3..0]
CLK[11..8]
RCLK[14..15]
RCLK[4..5]
CLK[7..4]
Regional Clocks Only Drive a Device
Quadrant from Specified CLK Pins or
PLLs within that Quadrant
RCLK[6..7]
RCLK[12..13]
Clock Input Connections
Two CLK pins drive each enhanced PLL. You can use either one or both
pins for clock switchover inputs into the PLL. Either pin can be the
primary clock source for clock switchover, which is controlled in the
Quartus II software. Enhanced PLLs 5 and 6 also have feedback input
pins as shown in Table 1–14.
Altera Corporation
July 2005
1–41
Stratix Device Handbook, Volume 2
Clocking
Input clocks for fast PLLs 1, 2, 3, and 4 come from CLK pins. Stratix GX
devices use PLLs 3 and 4 in the HSSI block only. A multiplexer chooses
one of two possible CLK pins to drive each PLL. This multiplexer is not a
clock switchover multiplexer and is only used for clock input
connectivity.
Either a FPLLCLK input pin or a CLK pin can drive the fast PLLs in the
corners (7, 8, 9, and 10) when used for general purpose. CLK pins cannot
drive these fast PLLs in high-speed differential I/O mode. PLLs 9 and 10
are used for the HSSI block in Stratix GX devices and are not available.
Table 1–14 shows which PLLs are available for each Stratix device and
which input clock pin drives which PLLs.
Table 1–14. Stratix Clock Input Sources For Enhanced & Fast PLLs (Part 1 of 2)
EP1S30, EP1S40, EP1S60 &
EP1S80 Devices Only
All Stratix Devices
Clock Input
Pins
EP1S40 (3),
EP1S60 &
EP1S80
Devices Only
PLL 1 PLL 2 PLL 3 PLL 4 PLL 5 PLL 6 PLL 7 PLL 8 PLL 9 PLL
PLL
PLL
(1)
(1)
(1)
(1)
(2)
(2)
(1)
(1)
(1) 10 (1) 11 (2) 12 (2)
CLK0p/n
v
CLK1p/n
v
v
CLK2p/n
v
CLK3p/n
v
v
CLK4p/n
v
CLK5p/n
v
CLK6p/n
v
CLK7p/n
v
CLK8p/n
v
CLK9p/n
v
v
CLK10p/n
v
CLK11p/n
v
v
CLK12p/n
v
CLK13p/n
v
CLK14p/n
v
CLK15p/n
v
1–42
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
Table 1–14. Stratix Clock Input Sources For Enhanced & Fast PLLs (Part 2 of 2)
EP1S30, EP1S40, EP1S60 &
EP1S80 Devices Only
All Stratix Devices
Clock Input
Pins
EP1S40 (3),
EP1S60 &
EP1S80
Devices Only
PLL 1 PLL 2 PLL 3 PLL 4 PLL 5 PLL 6 PLL 7 PLL 8 PLL 9 PLL
PLL
PLL
(1)
(1)
(1)
(1)
(2)
(2)
(1)
(1)
(1) 10 (1) 11 (2) 12 (2)
v
FPll7clk
v
FPll8clk
v
FPll9clk
v
FPll10clk
Clock Feedback Input Pins
Pll5_fbp/n
Pll6_fbp/n
v
v
Notes to Table 1–14:
(1)
(2)
(3)
This is a fast PLL. The global or regional clocks in a fast PLL’s quadrant can drive the fast PLL input. A pin or other
PLL must drive the global or regional source. The source cannot be driven by internally generated logic before
driving the fast PLL.
This is an enhanced PLL.
The EP1S40 device in the F780 package does not support PLLs 11 and 12.
Clock Output Connections
Enhanced PLLs have outputs for two regional clock outputs and four
global outputs. There is line sharing between clock pins, global and
regional clock networks and all PLL outputs. Check Tables 1–15 and 1–16
and Figures 1–21 and 1–22 to make sure that the clocking scheme is valid.
The Quartus II software automatically maps to regional and global clocks
to avoid any restrictions. Enhanced PLLs 5 and 6 drive out to singleended pins as shown in Table 1–15. PLLs 11 and 12 drive out to singleended pins.
You can connect each fast PLL 1, 2, 3, or 4 outputs (g0, l0, and l1) to either
a global or a regional clock. (PLLs 3 and 4 are not available on Stratix GX
devices.) There is line sharing between clock pins, FPLLCLK pins, global
and regional clock networks and all PLL outputs. Check Figures 1–21 and
1–22 to make sure that the clocking is valid. The Quartus II software
automatically maps to regional and global clocks to avoid any
restrictions.
Altera Corporation
July 2005
1–43
Stratix Device Handbook, Volume 2
Clocking
Table 1–15 shows the global and regional clocks that each PLL drives
outputs to for Stratix devices. Table 1–16 shows the global and regional
clock network each of the CLK and FPLLCLK pins drive when bypassing
the PLL.
Table 1–15. Stratix Global & Regional Clock Output Line Sharing for Enhanced & Fast PLLs (Part 1 of 2)
EP1S30, EP1S40, EP1S60 &
EP1S80 Devices Only
All Devices
Clock
Network
EP1S40 (5),
EP1S60 &
EP1S80
Devices Only
PLL 1 PLL 2 PLL 3 PLL 4 PLL 5 PLL 6 PLL 7 PLL 8 PLL PLL
PLL
PLL
(1)
(1)
(1)
(1)
(2)
(2)
(1)
(1) 9 (1) 10 (1) 11 (2) 12 (2)
GCLK0
v
v
v
v
GCLK1
v
v
v
v
GCLK2
v
v
v
v
GCLK3
v
v
v
v
GCLK4
v
v
GCLK5
v
v
GCLK6
v
v
GCLK7
v
v
GCLK8
v
v
v
v
GCLK9
v
v
v
v
GCLK10
v
v
v
v
GCLK11
v
v
v
v
GCLK12
v
v
GCLK13
v
v
GCLK14
v
v
GCLK15
v
v
RCLK0
v
v
v
RCLK1
v
v
v
RCLK2
v
v
RCLK3
v
v
RCLK4
v
v
v
RCLK5
v
v
v
1–44
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
Table 1–15. Stratix Global & Regional Clock Output Line Sharing for Enhanced & Fast PLLs (Part 2 of 2)
EP1S30, EP1S40, EP1S60 &
EP1S80 Devices Only
All Devices
Clock
Network
EP1S40 (5),
EP1S60 &
EP1S80
Devices Only
PLL 1 PLL 2 PLL 3 PLL 4 PLL 5 PLL 6 PLL 7 PLL 8 PLL PLL
PLL
PLL
(1)
(1)
(1)
(1)
(2)
(2)
(1)
(1) 9 (1) 10 (1) 11 (2) 12 (2)
RCLK6
v
v
RCLK7
v
v
RCLK8
v
v
v
RCLK9
v
v
v
RCLK10
v
v
RCLK11
v
v
RCLK12
v
v
RCLK13
v
v
RCLK14
v
v
v
RCLK15
v
v
v
External Clock Output
PLL5_OUT
[3..0]p/n
PLL6_OUT
[3..0]p/n
v
v
PLL11_OUT
v
(3)
PLL12_OUT
v
(4)
Notes to Table 1–15:
(1)
(2)
(3)
(4)
(5)
This is a fast PLL.
This is an enhanced PLL.
This pin is a tri-purpose pin; it can be an I/O pin, CLK13n, or used for PLL 11 output.
This pin is a tri-purpose pin; it can be an I/O pin, CLK7n, or used for PLL 12 output.
The EP1S40 device in the F780 package does not support PLLs 11 and 12.
Altera Corporation
July 2005
1–45
Stratix Device Handbook, Volume 2
Clocking
Table 1–16. Stratix CLK & FPLLCLK Input Pin Connections to Global & Regional Clock Networks Note (1)
CLK Pins
FPLLCLK (2)
Clock Network
0
GCLK0
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15
v
v v
v
GCLK4
v
GCLK5
v
GCLK6
v
GCLK7
v
GCLK8
v v
v
GCLK9
v v
v
GCLK10
v v
v
GCLK11
v v
v
GCLK12
v
GCLK13
v
GCLK14
v
GCLK15
v
v
v
v
v
RCLK2
RCLK5
v
v
RCLK3
RCLK4
10
v v
v
GCLK3
RCLK1
9
v v
v
GCLK2
RCLK0
8
v v
v
GCLK1
7
v
v
v
RCLK6
RCLK7
v
v
v
RCLK8
v
RCLK9
v
v
RCLK11
1–46
Stratix Device Handbook, Volume 2
v
v
RCLK10
RCLK12
v
v
v
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
Table 1–16. Stratix CLK & FPLLCLK Input Pin Connections to Global & Regional Clock Networks Note (1)
CLK Pins
FPLLCLK (2)
Clock Network
0
RCLK13
RCLK14
RCLK15
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15
7
8
9
10
v
v
v
Notes to Table 1–16:
(1)
(2)
The CLK and FPLLCLK pins cannot drive.
The FPLLCLK pin is only available in EP1S80, EP1S60, EP1S40, and EP1S30 devices.
The fast PLLs also drive high-speed SERDES clocks for differential I/O
interfacing. For information on these FPLLCLK pins, see the High-Speed
Differential I/O Interfaces in Stratix Devices chapter.
Figure 1–21 shows the global and regional clock input and output
connections from the enhanced. Figure 1–21 shows graphically the same
information as Tables 1–15 and 1–16 but with the added detail of where
each specific PLL output port drives to.
Altera Corporation
July 2005
1–47
Stratix Device Handbook, Volume 2
Clocking
Figure 1–21. Global & Regional Clock Connections from Side Clock Pins & Fast PLL Outputs
RCLK1
RCLK0
FPLL7CLK
G3
G1
G0
G2
G8
G9
G10
G11
RCLK9
RCLK8
l0
l0
PLL 7 l1
CLK0
CLK1
g0
g0
l0
l0
PLL 1 l1
l1 PLL 4
g0
CLK2
CLK3
CLK10
CLK11
g0
l02
PLL 2 l1
g0
2l0
l1 PLL 3
g0
l0
FPLL8CLK
FPLL10CLK
l1 PLL 10
CLK8
CLK9
l0
PLL 8 l1
g0
l1 PLL 9
g0
RCLK4
RCLK5
Regional
Clocks
FPLL9CLK
RCLK14
Global
Clocks
RCLK15
Regional
Clocks
Notes to Figures 1–21:
(1)
(2)
The global or regional clocks in a fast PLL’s quadrant can drive the fast PLL input. A dedicated pin or other PLL
must drive the global or regional source. The source cannot be driven by internally generated logic before driving
the fast PLL.
PLLs 3, 4, 9, and 10 are used for the HSSI block in Stratix GX devices and are not available for this use.
When using a fast PLL to compensate for clock delays to drive logic on
the chip, the clock delay from the input pin to the clock input port of the
PLL is compensated only if the clock is fed by the dedicated input pin
closest to the PLL. If the fast PLL gets its input clock from a global or
regional clock or from another dedicated clock pin, which does not
directly feed the fast PLL, the clock signal is first routed onto a global
clock network. The signal then drives into the PLL. In this case, the clock
delay is not fully compensated and the delay compensation is equal to the
clock delay from the dedicated clock pin closest to the PLL to the clock
input port of the PLL.
For example, if you use CLK0 to feed PLL 7, the input clock path delay is
not fully compensated, but if FPLL7CLK feeds PLL 7, the input clock path
delay is fully compensated.
Figure 1–22 shows the global and regional clock input and output
connections from the fast PLLs. Figure 1–22 shows graphically the same
information as Tables 1–15 and 1–16 but with the added detail of where
each specific PLL output port drives to.
1–48
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
Figure 1–22. Global & Regional Clock Connections from Top Clock Pins & Enhanced PLL Outputs
PLL5_OUT[3..0] CLK14 (1)
PLL5_FB
CLK15 (2)
CLK12 (1)
CLK13 (2)
E[0..3]
PLL 5
PLL 11
L0 L1 G0 G1 G2 G3
G0 G1 G2 G3 L0 L1
PLL11_OUT
RCLK10
RCLK11
Regional
Clocks
RCLK2
RCLK3
G12
G13
G14
G15
Global
Clocks
Regional
Clocks
G4
G5
G6
G7
RCLK6
RCLK7
RCLK12
RCLK13
PLL12_OUT
L0 L1 G0 G1 G2 G3
G0 G1 G2 G3 L0 L1
PLL 6
PLL6_OUT[3..0]
PLL 12
PLL6_FB
CLK4 (1)
CLK6 (1)
CLK7 (2)
CLK5 (2)
Notes to Figures 1–22:
(1)
(2)
CLK4, CLK6, CLK12, and CLK14 feed the corresponding PLL’s inclk0 port.
CLK5, CLK7, CLK13, and CLK15 feed the corresponding PLL’s inclk1 port.
Altera Corporation
July 2005
1–49
Stratix Device Handbook, Volume 2
Board Layout
Board Layout
The enhanced and fast PLL circuits in Stratix and Stratix GX devices
contain analog components embedded in a digital device. These analog
components have separate power and ground pins to minimize noise
generated by the digital components. Both Stratix and Stratix GX
enhanced and fast PLLs use separate VCC and ground pins to isolate
circuitry and improve noise resistance.
VCCA & GNDA
Each enhanced and fast PLL uses separate VCC and ground pin pairs for
their analog circuitry. The analog circuit power and ground pin for each
PLL is called PLL<PLL number>_VCCA and PLL<PLL number>_GNDA.
Connect the VCCA power pin to a 1.5-V power supply, even if you do not
use the PLL. Isolate the power connected to VCCA from the power to the
rest of the Stratix and Stratix GX device or any other digital device on the
board. You can use one of three different methods of isolating the VCCA
pin: separate VCCA power planes, a partitioned VCCA island within the
VCCINT plane, and thick VCCA traces.
Separate VCCA Power Plane
A mixed signal system is already partitioned into analog and digital
sections, each with its own power planes on the board. To isolate the
VCCA pin using a separate VCCA power plane, connect the VCCA pin to the
analog 1.5-V power plane.
Partitioned VCCA Island within VCCINT Plane
Fully digital systems do not have a separate analog power plane on the
board. Because it is expensive to add new planes to the board, you can
create islands for VCCA_PLL. Figure 1–23 shows an example board layout
with an analog power island. The dielectric boundary that creates the
island should be 25 mils thick. Figure 1–23 shows a partitioned plane
within VCCINT for VCCA.
1–50
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
Figure 1–23. VCCINT Plane Partitioned for VCCA Island
Thick VCCA Trace
Because of board constraints, you might not be able to partition a VCCA
island. Instead, run a thick trace from the power supply to each VCCA pin.
The traces should be at least 20 mils thick.
In each of these three cases, you should filter each VCCA pin with a
decoupling circuit shown in Figure 1–24. Place a ferrite bead that exhibits
high impedance at frequencies of 50 MHz or higher and a 10-μF tantalum
parallel capacitor where the power enters the board. Decouple each VCCA
pin with a 0.1-μF and 0.001-μF parallel combination of ceramic capacitors
located as close as possible to the Stratix or Stratix GX device. You can
connect the GNDA pins directly to the same ground plane as the device’s
digital ground.
Altera Corporation
July 2005
1–51
Stratix Device Handbook, Volume 2
Board Layout
Figure 1–24. PLL Power Schematic for Stratix or Stratix GX PLLs
Ferrite
Bead
1.5-V Supply
10 μF
PLL<PLL number>_VCCA
0.1 μF
0.001 μF
PLL<PLL number>_GNDA
VCCINT
PLL<PLL number>_VCCG
Repeat for Each PLL
Power and Ground Set
PLL<PLL number>_GNDG
Stratix Device
VCCG & GNDG
The guard ring power and ground pins are called
PLL<PLL number>_VCCG and PLL<PLL number>_GNDG. The guard ring
isolates the PLL circuit from the rest of the device. Connect these guard
ring VCCG pins to the quietest digital supply on the board. In most
systems, this is the digital 1.5-V supply supplied to the device's VCCINT
pins. Connect the VCCG pins to a power supply even if you do not use the
PLL. You can connect the GNDG pins directly to the same ground plane as
the device’s digital ground. See Figure 1–24.
1–52
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
External Clock Output Power
Enhanced PLLs 5 and 6 also have isolated power pins for their dedicated
external clock outputs (VCC_PLL5_OUTA and VCC_PLL5_OUTB, or
VCC_PLL6_OUTA and VCC_PLL6_OUTB, respectively). PLLs 5 and 6 both
have two banks of outputs. Each bank is powered by a unique output
power, OUTA or OUTB, as illustrated in Figure 1–25. These outputs can by
powered by 3.3, 2.5, 1.8, or 1.5 V depending on the I/O standard for the
clock output in the A or B groups.
Altera Corporation
July 2005
1–53
Stratix Device Handbook, Volume 2
Board Layout
Figure 1–25. External Clock Output Pin Association to Output Power Note (1)
VCC_PLL5_OUTA
PLL5_OUT0p
PLL5_OUT0n
PLL5_OUT0p
PLL5_OUT0n
VCC_PLL5_OUTB
PLL5_OUT2p
PLL5_OUT2n
PLL5_OUT3p
PLL5_OUT3n
Note to Figure 1–25:
(1)
These pins apply to PLL 5. The figure for PLL 6 is similar, except that the pin names
begin with the prefix PLL6 instead of PLL5.
1–54
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
General-Purpose PLLs in Stratix & Stratix GX Devices
Filter each isolated power pin with a decoupling circuit shown in
Figure 1–26. Decouple the isolated power pins with a 0.1-μF and a
0.001-μF parallel combination of ceramic capacitors located as close as
possible to the Stratix device.
Figure 1–26. Stratix PLL External Clock Output Power Ball Connections
Note (1)
VCCIO
Supply
VCC_PLL5_OUTA
0.1 μF
0.001 μF
VCC_PLL5_OUTB
0.1 μF
0.001 μF
Stratix Device
Note to Figure 1–26:
(1)
Altera Corporation
July 2005
Figure 1–26 also applies to VCC_PLL6_OUTA/B.
1–55
Stratix Device Handbook, Volume 2
Conclusion
Guidelines
Use the following guidelines for optimal jitter performance on the
external clock outputs from enhanced PLLs 5 and 6. If all outputs are
running at the same frequency, these guidelines are not necessary to
improve performance.
■
■
■
When driving two or more clock outputs from PLL 5 or 6, separate
the outputs into the two groups shown in Figure 1–24. For example,
if you are driving 100- and 200-MHz clock outputs off-chip from PLL
5, place one output on PLL5_OUT0p (powered by VCC_PLL5_OUTA)
and the other output on PLL5_OUT2p (powered by
VCC_PLL5_OUTB). Since the output buffers are powered by different
pins, they are less susceptible to bimodal jitter. Bimodal jitter is a
deterministic jitter not caused by the PLL but rather by coincident
edges of clock outputs that are multiples of each other.
Use phase shift to ensure edges are not coincident on all the clock
outputs.
Use phase shift to skew clock edges with respect to each other for
best jitter performance.
1
■
Conclusion
Delay shift (time delay elements) are no longer supported
in Stratix PLLs. Use the phase shift feature to implement the
desired time shift.
If you cannot drive multiple clocks of different frequencies and
phase shifts or isolate banks, you should control the drive capability
on the lower frequency clock. Reducing how much current the
output buffer has to supply can reduce the noise. Minimize
capacitive load on the slower frequency output and configure the
output buffer to drive slow slew rate and lower current strength. The
higher frequency output should have an improved performance, but
this may degrade the performance of your lower frequency clock
output.
Stratix and Stratix GX device enhanced PLLs provide you with complete
control of your clocks and system timing. These PLLs are capable of
offering flexible system level clock management that was previously only
available in discrete PLL devices. The embedded PLLs meet and exceed
the features offered by these high-end discrete devices, reducing the need
for other timing devices in the system.
1–56
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Section II. Memory
This section provides information on the TriMatrix™ Embedded Memory
blocks internal to Stratix® devices and the supported external memory
interfaces.
It contains the following chapters:
■
Chapter 2, TriMatrix Embedded Memory Blocks in
Stratix & Stratix GX Devices
■
Chapter 3, External Memory Interfaces in Stratix & Stratix GX
Devices
The QDR SRAM Controller Reference Design for Stratix & Stratix GX Devices
chapter is removed in this version of the Stratix Device Handbook. The
information is available in AN 349: Interfacing QDR SRAM with Stratix and
Stratix GX Devices.
Revision History
The table below shows the revision history for Chapters 2 and 3.
Chapter
Date/Version
Changes Made
2
July 2005, v3.3
●
Updated “Implementing True Dual-Port Mode” section.
January 2005,
v3.2
●
Minor technical content update.
September
2004, v3.1
●
Updated Note 1 in Figure 2–12 on page 2–22.
Updated description about using two different clocks in a
dual-port RAM on page 2–27.
Deleted description of M-RAM block and document
references on page 2–27.
●
●
April 2004, v3.0
●
●
●
July 2003, v2.0
●
●
●
Altera Corporation
Comments
Synchronous occurrences are renamed to pipelined.
Pseudo-synchronous occurrences are renamed flowthrough.
Added AND gate to Figure 2–12.
Updated performance specification for TriMatrix memory
in Table 2-1.
Added addressing example for a RAM that is using
mixed-width mode, page 2-9.
Added Note 1 to Tables 2-9 and 2-10, Note 3 to Figure 211, and Note 2 to Figures 2-12 and 2-13.
Section II–1
Memory
Stratix Device Handbook, Volume 2
Chapter
Date/Version
3
June 2006, v3.3
Changes Made
●
●
July 2005, v3.2
●
●
September
2004, v3.1
●
●
●
●
●
●
●
April 2004, v3.0
●
●
●
●
●
●
Updated mathematical symbols in Table 3–3.
Updated “DQS Phase-Shift Circuitry” section.
Moved Figure 8 to become Figure 1, “Example of Where
a DQS Signal is Center-Aligned in the IOE” on page 3–3.
Updated Table 3–1 on page 3–10, updated Note 4. Note
4, 5, and 6, are now Note 5, 6, and 7, respectively.
Updated Table 3–2 on page 3–10.
Updated Table 3–3 on page 3–13.
Updated Note on page 3–14.
Moved the “External Memory Standards” on page 3–1 to
follow the Introduction section.
Moved “Conclusion” on page 3–27 to end of chapter.
Chapter renamed Chapter 3, External Memory Interfaces
in Stratix & Stratix GX Devices.
Table 3–1: DDR SDRAM - side banks row added, ZBT
SRAM row updated.
Added Tables 3–2 and 3–4.
DQSn pins removed (page 3-5)
Deleted “QDR SRAM Interfacing” figure.
Replaced “tZX & tXZ Timing Diagram.”
●
Removed support for series and parallel on-chip
termination.
July 2003, v2.0
●
altddio_bidir function is used for DQS in versions before
Quartus II 3.0. (page 3-2)
Updated naming convention for DQS pins on page 3-9 to
match pin tables.
Clarified input clock to PLL must come from an external
input pin on page 3-12.
●
Section II–2
Changed the name of the chapter from External Memory
Interfaces to External Memory Interfaces in Stratix &
Stratix GX Devices to reflect its shared status between
those device handbooks.
Added cross reference regarding frequency limits for 72
and 90° phase shift for DQS.
November 2003,
v2.1
●
Comments
Altera Corporation
2. TriMatrix Embedded
Memory Blocks in
Stratix & Stratix GX Devices
S52003-3.3
Introduction
Stratix® and Stratix GX devices feature the TriMatrix™ memory
structure, composed of three sizes of embedded RAM blocks. TriMatrix
memory includes 512-bit M512 blocks, 4-Kbit M4K blocks, and 512-Kbit
M-RAM blocks, each of which is configurable to support a wide range of
features. Offering up to 10 Mbits of RAM and up to 12 terabits per second
of device memory bandwidth, the TriMatrix memory structure makes the
Stratix and Stratix GX families ideal for memory-intensive applications.
TriMatrix
Memory
TriMatrix memory structures can implement a wide variety of complex
memory functions. For example, use the small M512 blocks for first-in
first-out (FIFO) functions and clock domain buffering where memory
bandwidth is critical. The M4K blocks are an ideal size for applications
requiring medium-sized memory, such as asynchronous transfer mode
(ATM) cell processing. M-RAM blocks enhance programmable logic
device (PLD) memory capabilities for large buffering applications, such
as internet protocol (IP) packet buffering and system cache.
TriMatrix memory blocks support various memory configurations,
including single-port, simple dual-port, true dual-port (also known as
bidirectional dual-port), shift-register, ROM, and FIFO mode. The
TriMatrix memory architecture also includes advanced features and
capabilities, such as byte enable support, parity-bit support, and mixedport width support. This chapter describes the various TriMatrix memory
modes and features.
Table 2–1 summarizes the features supported by the three sizes of
TriMatrix memory.
f
Altera Corporation
July 2005
For more information on selecting which memory block to use, see
AN 207: TriMatrix Memory Selection Using the Quartus II Software.
2–1
TriMatrix Memory
Table 2–1. Summary of TriMatrix Memory Features
Feature
Performance
Total RAM bits (including parity bits)
Configurations
Parity bits
M512 Block
M4K Block
M-RAM Block
319 MHz
290 MHz
287 MHz
576
4,608
589,824
512 × 1
256 × 2
128 × 4
64 × 8
64 × 9
32 × 16
32 × 18
4K × 1
2K × 2
1K × 4
512 × 8
512 × 9
256 × 16
256 × 18
128 × 32
128 × 36
64K × 8
64K × 9
32K × 16
32K × 18
16K × 32
16K × 36
8K × 64
8K × 72
4K × 128
4K × 144
v
v
v
v
v
Byte enable
Single-port memory
v
v
v
Simple dual-port memory
v
v
v
v
v
True dual-port memory
Embedded shift register
v
v
ROM
v
v
FIFO buffer
v
v
v
Simple dual-port mixed width support
v
v
v
v
v
True dual-port mixed width support
Memory initialization file (.mif)
Mixed-clock mode
v
v
v
v
v
Power-up condition
Outputs cleared
Outputs cleared
Outputs unknown
Register clears
Input and output
registers (1)
Input and output
registers (2)
Output registers
Same-port read-during-write
New data available at
positive clock edge
New data available at
positive clock edge
New data available at
positive clock edge
Mixed-port read-during-write
Outputs set to
unknown or old data
Outputs set to
unknown or old data
Unknown output
Notes to Table 2–1:
(1)
(2)
The rden register on the M512 memory block does not have a clear port.
On the M4K block, asserting the clear port of the rden and byte enable registers drives the output of these registers
high.
2–2
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
TriMatrix Embedded Memory Blocks in Stratix & Stratix GX Devices
The extremely high memory bandwidth of the Stratix and Stratix GX
device families is a result of increased memory capacity and speed.
Table 2–2 shows the memory capacity for TriMatrix memory blocks in
each Stratix device. Table 2–3 shows the memory capacity for TriMatrix
memory blocks in each Stratix GX device.
Table 2–2. TriMatrix Memory Distribution in Stratix Devices
Device
M512
M4K
Columns/Blocks Columns/Blocks
EP1S10
4 / 94
EP1S20
EP1S25
M-RAM
Blocks
Total RAM Bits
2 / 60
1
920,448
6 / 194
2 / 82
2
1,669,248
6 / 224
3 / 138
2
1,944,576
EP1S30
7 / 295
3 / 171
4
3,317,184
EP1S40
8 / 384
3 / 183
4
3,423,744
EP1S60
10 / 574
4 / 292
6
5,215,104
EP1S80
11 / 767
4 / 364
9
7,427,520
Table 2–3. TriMatrix Memory Distribution in Stratix GX Devices
Device
M512
M4K
Columns/Blocks Columns/Blocks
M-RAM
Blocks
Total RAM Bits
EP1SGX10
4 / 94
2 / 60
1
920,448
EP1SGX25
6 / 224
3 / 138
2
1,944,576
EP1SGX40
8 / 384
3 / 183
4
3,423,744
Clear Signals
When applied to input registers, the asynchronous clear signal for the
TriMatrix embedded memory immediately clears the input registers.
However, the output of the memory block does not show the effects until
the next clock edge. When applied to output registers, the asynchronous
clear signal clears the output registers and the effects are seen
immediately.
Parity Bit Support
The memory blocks support a parity bit for each byte. Parity bits are in
addition to the amount of memory in each RAM block. For example, the
M512 block has 576 bits, 64 of which are optionally used for parity bit
Altera Corporation
July 2005
2–3
Stratix Device Handbook, Volume 2
TriMatrix Memory
storage. The parity bit, along with logic implemented in logic elements
(LEs), can implement parity checking for error detection to ensure data
integrity. Parity-size data words can also store user-specified control bits.
Byte Enable Support
In the M4K and M-RAM blocks, byte enables can mask the input data so
that only specific bytes of data are written. The unwritten bytes retain the
previous written value. The write enable signals (wren), in conjunction
with the byte enable signals (byteena), controls the RAM block’s write
operations. The default value for the byteena signals is high (enabled),
in which case writing is controlled only by the wren signals.
Asserting the clear port of the byte enable registers drives the byte enable
signals to their default high level.
M4K Blocks
M4K blocks support byte writes when the write port has a data width of
16, 18, 32, or 36 bits. Table 2–4 summarizes the byte selection.
Table 2–4. Byte Enable for M4K Blocks Notes (1), (2)
byteena
datain × 18
datain × 36
[0] = 1
[8..0]
[8..0]
[1] = 1
[17..9]
[17..9]
[2] = 1
–
[26..18]
[3] = 1
–
[35..27]
Notes to Table 2–4:
(1)
(2)
Any combination of byte enables is possible.
Byte enables can be used in the same manner with 8-bit words, i.e., in × 16 and × 32
modes.
2–4
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
TriMatrix Embedded Memory Blocks in Stratix & Stratix GX Devices
M-RAM Blocks
M-RAM blocks support byte enables for the × 16, × 18, × 32, × 36, × 64, and
× 72 modes. In the × 128 or × 144 simple dual-port mode, the two sets of
byteena signals (byteena_a and byteena_b) combine to form the
necessary 16 byte enables. Tables 2–5 and 2–6 summarize the byte
selection.
Table 2–5. Byte Enable for M-RAM Blocks Notes (1), (2)
byteena
datain × 18
datain × 36
datain × 72
[0] = 1
[8..0]
[8..0]
[8..0]
[1] = 1
[17..9]
[17..9]
[17..9]
[2] = 1
–
[26..18]
[26..18]
[3] = 1
–
[35..27]
[35..27]
[4] = 1
–
–
[44..36]
[5] = 1
–
–
[53..45]
[6] = 1
–
–
[62..54]
[7] = 1
–
–
[71..63]
Notes to Table 2–5:
(1)
(2)
Any combination of byte enables is possible.
Byte enables can be used in the same manner with 8-bit words, that is, in × 16, × 32,
and × 64 modes.
Table 2–6. M-RAM Combined Byte Selection for × 144 Mode (Part 1 of 2),
Notes (1), (2)
byteena_a
Altera Corporation
July 2005
datain × 144
[0] = 1
[8..0]
[1] = 1
[17..9]
[2] = 1
[26..18]
[3] = 1
[35..27]
[4] = 1
[44..36]
[5] = 1
[53..45]
[6] = 1
[62..54]
[7] = 1
[71..63]
[8] = 1
[80..72]
[9] = 1
[89..81]
[10] = 1
[98..90]
[11] = 1
[107..99]
2–5
Stratix Device Handbook, Volume 2
TriMatrix Memory
Table 2–6. M-RAM Combined Byte Selection for × 144 Mode (Part 2 of 2),
Notes (1), (2)
byteena_a
datain × 144
[12] = 1
[116..108]
[13] = 1
[125..117]
[14] = 1
[134..126]
[15] = 1
[143..135]
Notes to Table 2–6:
(1)
(2)
Any combination of byte enables is possible.
Byte enables can be used in the same manner with 8-bit words, i.e., in × 16, × 32,
× 64, and × 128 modes.
Byte Enable Functional Waveform
Figure 2–1 shows how both the wren and the byteena signals control
the write operations of the RAM.
Figure 2–1. Byte Enable Functional Waveform Note (1)
inclock
wren
a0
address
an
data_in
XXXX
byteena
XX
contents at a0
contents at a1
a2
a0
a1
ABCD
10
a2
XXXX
01
11
FFFF
XX
ABFF
FFFF
FFCD
FFFF
contents at a2
asynch_data_out
a1
doutn
ABXX
ABCD
XXCD
ABCD
ABFF
FFCD
ABCD
Note to Figure 2–1:
(1)
For more information on simulation output when a read-during-write occurs at the same address location, see
“Read-During-Write Operation at the Same Address” on page 2–25.
2–6
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
TriMatrix Embedded Memory Blocks in Stratix & Stratix GX Devices
Using TriMatrix
Memory
f
The TriMatrix memory blocks include input registers that synchronize
writes and output registers to pipeline designs and improve system
performance. All TriMatrix memory blocks are pipelined, meaning that
all inputs are registered, but outputs are either registered or
combinatorial. TriMatrix memory can emulate a flow-through memory
by using combinatorial outputs.
For more information, see AN 210: Converting Memory from Asynchronous
to Synchronous for Stratix & Stratix GX Designs.
Depending on the TriMatrix memory block type, the memory can have
various modes, including:
■
■
■
■
■
■
Single-port
Simple dual-port
True dual-port (bidirectional dual-port)
Shift-register
ROM
FIFO
Implementing Single-Port Mode
Single-port mode supports non-simultaneous reads and writes.
Figure 2–2 shows the single-port memory configuration for TriMatrix
memory. All memory block types support the single-port mode.
Figure 2–2. Single-Port Memory Note (1)
data[ ]
address[ ]
wren
inclock
inclocken
inaclr
q[ ]
outclock
outclocken
outaclr
Note to Figure 2–2:
(1)
Two single-port memory blocks can be implemented in a single M4K block.
M4K memory blocks can also be divided in half and used for two
independent single-port RAM blocks. The Altera Quartus II software
automatically uses this single-port memory packing when running low
on memory resources. To force two single-port memories into one M4K
block, first ensure that each of the two independent RAM blocks is equal
to or less than half the size of the M4K block. Second, assign both singleport RAMs to the same M4K block.
Altera Corporation
July 2005
2–7
Stratix Device Handbook, Volume 2
Using TriMatrix Memory
In the single-port RAM configuration, the outputs can only be in
read-during-write mode, which means that during the write operation,
data written to the RAM flows through to the RAM outputs. When the
output registers are bypassed, the new data is available on the rising edge
of the same clock cycle it was written on. For more information about
read-during-write mode, see “Read-During-Write Operation at the Same
Address” on page 2–25.
Figure 2–3 shows timing waveforms for read and write operations in
single-port mode.
Figure 2–3. Single-Port Timing Waveforms
in clock
wren
address
an-1
an
data_in
din-1
din
synch_data_out
asynch_data_out
a0
din-2
din
din-1
din
din-1
a1
a2
dout0
dout0
dout1
a3
dout1
dout2
a4
a5
a6
din4
din5
din6
dout2
dout3
dout3
din4
din4
din5
Implementing Simple Dual-Port Mode
Simple dual-port memory supports a simultaneous read and write.
Figure 2–4 shows the simple dual-port memory configuration for
TriMatrix memory. All memory block types support this configuration.
Figure 2–4. Simple Dual-Port Memory Note (1)
Dual-Port Memory
data[ ]
wraddress[ ]
wren
inclock
inclocken
inaclr
rdaddress[ ]
rden
q[ ]
outclock
outclocken
outaclr
Note to Figure 2–4:
(1)
Simple dual-port RAM supports read/write clock mode in addition to the
input/output clock mode shown.
2–8
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
TriMatrix Embedded Memory Blocks in Stratix & Stratix GX Devices
TriMatrix memory supports mixed-width configurations, allowing
different read and write port widths. When using mixed-width mode, the
LSB is written to or read from first. For example, take a RAM that is set up
in mixed-width mode with write data width ×8 and read data width ×2.
If a binary 00000001 is written to write dress 0, the following is read out
of the ×2 output side:
Read Address
×2 data
00
01(LSB of ×8 data)
01
00
10
00
11
00(MSB of ×8 data)
Tables 2–7 to 2–9 show the mixed width configurations for the M512,
M4K, and M-RAM blocks, respectively.
Table 2–7. M512 Block Mixed-Width Configurations (Simple Dual-Port Mode)
Write Port
Read Port
512 × 1
256 × 2
128 × 4
64 × 8
32 × 16
v
v
v
v
v
v
v
512 × 1
256 × 2
v
v
v
128 × 4
v
v
v
64 × 8
v
v
32 × 16
v
v
64 × 9
32 × 18
v
v
v
v
64 × 9
v
32 × 18
v
Table 2–8. M4K Block Mixed-Width Configurations (Simple Dual-Port Mode) (Part 1 of 2)
Write Port
Read Port
4K × 1 2K × 2 1K × 4 512 × 8 256 × 16
128 × 32 512 × 9 256 × 18
4K × 1
v
v
v
v
v
v
2K × 2
v
v
v
v
v
v
1K × 4
v
v
v
v
v
v
512 × 8
v
v
v
v
v
v
256 × 16
v
v
v
v
v
v
Altera Corporation
July 2005
128 × 36
2–9
Stratix Device Handbook, Volume 2
Using TriMatrix Memory
Table 2–8. M4K Block Mixed-Width Configurations (Simple Dual-Port Mode) (Part 2 of 2)
Write Port
Read Port
128 × 32
4K × 1 2K × 2 1K × 4 512 × 8 256 × 16
v
v
v
v
v
128 × 32 512 × 9 256 × 18
128 × 36
v
512 × 9
v
v
v
256 × 18
v
v
v
128 × 36
v
v
v
Table 2–9. M-RAM Block Mixed-Width Configurations (Simple Dual-Port Mode)
Write Port
Read Port
64K × 9
32K × 18
16K × 36
8K × 72
64K × 9
v
v
v
v
32K × 18
v
v
v
v
16K × 36
v
v
v
v
8K × 72
v
v
v
v
4K × 144
4K × 144
v
M512 blocks support serializer and deserializer (SERDES) applications.
By using the mixed-width support in combination with double data rate
(DDR) I/O standards, the block can function as a SERDES to support lowspeed serial I/O standards using global or regional clocks.
f
For more information on Stratix device I/O structure see the Stratix
Device Family Data Sheet section of the Stratix Device Handbook, Volume 1.
For more information on Stratix GX device I/O structure see the
Stratix GX Device Family Data Sheet section of the Stratix GX Device
Handbook, Volume 1.
In simple dual-port mode, the M512 and M4K blocks have one write
enable and one read enable signal. The M512 does not support a clear port
on the rden register. On the M4K block, asserting the clear port of the
rden register drives rden high, which allows the read operation to occur.
When the read enable is deactivated, the current data is retained at the
output ports. If the read enable is activated during a write operation with
the same address location selected, the simple dual-port RAM output is
either unknown or can be set to output the old data stored at the memory
address. For more information, see “Read-During-Write Operation at the
Same Address” on page 2–25.
2–10
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
TriMatrix Embedded Memory Blocks in Stratix & Stratix GX Devices
M-RAM blocks have one write enable signal in simple dual-port mode. To
perform a write operation, the write enable is held high. The M-RAM
block is always enabled for read operation. If the read address and the
write address select the same address location during a write operation,
the M-RAM block output is unknown.
Figure 2–5 shows timing waveforms for read and write operations in
simple dual-port mode.
Figure 2–5. Simple Dual-Port Timing Waveforms Note (1)
wrclock
wren
wraddress
an-1
an
data_in
din-1
din
a0
a1
a2
a3
a4
a5
a6
din4
din5
din6
rdclock
rden
rdaddress
synch_data_out
asynch_data_out
bn
doutn-2
doutn-1
b1
b0
doutn-1
doutn
b2
doutn
b3
dout0
dout0
Note to Figure 2–5:
(1)
The rden signal is not available in the M-RAM block. A M-RAM block in simple dual-port mode is always reading
out the data stored at the current read address location.
Implementing True Dual-Port Mode
M4K and M-RAM blocks offer a true dual-port mode to support any
combination of two-port operations: two reads, two writes, or one read
and one write at two different clock frequencies. Figure 2–6 shows the
true dual-port memory configuration for TriMatrix memory.
Altera Corporation
July 2005
2–11
Stratix Device Handbook, Volume 2
Using TriMatrix Memory
Figure 2–6. True Dual-Port Memory Note (1)
A
B
dataA[ ]
addressA[ ]
wrenA
clockA
clockenA
qA[ ]
aclrA
dataB[ ]
addressB[ ]
wrenB
clockB
clockenB
qB[ ]
aclrB
Note to Figure 2–6:
(1)
True dual-port memory supports input/output clock mode in addition to the
independent clock mode shown.
The widest bit configuration of the M4K and M-RAM blocks in true dualport mode is 256 × 16-bit (× 18-bit with parity) and 8K × 64-bit (× 72-bit
with parity), respectively. The 128 × 32-bit (× 36-bit with parity)
configuration of the M4K block and the 4K × 128-bit (× 144-bit with parity)
configuration of the M-RAM block are unavailable because the number of
output drivers is equivalent to the maximum bit width of the respective
memory block. Because true dual-port RAM has outputs on two ports,
the maximum width of the true dual-port RAM equals half of the total
number of output drivers. Tables 2–10 and 2–11 list the possible M4K
RAM block and M-RAM block configurations, respectively.
Table 2–10. M4K Block Mixed-Port Width Configurations (True Dual-Port)
Port B
Port A
4K × 1
2K × 2
1K × 4
512 × 8
256 × 16
512 × 9
256 × 18
4K × 1
v
v
v
v
v
2K × 2
v
v
v
v
v
1K × 4
v
v
v
v
v
512 × 8
v
v
v
v
v
256 × 16
v
v
v
v
v
512 × 9
v
v
256 × 18
v
v
2–12
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
TriMatrix Embedded Memory Blocks in Stratix & Stratix GX Devices
Table 2–11. M-RAM Block Mixed-Port Width Configurations (True Dual-Port)
Port B
Port A
64K × 9
32K × 18
16K × 36
8K × 72
64K × 9
v
v
v
v
32K × 18
v
v
v
v
16K × 36
v
v
v
v
8K × 72
v
v
v
v
In true dual-port configuration, the RAM outputs can only be configured
for read-during-write mode. This means that during write operation,
data being written to the A or B port of the RAM flows through to the A
or B outputs, respectively. When the output registers are bypassed, the
new data is available on the rising edge of the same clock cycle it was
written on. For waveforms and information on mixed-port read-duringwrite mode, see “Read-During-Write Operation at the Same Address” on
page 2–25.
Potential write contentions must be resolved external to the RAM because
writing to the same address location at both ports results in unknown
data storage at that location. Data is written on the rising edge of the write
clock for the M-RAM block. For a valid write operation to the same
address of the M-RAM block, the rising edge of the write clock for port A
must occur following the maximum write cycle time interval after the
rising edge of the write clock for port B. Since data is written into the
M512 and M4K blocks at the falling edge of the write clock, the rising
edge of the write clock for port A should occur following half of the
maximum write cycle time interval after the falling edge of the write clock
for port B. If this timing is not met, the data stored in that particular
address is invalid.
f
See the Stratix Device Family Data Sheet section of the Stratix Device
Handbook, Volume 1 or the Stratix GX Device Family Data Sheet section of
the Stratix GX Device Handbook, Volume 1 for the maximum synchronous
write cycle time.
Figure 2–7 shows true dual-port timing waveforms for write operation at
port A and read operation at port B.
Altera Corporation
July 2005
2–13
Stratix Device Handbook, Volume 2
Using TriMatrix Memory
Figure 2–7. True Dual-Port Timing Waveforms
A_clk
A_wren
A_address
A_data_in
an-1
an
din-1
din
A_synch_data_out
din-2
A_asynch_data_out
a0
din
din-1
din
din-1
a1
a2
dout0
dout1
dout1
dout0
a3
a4
a5
a6
din4
din5
din6
dout2
dout2
din4
dout3
dout3
din5
din4
B_clk
B_wren
B_address
B_synch_data_out
B_asynch_data_out
bn
doutn-2
doutn-1
b1
b0
doutn-1
doutn
b2
dout0
doutn
dout0
b3
dout1
dout2
dout1
Implementing Shift-Register Mode
Embedded memory block configurations can implement shift registers
for digital signal processing (DSP) applications, such as finite impulse
response (FIR) filters, pseudo-random number generators, multi-channel
filtering, and auto-correlation and cross-correlation functions. These and
other DSP applications require local data storage, traditionally
implemented with standard flip-flops that can quickly consume many
logic cells for large shift registers. A more efficient alternative is to use
embedded memory as a shift register block, which saves logic cell and
routing resources and provides a more efficient implementation.
The size of a (w × m × n) shift register is determined by the input data
width (w), the length of the taps (m), and the number of taps (n). The size
of a (w × m × n) shift register must be less than or equal to the maximum
number of memory bits in the respective block: 576 bits for the M512
block and 4,608 bits for the M4K block. In addition, the size of w × n must
be less than or equal to the maximum width of the respective block: 18
bits for the M512 block and 36 bits for the M4K block. If a larger shift
register is required, the memory blocks can be cascaded together.
1
2–14
Stratix Device Handbook, Volume 2
M-RAM blocks do not support the shift-register mode.
Altera Corporation
July 2005
TriMatrix Embedded Memory Blocks in Stratix & Stratix GX Devices
Data is written into each address location at the falling edge of the clock
and read from the address at the rising edge of the clock. The shift-register
mode logic automatically controls the positive and negative edge
clocking to shift the data in one clock cycle. Figure 2–8 shows the
TriMatrix memory block in the shift-register mode.
Figure 2–8. Shift-Register Memory Configuration
w × m × n Shift Register
m-Bit Shift Register
w
w
m-Bit Shift Register
w
w
n Number
of Taps
m-Bit Shift Register
w
w
m-Bit Shift Register
w
w
Implementing ROM Mode
The M512 and the M4K blocks support ROM mode. Use a memory
initialization file (.mif) to initialize the ROM contents of M512 and M4K
blocks. The M-RAM block does not support ROM mode.
All Stratix memory configurations must have synchronous inputs;
therefore, the address lines of the ROM are registered. The outputs can be
registered or combinatorial. The ROM read operation is identical to the
read operation in the single-port RAM configuration.
Altera Corporation
July 2005
2–15
Stratix Device Handbook, Volume 2
Clock Modes
Implementing FIFO Buffers
While the small M512 memory blocks are ideal for designs with many
shallow FIFO buffers, all three memory sizes support FIFO mode.
All memory configurations have synchronous inputs; however, the FIFO
buffer outputs are always combinatorial. Simultaneous read and write
from an empty FIFO is not supported.
Clock Modes
Depending on the TriMatrix memory mode, independent, input/output,
read/write, and/or single-port clock modes are available. Table 2–12
shows the clock modes supported by the TriMatrix memory modes.
Table 2–12. TriMatrix Memory Clock Modes
Clocking Mode
True-Dual Port
Mode
Independent
v
Input/output
v
Read/write
Single-port
Simple DualPort Mode
Single-Port
Mode
v
v
v
Independent Clock Mode
The TriMatrix memory blocks can implement independent clock mode
for true dual-port memory. In this mode, a separate clock is available for
each port (A and B). Clock A controls all registers on the port A side,
while clock B controls all registers on the port B side. Each port also
supports independent clock enables and asynchronous clear signals for
port A and B registers. Figure 2–9 shows a TriMatrix memory block in
independent clock mode.
2–16
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
TriMatrix Embedded Memory Blocks in Stratix & Stratix GX Devices
clockB
clkenB
wrenB
addressB[ ]
D
ENA
D
D
Q
ENA
ENA
Q
Q
ENA
D
Q
ENA
ENA
D
clockA
clkenA
wrenA
addressA[ ]
byteenaA[ ]
dataA[ ]
8
D
8 LAB Row Clocks
Q
Write
Pulse
Generator
D
Data Out
Address A
Write/Read
Enable
qA[ ]
qB[ ]
Q
Data Out
Write/Read
Enable
Address B
Byte Enable B
Byte Enable A
B
Data In
Memory Block
256 ´ 16 (2)
512 ´ 8
1,024 ´ 4
2,048 ´ 2
4,096 ´ 1
A
Data In
ENA
Write
Pulse
Generator
Q
D
ENA
Q
D
ENA
Q
D
Q
ENA
8
dataB[ ]
byteenaB[ ]
Figure 2–9. Independent Clock Mode Note (1), (2)
Note to Figure 2–9:
(1)
(2)
Violating the setup or hold time on the address registers could corrupt the memory contents. This applies to both
read and write operations.
All registers shown have asynchronous clear ports, except when using the M-RAM. M-RAM blocks have
asynchronous clear ports on their output registers only.
Altera Corporation
July 2005
2–17
Stratix Device Handbook, Volume 2
Clock Modes
Input/Output Clock Mode
The TriMatrix memory blocks can implement input/output clock mode
for true and simple dual-port memory. On each of the two ports, A and B,
one clock controls all registers for inputs into the memory block: data
input, wren, and address. The other clock controls the block’s data output
registers. Each memory block port also supports independent clock
enables and asynchronous clear signals for input and output registers.
Figures 2–10 and 2–11 show the memory block in input/output clock
mode for true and simple dual-port modes, respectively.
2–18
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
(1)
Altera Corporation
July 2005
clockA
clkenA
wrenA
addressA[ ]
byteenaA[ ]
dataA[ ]
8
ENA
D
ENA
D
ENA
D
ENA
D
8 LAB Row Clocks
Q
Q
Q
Q
Write
Pulse
Generator
Q
Data Out
Write/Read
Enable
Address A
ENA
D
A
qA[ ]
Data In
B
qB[ ]
Q
D
ENA
Data Out
Write/Read
Enable
Address B
Byte Enable B
Memory Block
256 × 16 (2)
512 × 8
1,024 × 4
2,048 × 2
4,096 × 1
Byte Enable A
Data In
Write
Pulse
Generator
Q
Q
Q
Q
ENA
D
ENA
D
ENA
D
ENA
D
8
clockB
clkenB
wrenB
addressB[ ]
byteenaB[ ]
dataB[ ]
TriMatrix Embedded Memory Blocks in Stratix & Stratix GX Devices
Figure 2–10. Input/Output Clock Mode in True Dual-Port Mode Note (1)
Note to Figure 2–10:
Violating the setup or hold time on the address registers could corrupt the memory contents. This applies to both
read and write operations.
2–19
Stratix Device Handbook, Volume 2
Clock Modes
All registers shown have asynchronous clear ports, except when using
the M-RAM. M-RAM blocks have asynchronous clear ports on their
output registers only.
Figure 2–11. Input/Output Clock Mode in Simple Dual-Port Mode Notes (1), (2), (3), (4)
8 LAB Row
Clocks
Memory Block
256 ´ 16
512 ´ 8
1,024 ´ 4
2,048 ´ 2
4,096 ´ 1
8
data[ ]
D
Q
ENA
Data In
address[ ]
D
Q
ENA
Read Address
Data Out
byteena[ ]
D
Q
ENA
Byte Enable
wraddress[ ]
D
Q
ENA
Write Address
D
Q
ENA
Read Enable
D
Q
ENA
To MultiTrack
Interconnect
rden
wren
outclken
inclken
wrclock
D
Q
ENA
Write
Pulse
Generator
Write Enable
rdclock
Notes to Figure 2–11:
(1)
(2)
(3)
(4)
The rden signal is not available in the M-RAM block. A M-RAM block in simple dual-port mode is always reading
out the data stored at the current read address location.
For more information on the MultiTrack™ interconnect, see the Stratix Device Family Data Sheet section of the Stratix
Device Handbook, Volume 1 or the Stratix GX Device Family Data Sheet section of the Stratix GX Device Handbook,
Volume 1.
All registers shown have asynchronous clear ports, except when using the M-RAM. M-RAM blocks have
asynchronous clear ports on their output registers only.
Violating the setup or hold time on the address registers could corrupt the memory contents. This applies to both
read and write operations.
2–20
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
TriMatrix Embedded Memory Blocks in Stratix & Stratix GX Devices
Read/Write Clock Mode
The TriMatrix memory blocks can implement read/write clock mode for
simple dual-port memory. This mode can use up to two clocks. The write
clock controls the block’s data inputs, wraddress, and wren. The read
clock controls the data output, rdaddress, and rden. The memory
blocks support independent clock enables for each clock and
asynchronous clear signals for the read- and write-side registers.
Figure 2–12 shows a memory block in read/write clock mode.
Altera Corporation
July 2005
2–21
Stratix Device Handbook, Volume 2
Clock Modes
Figure 2–12. Read/Write Clock Mode in Simple Dual-Port Mode Notes (1), (2), (3)
8 LAB Row
Clocks
Memory Block
256 × 16
512 × 8
1,024 × 4
Data In
2,048 × 2
4,096 × 1
8
data[ ]
D
Q
ENA
Data Out
address[ ]
D
Q
ENA
Read Address
wraddress[ ]
D
Q
ENA
Write Address
byteena[ ]
D
Q
ENA
Byte Enable
D
Q
ENA
To MultiTrack
Interconnect
rden
D
Q
ENA
wren
Read
Pulse
Generator
Read Enable
rdclocken
wrclocken
wrclock
D
Q
ENA
Write
Pulse
Generator
Write Enable
rdclock
Notes to Figure 2–12:
(1)
(2)
(3)
For more information on the MultiTrack interconnect, see the Stratix Device Family Data Sheet section of the Stratix
Device Handbook, Volume 1 or the Stratix GX Device Family Data Sheet section of the Stratix GX Device Handbook,
Volume 1.
All registers shown have asynchronous clear ports, except when using the M-RAM. M-RAM blocks have
asynchronous clear ports on their output registers only.
Violating the setup or hold time on the address registers could corrupt the memory contents. This applies to both
read and write operations.
2–22
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
TriMatrix Embedded Memory Blocks in Stratix & Stratix GX Devices
Single-Port Mode
The TriMatrix memory blocks can implement single-port clock mode for
single-port memory mode. Single-port mode is used when simultaneous
reads and writes are not required. See Figure 2–13. A single block in a
memory block can support up to two single-port mode RAM blocks in
M4K blocks.
Figure 2–13. Single-Port Mode Notes (1), (2), (3)
8 LAB Row
Clocks
RAM/ROM
256 × 16
512 × 8
1,024 × 4
Data In
2,048 × 2
4,096 × 1
8
data[ ]
D
Q
ENA
Data Out
address[ ]
D
Q
ENA
Address
D
Q
ENA
To MultiTrack
Interconnect
wren
Write Enable
outclken
inclken
inclock
D
Q
ENA
Write
Pulse
Generator
outclock
Notes to Figure 2–13:
(1)
(2)
(3)
For more information on the MultiTrack interconnect, see the Stratix Device Family Data Sheet section of the Stratix
Device Handbook, Volume 1 or the Stratix GX Device Family Data Sheet section of the Stratix GX Device Handbook,
Volume 1.
All registers shown have asynchronous clear ports, except when using the M-RAM. M-RAM blocks have
asynchronous clear ports on their output registers only.
Violating the setup or hold time on the address registers could corrupt the memory contents. This applies to both
read and write operations.
Designing With
TriMatrix
Memory
Altera Corporation
July 2005
When instantiating TriMatrix memory you must understand the various
features that set it apart from other memory architectures. The following
sections describe some of the important attributes and functionality of
TriMatrix memory.
2–23
Stratix Device Handbook, Volume 2
Designing With TriMatrix Memory
f
For information on the difference between APEX-style memory and
TriMatrix memory, see the Transitioning APEX Designs to Stratix Devices
chapter.
Selecting TriMatrix Memory Blocks
The Quartus II software automatically partitions user-defined memory
into embedded memory blocks using the most efficient size
combinations. The memory can also be manually assigned to a specific
block size or a mixture of block sizes. Table 2–1 on page 2–2 is a guide for
selecting a TriMatrix memory block size based on supported features.
1
f
Violating the setup or hold time on the address registers could
corrupt the memory contents. This applies to both read and
write operations.
For more information on selecting which memory block to use, see
AN 207: TriMatrix Memory Selection Using the Quartus II Software.
1
Violating the setup or hold time on the address registers could
corrupt the memory contents. This applies to both read and
write operations.
Pipeline & Flow-Through Modes
TriMatrix memory architecture implements synchronous (pipelined)
RAM by registering both the input and output signals to the RAM block.
All TriMatrix memory inputs are registered providing synchronous write
cycles. In synchronous operation, RAM generates its own self-timed
strobe write enable (wren) signal derived from the global or regional
clock. In contrast, a circuit using asynchronous RAM must generate the
RAM wren signal while ensuring its data and address signals meet setup
and hold time specifications relative to the wren signal. The output
registers can be bypassed.
In an asynchronous memory neither the input nor the output is
registered. While Stratix and Stratix GX devices do not support
asynchronous memory, they do support a flow-through read where the
output data is available during the clock cycle when the read address is
driven into it. Flow-through reading is possible in the simple and true
dual-port modes of the M512 and M4K blocks by clocking the read enable
and read address registers on the negative clock edge and bypassing the
output registers.
f
For more information, see AN 210: Converting Memory from Asynchronous
to Synchronous for Stratix & Stratix GX Devices.
2–24
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
TriMatrix Embedded Memory Blocks in Stratix & Stratix GX Devices
Power-up Conditions & Memory Initialization
Upon power-up, TriMatrix memory is in an idle state. The M512 and M4K
block outputs always power-up to zero, regardless of whether the output
registers are used or bypassed. Even if a memory initialization file is used
to pre-load the contents of the RAM block, the outputs still power-up
cleared. For example, if address 0 is pre-initialized to FF, the M512 and
M4K blocks power-up with the output at 00.
M-RAM blocks do not support memory initialization files; therefore, they
cannot be pre-loaded with data upon power-up. M-RAM blocks
combinatorial outputs and memory controls always power-up to an
unknown state. If M-RAM block outputs are registered, the registers
power-up cleared. The undefined output appears one clock cycle later.
The output remains undefined until a read operation is performed on an
address that has been written to.
Read-DuringWrite Operation
at the Same
Address
The following two sections describe the functionality of the various RAM
configurations when reading from an address during a write operation at
that same address. There are two types of read-during-write operations:
same-port and mixed-port. Figure 2–14 illustrates the difference in data
flow between same-port and mixed-port read-during-write.
Figure 2–14. Read-During-Write Data Flow
Port A
data in
Port B
data in
Mixed-port
data flow
Same-port
data flow
Port A
data out
Port B
data out
Same-Port Read-During-Write Mode
For read-during-write operation of a single-port RAM or the same port of
a true dual-port RAM, the new data is available on the rising edge of the
same clock cycle it was written on. This behavior is valid on all memoryblock sizes. See Figure 2–15 for a sample functional waveform.
Altera Corporation
July 2005
2–25
Stratix Device Handbook, Volume 2
Read-During-Write Operation at the Same Address
When using byte enables in true dual-port RAM mode, the outputs for
the masked bytes on the same port are unknown. (See Figure 2–1 on
page 2–6.) The non-masked bytes are read out as shown in Figure 2–15.
Figure 2–15. Same-Port Read-During-Write Functionality Note (1)
inclock
data_in
A
B
wren
data_out Old
A
Note to Figure 2–15:
(1)
Outputs are not registered.
Mixed-Port Read-During-Write Mode
This mode is used when a RAM in simple or true dual-port mode has one
port reading and the other port writing to the same address location with
the same clock.
The READ_DURING_WRITE_MODE_MIXED_PORTS parameter for M512
and M4K memory blocks determines whether to output the old data at
the address or a “don’t care” value. Setting this parameter to OLD_DATA
outputs the old data at that address. Setting this parameter to DONT_CARE
outputs a “don’t care” or unknown value. See Figures 2–16 and 2–17 for
sample functional waveforms showing this operation. These figures
assume that the outputs are not registered.
The DONT_CARE setting allows memory implementation in any TriMatrix
memory block. The OLD_DATA setting restricts memory implementation
to only M512 or M4K memory blocks. Selecting DONT_CARE gives the
compiler more flexibility when placing memory functions into TriMatrix
memory.
2–26
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
TriMatrix Embedded Memory Blocks in Stratix & Stratix GX Devices
Figure 2–16. Mixed-Port Read-During-Write: OLD_DATA
inclock
addressA and
addressB
Port A
data_in
Address Q
A
B
Port A
wren
Port B
wren
Port B
data_out
Old
A
B
For mixed-port read-during-write operation of the same address location
of a M-RAM block, the RAM outputs are unknown, as shown in
Figure 2–17.
Figure 2–17. Mixed-Port Read-During-Write: DONT_CARE
inclock
addressA and
addressB
Port A
data_in
Address Q
A
B
Port A
wren
Port B
wren
Port B
data_out
Unknown
B
Mixed-port read-during-write is not supported when two different clocks
are used in a dual-port RAM. The output value will be unknown during
a mixed-port read-during-write operation.
Conclusion
Altera Corporation
July 2005
TriMatrix memory, an enhanced RAM architecture with extremely high
memory bandwidth in Stratix and Stratix GX devices, gives advanced
control of memory applications with features such as byte enables, parity
bit storage, and shift-register mode, as well as mixed-port width support
and true dual-port mode.
2–27
Stratix Device Handbook, Volume 2
Conclusion
2–28
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
3. External Memory
Interfaces in Stratix &
Stratix GX Devices
S52008-3.3
Introduction
Stratix® and Stratix GX devices support a broad range of external
memory interfaces such as double data rate (DDR) SDRAM, RLDRAM II,
quad data rate (QDR) SRAM, QDRII SRAM, zero bus turnaround (ZBT)
SRAM, and single data rate (SDR) SDRAM. The dedicated phase-shift
circuitry allows the Stratix and Stratix GX devices to interface at twice the
system clock speed with an external memory (up to 200 MHz/400 Mbps).
Typical I/O architectures transmit a single data word on each positive
clock edge and are limited to the associated clock speed using this
protocol. To achieve a 400-megabits per second (Mbps) transfer rate, a
SDR system requires a 400-MHz clock. Many new applications have
introduced a DDR I/O architecture as an alternative to SDR architectures.
While SDR architectures capture data on one edge of a clock, the DDR
architectures captures data on both the rising and falling edges of the
clock, doubling the throughput for a given clock frequency and
accelerating performance. For example, a 200-MHz clock can capture a
400-Mbps data stream, enhancing system performance and simplifying
board design.
Most current memory architectures use a DDR I/O interface. These DDR
memory standards cover a broad range of applications for embedded
processor systems, image processing, storage, communications, and
networking. This chapter describes the hardware features in Stratix and
Stratix GX devices that facilitate the high-speed memory interfacing for
each memory standard. It then briefly explains how each memory
standard uses the features of the Stratix and Stratix GX devices.
f
External
Memory
Standards
You can use this document with AN 329: ZBT SRAM Controller Reference
Design for Stratix & Stratix GX Devices, AN 342: Interfacing DDR SDRAM
with Stratix & Stratix GX Devices, and AN 349: QDR SRAM Controller
Reference Design for Stratix & Stratix GX Devices.
The following sections provide an overview on using the Stratix and
Stratix GX device external memory interfacing features.
DDR SDRAM
DDR SDRAM is a memory architecture that transmits and receives data
at twice the clock speed of traditional SDR architectures. These devices
transfer data on both the rising and falling edge of the clock signal.
Altera Corporation
June 2006
3–1
External Memory Standards
Interface Pins
DDR devices use interface pins including data, data strobe, clock,
command, and address pins. Data is sent and captured at twice the clock
rate by transferring data on both the positive and negative edge of a clock.
The commands and addresses only use one active edge of a clock.
Connect the memory device’s DQ and DQS pins to the DQ and DQS pins,
respectively, as listed in the Stratix and Stratix GX devices pin table. DDR
SDRAM also uses active-high data mask pins for writes. You can connect
DM pins to any of the I/O pins in the same bank as the DQ pins of the
FPGA. There is one DM pin per DQS/DQ group.
DDR SDRAM ×16 devices use two DQS pins, and each DQS pin is
associated with eight DQ pins. However, this is not the same as the
×16 mode in Stratix and Stratix GX devices. To support a ×16 DDR
SDRAM, you need to configure the Stratix and Stratix GX FPGAs to use
two sets of DQ pins in ×8 mode. Similarly if your ×32 memory device uses
four DQS pins where each DQS pin is associated with eight DQ pins, you
need to configure the Stratix and Stratix GX FPGA to use four sets of pins
in ×8 mode.
You can also use any I/O pins in banks 1, 2, 5, or 6 to interface with
DDR SDRAM devices. These banks do not have dedicated circuitry,
though.
You can also use any of the user I/O pins for commands and addresses to
the DDR SDRAM.
f
For more information, see AN 342: Interfacing DDR SDRAM with Stratix
& Stratix GX Devices.
If the DDR SDRAM device supports ECC, the design uses a DQS/DQ
group for ECC pins. You can use any of the user I/O pins for commands
and addresses.
Because of the symmetrical setup and hold time for the command and
address pins at the memory, you might need to generate these signals
from the system clock’s negative edge.
The clocks to the SDRAM device are called CK and CK#. Use any of the
user I/O pins via the DDR registers to generate the CK and CK# signals
to meet the DDR SDRAM tDQSS requirement. The memory device’s tDQSS
requires that the DQS signal’s positive edge write operations must be
within 25% of the positive edge of the DDR SDRAM clock input. Using
user I/O pins for CK and CK# ensures that any PVT variations seen by
the DQS signal are tracked by these pins, too.
3–2
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
External Memory Interfaces in Stratix & Stratix GX Devices
Read & Write Operations
When reading from the DDR SDRAM, the DQS signal coming into the
Stratix and Stratix GX device is edge-aligned with the DQ pins. The
dedicated circuitry center-aligns the DQS signal with respect to the DQ
signals and the shifted DQS bus drives the clock input of the DDR input
registers. The DDR input registers bring the data from the DQ signals to
the device. The system clock clocks the DQS output enable and output
paths. The -90° shifted clock clocks the DQ output enable and output
paths. Figure 3–1 shows an example of the DQ and DQS relationship
during a burst-of-two read. It shows where the DQS signal is
center-aligned in the IOE.
Figure 3–1. Example of Where a DQS Signal is Center-Aligned in the IOE
Pin to register
delay
DQS at
FPGA Pin
Postamble
Preamble
DQ at
FPGA Pin
DQS at DQ
IOE registers
DQ at DQ
IOE registers
90 degree shift
Pin to register
delay
When writing to the DDR SDRAM, the DQS signal must be centeraligned with the DQ pins. Two PLL outputs are needed to generate the
DQS signal and to clock the DQ pins. The DQS are clocked by the 0°
phase-shift PLL output, while the DQ pins are clocked by the -90° phaseshifted PLL output. Figure 3–2 shows the DQS and DQ relationship
during a DDR SDRAM burst-of-two write.
Figure 3–2. DQ & DQS Relationship During a Burst-of-Two Write
DQS at
FPGA Pin
DQ at
FPGA Pin
Altera Corporation
June 2006
3–3
Stratix Device Handbook, Volume 2
External Memory Standards
Figure 3–3 shows DDR SDRAM interfacing from the I/O through the
dedicated circuitry to the logic array. When the DQS pin acts as an input
strobe, the dedicated circuitry shifts the incoming DQS pin by either 72°
or 90° and clocks the DDR input registers. Because of the DDR input
registers architecture in Stratix and Stratix GX devices, the shifted DQS
signal must be inverted. The DDR registers outputs are sent to two LE
registers to be synchronized with the system clock.
f
Refer to the DC & Switching Characteristics chapter in volume 1 of the
Stratix Device Handbook for frequency limits regarding the 72 and 90°
phase shift for DQS.
Figure 3–3. DDR SDRAM Interfacing
DQ
DQS
Compensated
Delay Shift
OE
PLL
DDR
OE
Registers
User logic/ 2
GND
2
DDR
Output
Registers
OE
Δt
DDR
OE
Registers
2
DDR
Output
Registers
DDR
Input
Registers
I/O Elements &
Periphery
− 90˚
DQS Bus
LE
Register
LE
Register
Resynchronizing
Global Clock
f
Adjacent LAB LEs
For more information on DDR SDRAM specifications, see JEDEC
standard publications JESD79C from www.jedec.org, or see
AN 342: Interfacing DDR SDRAM with Stratix & Stratix GX Devices.
RLDRAM II
RLDRAM II provides fast random access as well as high bandwidth and
high density, making this memory technology ideal for high-speed
network and communication data storage applications. The fast random
access speeds in RLDRAM II devices make them a viable alternative to
SRAM devices at a lower cost. Additionally, RLDRAM II devices have
minimal latency to support designs that require fast response times.
3–4
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
External Memory Interfaces in Stratix & Stratix GX Devices
Interface Pins
RLDRAM II devices use interface pins such as data, clock, command, and
address pins. There are two types of RLDRAM II memory: common I/O
(CIO) and separate I/O (SIO). The data pins in RLDRAM II CIO device
are bidirectional while the data pins in a RLDRAM II SIO device are
uni-directional. Instead of bidirectional data strobes, RLDRAM II uses
differential free-running read and write clocks to accompany the data. As
in DDR SDRAM, data is sent and captured at twice the clock rate by
transferring data on both the positive and negative edge of a clock. The
commands and addresses still only use one active edge of a clock.
If the data pins are bidirectional, connect them to the Stratix and
Stratix GX device DQ pins. If the data pins are uni-directional, connect
the RLDRAM II device Q ports to the Stratix and Stratix GX device DQ
pins and connect the D ports to any user I/O pins in I/O banks 3, 4, 7, and
8. RLDRAM II also uses active-high data mask pins for writes. You can
connect DM pins to any of the I/O pins in the same bank as the DQ pins
of the FPGA. When interfacing with SIO devices, connect the DM pins to
any of the I/O pins in the same bank as the D pins. There is one DM pin
per DQS/DQ group.
Connect the read clock pins (QK) to Stratix and Stratix GX device DQS
pins. You must configure the DQS signals as bidirectional pins. However,
since QK pins are output-only pins from the memory, RLDRAM memory
interfacing in Stratix and Stratix GX devices requires that you ground the
DQS and DQSn pin output enables. The Stratix and Stratix GX devices
use the shifted QK signal from the DQS logic block to capture data. You
can leave the QK# signal of the RLDRAM II device unconnected.
RLDRAM II devices have both input clocks (CK and CK#) and write
clocks (DK and DK#). Use the external clock buffer to generate CK, CK#,
DK, and DK# to meet the CK, CK#, DK, and DK# skew requirements from
the RLDRAM II device. If you are interfacing with multiple RLDRAM II
devices, perform IBIS simulations to analyze the loading effects on the
clock pair.
You can use any of the user I/O pins for commands and addresses.
RLDRAM II also offers QVLD pins to indicate the read data availability.
Connect the QVLD pins to the Stratix and Stratix GX device DQVLD pins,
listed in the pin table.
Read & Write Operations
When reading from the RLDRAM II device, data is sent edge-aligned
with the read clock QK or QK# signal. When writing to the RLDRAM II
device, data must be center-aligned with the write clock (DK or DK#
signal). The Stratix and Stratix GX device RLDRAM II interface uses the
Altera Corporation
June 2006
3–5
Stratix Device Handbook, Volume 2
External Memory Standards
same scheme as in DDR SDRAM interfaces whereby the dedicated
circuitry is used during reads to center-align the data and the read clock
inside the FPGA and the PLL center-aligns the data and write clock
outputs. The data and clock relationship for reads and writes in
RLDRAM II is similar to those in DDR SDRAM as already depicted in
Figure 3–1 on page 3–3 and Figure 3–3 on page 3–4.
QDR & QDRII SRAM
QDR SRAM provides independent read and write ports that eliminate
the need for bus turnaround. The memory uses two sets of clocks: K and
Kn for write access, and optional C and Cn for read accesses, where Kn
and Cn are the inverse of the K and C clocks, respectively. You can use
differential HSTL I/O pins to drive the QDR SRAM clock into the Stratix
and Stratix GX devices. The separate write data and read data ports
permit a transfer rate up to four words on every cycle through the DDR
circuitry. Stratix and Stratix GX devices support both burst-of-two and
burst-of-four QDR SRAM architectures, with clock cycles up to 167 MHz
using the 1.5-V HSTL Class I or Class II I/O standard. Figure 3–4 shows
the block diagram for QDR SRAM burst-of-two architecture.
Figure 3–4. QDR SRAM Block Diagram for Burst-of-Two Architecture
Discrete QDR SRAM Device
A
18
BWSn
Write
Port
WPSn
D
36
Data
256K × 18
Memory
Array
256K × 18
Memory 36
Array
Data
Read
Port
RPSn
2
C, Cn
18
Q
18
2
K, Kn
VREF
Control
Logic
QDRII SRAM is a second generation of QDR SRAM devices. It can
transfer four words per clock cycle, fulfilling the requirements facing
next-generation communications system designers. QDRII SRAM
devices provide concurrent reads and writes, zero latency, and increased
data throughput. Stratix and Stratix GX devices support QDRII SRAM at
speeds up to 200 MHz since the timing requirements for QDRII SRAM
are not as strict as QDR SRAM.
3–6
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
External Memory Interfaces in Stratix & Stratix GX Devices
Interface Pins
QDR and QDRII SRAM uses two separate, uni-directional data ports for
read and write operations, enabling quad data-rate data transfer. Both
QDR and QDRII SRAM use shared address lines for reads and writes.
Stratix and Stratix GX devices utilize dedicated DDR I/O circuitry for the
input and output data bus and the K and Kn output clock signals.
Both QDR and QDRII SRAM burst-of-two devices sample the read
address on the rising edge of the K clock and sample the write address on
the rising edge of the Kn clock while QDR and QDRII SRAM burst-offour devices sample both read and write addresses on the K clock's rising
edge. You can use any of the Stratix and Stratix GX device user I/O pins
in I/O banks 3, 4, 7, and 8 for the D write data ports, commands, and
addresses.
QDR SRAM uses the following clock signals: input clocks K and Kn and
output clocks C and Cn. In addition to the aforementioned two pairs of
clocks, QDRII SRAM also uses echo clocks CQ and CQn. Clocks Cn, Kn,
and CQn are logical complements of clocks C, K, and CQ respectively.
Clocks C, Cn, K, and Kn are inputs to the QDRII SRAM while clocks CQ
and CQn are outputs from the QDRII SRAM. Stratix and Stratix GX
devices use single-clock mode for single-device QDR and QDRII SRAM
interfacing where the K and Kn are used for both read and write
operations, and the C and Cn clocks are unused. Use both C or Cn and K
or Kn clocks when interfacing with a bank of multiple QDRII SRAM
devices with a single controller.
You can generate C, Cn, K, and Kn clocks using any of the I/O registers
in I/O banks 3, 4, 7, or 8 via the DDR registers. Due to strict skew
requirements between K and Kn signals, use adjacent pins to generate the
clock pair. Surround the pair with buffer pins tied to VCC and ground for
better noise immunity from other signals.
In general, all output signals to the QDR and QDRII SRAM should use the
top and bottom banks (I/O banks 3, 4, 7, or 8). You can place the input
signals from the QDR and QDRII SRAM in any I/O banks.
Read & Write Operations
Figure 3–5 shows the data and clock relationships in QDRII SRAM
devices at the memory pins during reads. QDR and QDRII SRAM devices
send data within a tCO time after each rising edge of the input clock C or
Cn in multi-clock mode, or the input clock K or Kn in single clock mode.
Data is valid until tDOH time, after each rising edge of the C or Cn in multi-
Altera Corporation
June 2006
3–7
Stratix Device Handbook, Volume 2
External Memory Standards
clock mode, or K or Kn in single clock mode. The edge-aligned CQ and
CQn clocks accompany the read data for data capture in Stratix and
Stratix GX devices.
Figure 3–5. Data & Clock Relationship During a QDRII SRAM Read
Note (1)
C/K
Cn/Kn
tCO (2)
Q
tCO (2)
QA
tCLZ (3)
QA + 1
tDOH (2)
QA + 2
QA + 3
tCHZ (3)
CQ
tCQD (4)
CQn
tCCQO (5)
tCQOH (5)
tCQD (4)
Notes to Figure 3–5:
(1)
(2)
(3)
(4)
(5)
The timing parameter nomenclature is based on the Cypress QDRII SRAM data sheet for CY7C1313V18.
CO is the data clock-to-out time and tDOH is the data output hold time between burst.
tCLZ and tCHZ are bus turn-on and turn-off times respectively.
tCQD is the skew between CQn and data edges.
tCQQO and tCQOH are skew between the C or Cn (or K or Kn in single-clock mode) and the CQ or CQn clocks.
When writing to QDRII SRAM devices, data is generated by the write
clock, while the K clock is 90° shifted from the write clock, creating a
center-aligned arrangement.
f
Go to www.qdrsram.com for the QDR SRAM and QDRII SRAM
specifications. For more information on QDR and QDRII SRAM
interfaces in Stratix and Stratix GX devices, see AN 349: QDR SRAM
Controller Reference Design for Stratix & Stratix GX Devices.
ZBT SRAM
ZBT SRAM eliminate dead bus cycles when turning a bidirectional bus
around between reads and writes or between writes and reads. ZBT
allows for 100% bus utilization because ZBT SRAM can be read or written
on every clock cycle. Bus contention can occur when shifting from a write
cycle to a read cycle or vice versa with no idle cycles in between.
ZBT SRAM allows small amounts of bus contention. To avoid bus
contention, the output clock-to-low-impedance time (tZX) must be greater
3–8
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
External Memory Interfaces in Stratix & Stratix GX Devices
than the clock-to-high-impedance time (tXZ). Stratix and Stratix GX device
I/O pins can interface with ZBT SRAM devices at up to 200 MHz and can
meet ZBT tCO and tSU timing requirements by controlling phase delay in
clocks to the OE or output and input registers using an enhanced PLL.
Figure 3–6 shows a flow-through ZBT SRAM operation where A1 and A3
are read addresses and A2 and A4 are write addresses. For pipelined
ZBT SRAM operation, data is delayed by another clock cycle. Stratix and
Stratix GX devices support up to 200-MHz ZBT SRAM operation using
the 2.5-V or 3.3-V LVTTL I/O standard.
Figure 3–6. tZX & tXZ Timing Diagram
tZX
clock
addr
A1
A2
A3
A4
tXZ
dataout
Q(A1)
Q(A3)
ZBT Bus Sharing
Device tZX
datain
D(A3)
wren
Interface Pins
ZBT SRAM uses one system clock input for all clocking purposes. Only
the rising edge of this clock is used, since ZBT SRAM uses a single data
rate scheme. The data bus, DQ, is bidirectional. There are three control
signals to the ZBT SRAM: RW_N, BW_N, and ADV_LD_N. You can use any
of the Stratix and Stratix GX device user I/O pins to interface to the
ZBT SRAM device.
f
Altera Corporation
June 2006
For more information on ZBT SRAM Interfaces in Stratix devices, see
AN 329: ZBT SRAM Controller Reference Design for Stratix & Stratix GX
Devices.
3–9
Stratix Device Handbook, Volume 2
DDR Memory Support Overview
DDR Memory
Support
Overview
Table 3–1 shows the external RAM support in Stratix EP1S10 through
EP1S40 devices and all Stratix GX devices. Table 3–2 shows the external
RAM support in Stratix EP1S60 and EP1S80 devices.
Table 3–1. External RAM Support in Stratix EP1S10 through EP1S40 & All Stratix GX Devices
Maximum Clock Rate (MHz)
DDR Memory Type
I/O
Standard
-5 Speed
Grade
-6 Speed Grade
Flip-Chip Flip-Chip
-7 Speed Grade
-8 Speed Grade
WireBond
FlipChip
WireBond
FlipChip
WireBond
DDR SDRAM (1),
(2)
SSTL-2
200
167
133
133
100
100
100
DDR SDRAM - side
banks (2), (3), (4)
SSTL-2
150
133
110
133
100
100
100
RLDRAM II (4)
1.8-V HSTL
200
(5)
(5)
(5)
(5)
(5)
(5)
QDR SRAM (6)
1.5-V HSTL
167
167
133
133
100
100
100
QDRII SRAM (6)
1.5-V HSTL
200
167
133
133
100
100
100
ZBT SRAM (7)
LVTTL
200
200
200
167
167
133
133
Notes to Table 3–1:
(1)
(2)
(3)
(4)
(5)
(6)
(7)
These maximum clock rates apply if the Stratix device uses DQS phase-shift circuitry to interface with DDR
SDRAM. DQS phase-shift circuitry is only available on the top and bottom I/O banks (I/O banks 3, 4, 7, and 8).
For more information on DDR SDRAM, see AN 342: Interfacing DDR SDRAM with Stratix & Stratix GX Devices.
DDR SDRAM is supported on the Stratix device side I/O banks (I/O banks 1, 2, 5, and 6) without dedicated DQS
phase-shift circuitry. The read DQS signal is ignored in this mode.
These performance specifications are preliminary.
This device does not support RLDRAM II.
For more information on QDR or QDRII SRAM, see AN 349: QDR SRAM Controller Reference Design for Stratix &
Stratix GX Devices.
For more information on ZBT SRAM, see AN 329: ZBT SRAM Controller Reference Design for Stratix and Stratix GX
Devices.
Table 3–2. External RAM Support in Stratix EP1S60 & EP1S80
(Part 1 of 2)
Maximum Clock Rate (MHz)
DDR Memory Type
I/O Standard
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
SSTL-2
167
167
133
DDR SDRAM - side banks (2), (3) SSTL-2
150
133
133
QDR SRAM (4)
1.5-V HSTL
133
133
133
QDRII SRAM (4)
1.5-V HSTL
167
167
133
DDR SDRAM (1), (2)
3–10
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
External Memory Interfaces in Stratix & Stratix GX Devices
Table 3–2. External RAM Support in Stratix EP1S60 & EP1S80
(Part 2 of 2)
Maximum Clock Rate (MHz)
DDR Memory Type
ZBT SRAM (5)
I/O Standard
LVTTL
-5 Speed Grade
-6 Speed Grade
-7 Speed Grade
200
200
167
Notes to Table 3–2:
(1)
(2)
(3)
(4)
(5)
These maximum clock rates apply if the Stratix device uses DQS phase-shift circuitry to interface with DDR
SDRAM. DQS phase-shift circuitry is only available on the top and bottom I/O banks (I/O banks 3, 4, 7, and 8).
For more information on DDR SDRAM, see AN 342: Interfacing DDR SDRAM with Stratix & Stratix GX Devices.
DDR SDRAM is supported on the side banks (I/O banks 1, 2, 5, and 6) with no dedicated DQS phase-shift circuitry.
The read DQS signal is ignored in this mode.
For more information on QDR or QDRII SRAM, see AN 349: QDR SRAM Controller Reference Design for Stratix &
Stratix GX Devices.
For more information on ZBT SRAM, see AN 329: ZBT SRAM Controller Reference Design for Stratix and Stratix GX
Devices.
Stratix and Stratix GX devices support the data strobe or read clock signal
(DQS) used in DDR SDRAM, and RLDRAM II devices. DQS signals are
associated with a group of data (DQ) pins.
Stratix and Stratix GX devices contain dedicated circuitry to shift the
incoming DQS signals by 0°, 72°, and 90°. The DQS phase-shift circuitry
uses a frequency reference to dynamically generate control signals for the
delay chains in each of the DQS pins, allowing it to compensate for
process, voltage, and temperature (PVT) variations. The dedicated
circuitry also creates consistent margins that meet your data sampling
window requirements.
f
Refer to the DC & Switching Characteristics chapter in volume 1 of the
Stratix Device Handbook for frequency limits regarding the 72 and 90°
phase shift for DQS.
In addition to the DQS dedicated phase-shift circuitry, every I/O element
(IOE) in Stratix and Stratix GX devices contains six registers and one latch
to achieve DDR operation. There is also a programmable delay chain in
the IOE that can help reduce contention when interfacing with ZBT
SRAM devices.
DDR Memory Interface Pins
Stratix and Stratix GX devices use data (DQ), data strobe (DQS), and clock
pins to interface with DDR SDRAM and RLDRAM II devices. This section
explains the pins used in the DDR SDRAM and RLDRAM II interfaces.
For QDR, QDRII, and ZBT SRAM interfaces, see the “External Memory
Standards” section.
Altera Corporation
June 2006
3–11
Stratix Device Handbook, Volume 2
DDR Memory Support Overview
Figure 3–7 shows the DQ and DQS pins in ×8 mode.
Figure 3–7. Stratix & Stratix GX Device DQ & DQS Groups in × 8 Mode
Top or Bottom I/O Bank
DQ Pins (1)
DQS Pin
Note to Figure 3–7:
(1)
There are at least eight DQ pins per group.
Data & Data Strobe Pins
Stratix and Stratix GX data pins for the DDR memory interfaces are called
DQ pins. The Stratix and Stratix GX device I/O banks at the top (I/O
banks 3 and 4) and the bottom (I/O banks 7 and 8) of the device support
DDR SDRAM and RLDRAM II up to 200 MHz. These pins support DQS
signals with DQ bus modes of ×8, ×16, or ×32. Stratix and Stratix GX
devices can support either bidirectional data strobes or uni-directional
read clocks. Depending on the external memory interface, either the
memory device's read data strobes or read clocks feed the DQS pins.
For ×8 mode, there are up to 20 groups of programmable DQS and DQ
pins—10 groups in I/O banks 3 and 4 and 10 groups in I/O banks 7 and 8
(see Table 3–3). Each group consists of one DQS pin and a set of eight DQ
pins.
For ×16 mode, there are up to eight groups of programmable DQS and
DQ pins—four groups in I/O banks 3 and 4, and four groups in I/O
banks 7 and 8. The EP1S20 device supports seven ×16 mode groups. The
EP1S10 device does not support ×16 mode. All other devices support the
full eight groups. See Table 3–3. Each group consists of one DQS and 16
DQ pins. In ×16 mode, DQS1T, DQS3T, DQS6T, and DQS8T pins on the top
side of the device, and DQS1B, DQS3B, DQS6B, and DQS8B pins on the
3–12
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
External Memory Interfaces in Stratix & Stratix GX Devices
bottom side of the device are dedicated DQS pins. The DQS2T, DQS7T,
DQS2B, and DQS7B pins are dedicated DQS pins for ×32 mode, and each
group consists of one DQS and 32 DQ pins.
Table 3–3. DQS & DQ Bus Mode Support
Device
Package
Note (1)
Number of ×8 Number of ×16
Groups
Groups
Number of ×32
Groups
672-pin BGA
672-pin FineLine BGA®
12 (2)
0
0
484-pin FineLine BGA
780-pin FineLine BGA
16 (3)
0
4
484-pin FineLine BGA
18 (4)
7 (5)
4
672-pin BGA
672-pin FineLine BGA
16 (3)
7 (5)
4
780-pin FineLine BGA
20
7 (5)
4
672-pin BGA
672-pin FineLine BGA
16 (3)
8
4
780-pin FineLine BGA
1,020-pin FineLine BGA
20
8
4
EP1S30
956-pin BGA
780-pin FineLine BGA
1,020-pin FineLine BGA
20
8
4
EP1S40
956-pin BGA
1,020-pin FineLine BGA
1,508-pin FineLine BGA
20
8
4
EP1S60
956-pin BGA
1,020-pin FineLine BGA
1,508-pin FineLine BGA
20
8
4
EP1S80
956-pin BGA
1,508-pin FineLine BGA
1,923-pin FineLine BGA
20
8
4
EP1S10
EP1S20
EP1S25
Notes to Table 3–3:
(1)
(2)
(3)
(4)
(5)
For VREF guidelines, see the Selectable I/O Standards in Stratix & Stratix GX Devices chapter of the Stratix Device
Handbook, Volume 2 or the Stratix GX Handbook, Volume 2.
These packages have six groups in I/O banks 3 and 4 and six groups in I/O banks 7 and 8.
These packages have eight groups in I/O banks 3 and 4 and eight groups in I/O banks 7 and 8.
This package has nine groups in I/O banks 3 and 4 and nine groups in I/O banks 7 and 8.
These packages have three groups in I/O banks 3 and 4 and four groups in I/O banks 7 and 8.
Altera Corporation
June 2006
3–13
Stratix Device Handbook, Volume 2
DDR Memory Support Overview
The DQS pins are marked in the Stratix and Stratix GX device pin table as
DQS[9..0]T or DQS[9..0]B, where T stands for top and B for bottom.
The corresponding DQ pins are marked as DQ[9..0]T[7..0], where
[9..0] indicates which DQS group the pins belong to. The numbering
scheme starts from right to left on the package bottom view. When not
used as DQ or DQS pins, these pins are available as user I/O pins.
You can also create a design in a mode other than the ×8, ×16, or ×32
mode. The Quartus® II software uses the next larger mode with the
unused DQ pins available as regular use I/O pins. For example, if you
create a design for ×9 mode for an RLDRAM II interface (nine DQ pins
driven by one DQS pin), the Quartus II software implements a ×16 mode
with seven DQ pins unconnected to the DQS bus. These seven unused
DQ pins can be used as regular I/O pins.
1
On the top and bottom side of the device, the DQ and DQS pins
must be configured as bidirectional DDR pins to enable the DQS
phase-shift circuitry. If you only want to use the DQ and/or
DQS pins as inputs, you need to set the output enable of the DQ
and/or DQS pins to ground. Use the altdqs and altdq
megafunctions to configure the DQS and DQ pins, respectively.
However, you should use the Altera® IP Toolbench to create the
data path for your memory interfaces.
Stratix and Stratix GX device side I/O banks (I/O banks 1, 2, 5, and 6)
support SDR SDRAM, ZBT SRAM, QDR SRAM, QDRII SRAM, and DDR
SDRAM interfaces and can use any of the user I/O pins in these banks for
the interface. Since these I/O banks do not have any dedicated circuitry
for memory interfacing, they can support DDR SDRAM up to 150 MHz
in -5 speed grade devices. However, these I/O banks do not support the
HSTL-18 Class II I/O standard, which is required to interface with
RLDRAM II.
Clock Pins
You can use any of the DDR I/O registers in the top or bottom bank of the
device (I/O banks 3, 4, 7, or 8) to generate clocks to the memory device.
You can also use any of the DDR I/O registers in the side I/O banks 1, 2,
5, or 6 to generate clocks for DDR SDRAM interfaces on the side I/O
banks (not using the DQS circuitry).
3–14
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
External Memory Interfaces in Stratix & Stratix GX Devices
Command & Address Pins
You can use any of the user I/O pins in the top or bottom bank of the
device (I/O banks 3, 4, 7, or 8) for commands and addresses. For DDR
SDRAM, you can also use any of the user I/O pins in the side I/O banks
1, 2, 5, or 6, regardless of whether you use the DQS phase-shift circuitry
or not.
Other Pins (Parity, DM, ECC & QVLD Pins)
You can use any of the DQ pins for the parity pins in Stratix and
Stratix GX devices. However, this may mean that you are using the next
larger DQS/DQ mode. For example, if you need a parity bit for each byte
of data, you are actually going to have nine DQ pins per DQS pin. The
Quartus II software then implements a ×16 mode, with the seven unused
DQ pins available as user I/O pins.
The data mask (DM) pins are only required when writing to
DDR SDRAM and RLDRAM II devices. A low signal on the DM pins
indicates that the write is valid. If the DM signal is high, the memory
masks the DQ signals. You can use any of the I/O pins in the same bank
as the DQ pins for the DM signals. Each group of DQS and DQ signals
requires a DM pin. The DDR register, clocked by the –90° shifted clock,
creates the DM signals, similar to DQ output signals.
Some DDR SDRAM devices support error correction coding (ECC),
which is a method of detecting and automatically correcting errors in
data transmission. Connect the DDR ECC pins to a Stratix and Stratix GX
device DQS/DQ group. In 72-bit DDR SDRAM, there are eight ECC pins
in addition to the 64 data pins. The memory controller needs extra logic
to encode and decode the ECC data.
QVLD pins are used in RLDRAM II interfacing to indicate the read data
availability. There is one QVLD pin per RLDRAM II device. A high on
QVLD indicates that the memory is outputting the data requested.
Similar to DQ inputs, this signal is edge-aligned with the RLDRAM II
read clocks, QK and QK#, and is sent half a clock cycle before data starts
coming out of the memory. You can connect QVLD pins to any of the I/O
pins in the same bank as the DQ pins for the QVLD signals.
DQS Phase-Shift Circuitry
Two single phase-shifting reference circuits are located on the top and
bottom of the Stratix and Stratix GX devices. Each circuit is driven by a
system reference clock that is of the same frequency as the DQS signal.
Clock pins CLK[15..12]p feed the phase-shift circuitry on the top of the
device and clock pins CLK[7..4]p feed the phase-shift circuitry on the
Altera Corporation
June 2006
3–15
Stratix Device Handbook, Volume 2
DDR Memory Support Overview
bottom of the device. The phase-shift circuitry cannot be fed from other
sources such as the LE or the PLL internal output clocks. This phase-shift
circuitry is used for DDR SDRAM and RLDRAM II interfaces. For best
performance, turn off the input reference clock to the DQS phase-shift
circuitry when reading from the DDR SDRAM or RLDRAM II. This is to
avoid any DLL jitter incorrectly shifting the DQS signal while the FPGA
is capturing data.
1
The I/O pins in I/O banks 1, 2, 5, and 6 can interface with the
DDR SDRAM at up to 150 MHz. See AN 342: Interfacing DDR
SDRAM with Stratix & Stratix GX Devices.
A compensated delay element on each DQS pin allows for either a 90° or
a 72° phase shift, which automatically centers input DQS signals with the
data valid window of their corresponding DQ data signals. The DQS
signals drive a local DQS bus within the top and bottom I/O banks. This
DQS bus is an additional resource to the I/O clocks and clocks DQ input
registers with the DQS signal.
f
Refer to the DC & Switching Characteristics chapter in volume 1 of the
Stratix Device Handbook for frequency limits regarding the 72 and 90°
phase shift for DQS.
The phase-shifting reference circuit on the top of the device controls the
compensated delay elements for all 10 DQS pins located at the top of the
device. The phase-shifting reference circuit on the bottom of the device
controls the compensated delay elements for all 10 DQS pins located on
the bottom of the device. All 10 delay elements (DQS signals) on either the
top or bottom of the device shift by the same degree amount. For
example, all 10 DQS pins on the top of the device can be shifted by 90° and
all 10 DQS pins on the bottom of the device can be shifted by 72°. The
reference circuit requires a maximum of 256 system reference clock cycles
to set the correct phase on the DQS delay elements.
1
This applies only to the initial phase calculation. Altera
recommends that you enable the DLL during the refresh cycle of
the DDR SDRAM. Enabling the DLL for the duration of the
minimum refresh time is sufficient for recalculating the phase
shift.
Figure 3–8 shows the phase-shift reference circuit control of each DQS
delay shift on the top of the device. This same circuit is duplicated on the
bottom of the device.
3–16
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
External Memory Interfaces in Stratix & Stratix GX Devices
Figure 3–8. DQS & DQSn Pins & the DQS Phase-Shift Circuitry
Note (1)
CLK[15..12] (2)
DQS
Pin
DQS
Pin
DQS
Pin
DQS
Pin
DQS
Pin
DQS
Pin
DQS
Pin
DQS
Pin
DQS
Pin
DQS
Pin
Compensated
Delay Element
Δt
Δt
Δt
Δt
Phase Shift
Reference
Circuit
Δt
Δt
Δt
Δt
Δt
Δt
DQS Bus
Notes to Figure 3–8:
(1)
(2)
There are up to 10 DQS and DQSn pins available on the top or the bottom of the Stratix and Stratix GX devices.
Clock pins CLK[15..12]p feed the phase-shift circuitry on the top of the device and clock pins CLK[7..4]p feed
the phase circuitry on the bottom of the device. The reference clock can also be used in the logic array.
The phase-shift circuitry is only used during read transactions where the
DQS pins are acting as input clocks or strobes. The phase-shift circuitry
can shift the incoming DQS signal by 0°, 72°, and 90°. The shifted DQS
signal is then inverted and used as a clock or a strobe at the DQ IOE input
registers.
f
Refer to the DC & Switching Characteristics chapter in volume 1 of the
Stratix Device Handbook for frequency limits regarding the 72 and 90°
phase shift for DQS.
The DQS phase-shift circuitry is bypassed when 0° shift is chosen. The
routing delay between the pins and the IOE registers is matched with
high precision for both the DQ and DQS signal when the 72° or 90° phase
shift is used. With the 0° phase shift, the skew between DQ and the DQS
signals at the IOE register has been minimized. See Table 3–4 for the
Quartus II software reported number on the DQ and DQS path to the IOE
when the DQS is set to 0° phase shift.
Table 3–4. Quartus II Reported Number on the DQS Path to the
IOE Note (1)
Altera Corporation
June 2006
Speed Grade
DQ2IOE
DQS2IOE
Unit
-5
0.908
1.008
ns
-6
0.956
1.061
ns
-7
1.098
1.281
ns
3–17
Stratix Device Handbook, Volume 2
DDR Memory Support Overview
Table 3–4. Quartus II Reported Number on the DQS Path to the
IOE Note (1)
Speed Grade
DQ2IOE
DQS2IOE
Unit
-8
1.293
1.635
ns
Note to Table 3–4:
(1)
These are reported by Quartus II version 4.0. Check the latest version of the
Quartus II software for the most current information.
To generate the correct phase shift, you must provide a clock signal of the
same frequency as the DQS signal to the DQS phase-shift circuitry. Any
of the CLK[15..12]p clock pins can feed the phase circuitry on the top
of the device (I/O banks 3 and 4) and any of the CLK[7..4]p clock pins
can feed the phase circuitry on the bottom of the device (I/O banks 7
and 8). Both the top and bottom phase-shift circuits need unique clock
pins for the reference clock. You cannot use any internal clock sources to
feed the phase-shift circuitry, but you can route internal clock sources
off-chip and then back into one of the allowable clock input pins.
DLL
The DQS phase-shift circuitry uses a DLL to dynamically measure the
clock period needed by the DQS pin (see Figure 3–9). The DQS
phase-shift circuitry then uses the clock period to generate the correct
phase shift. The DLL in the Stratix and Stratix GX devices DQS phaseshift circuitry can operate between 100 and 200 MHz. The phase-shift
circuitry needs a maximum of 256 clock cycles to calculate the correct
phase shift. Data sent during these clock cycles may not be properly
captured.
1
3–18
Stratix Device Handbook, Volume 2
You can still use the DQS phase-shift circuitry for DDR SDRAM
interfaces that are less than 100 MHz. The DQS signal is shifted
by about 2.5 ns. This shifted DQS signal is not in the center of the
DQ signals, but it is shifted enough to capture the correct data in
this low-frequency application.
Altera Corporation
June 2006
External Memory Interfaces in Stratix & Stratix GX Devices
Figure 3–9. Simplified Diagram of the DQS Phase-Shift Circuitry
Input
Reference
Clock
Phase
Comparator
Up/Down
Counter
Delay Chains
6
Control Signals
to DQS Pins
The input reference clock goes into the DLL to a chain of delay elements.
The phase comparator compares the signal coming out of the end of the
delay element chain to the input reference clock. The phase comparator
then issues the upndn signal to the up/down counter. This signal
increments or decrements a six-bit delay setting (control signals to DQS
pins) that increases or decreases the delay through the delay element
chain to bring the input reference clock and the signals coming out of the
delay element chain in phase.
The shifted DQS signal then goes to the DQS bus to clock the IOE input
registers of the DQ pins. It cannot go into the logic array for other
purposes.
For external memory interfaces that use a bidirectional read strobe like
DDR SDRAM, the DQS signal is low before going to or coming from a
high-impedance state (see Figure 3–1 on page 3–3). The state where DQS
is low just after a high-impedance state is called the preamble and the
state where DQS is low just before it returns to high-impedance state is
called the postamble. There are preamble and postamble specifications
for both read and write operations in DDR SDRAM. To ensure data is not
lost when there is noise on the DQS line at the end of a read postamble
time, you need to add soft postamble circuitry to disable the clocks at the
DQ IOE registers.
f
Altera Corporation
June 2006
For more information, the DQS Postamble soft logic is described in
AN 342: Interfacing DDR SDRAM with Stratix & Stratix GX Devices. The
Altera DDR SDRAM controller MegaCore® generates this logic as
open-source code.
3–19
Stratix Device Handbook, Volume 2
DDR Memory Support Overview
DDR Registers
Each Stratix and Stratix GX IOE contains six registers and one latch. Two
registers and a latch are used for input, two registers are used for output,
and two registers are used for output enable control. The second output
enable register provides the write preamble for the DQS strobe in the
DDR external memory interfaces. This negative-edge output enable
register extends the high-impedance state of the pin by a half clock cycle
to provide the external memory's DQS preamble time specification.
Figure 3–10 shows the six registers and the latch in the Stratix and
Stratix GX IOE and Figure 3–11 shows how the second OE register
extends the DQS high impedance state by half a clock cycle during a write
operation.
3–20
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
External Memory Interfaces in Stratix & Stratix GX Devices
Figure 3–10. Bidirectional DDR I/O Path in Stratix & Stratix GX Devices
Note (1)
DFF
OE
(2)
D
Q
OR2
OE Register AOE (3)
1
0
(4)
DFF
D
Q
OE Register BOE (5)
DFF
datain_l
D
Q
0
1
TRI (6)
I/O Pin (7)
Output Register AO
DFF
Logic Array
datain_h
D
Q
Output Register BO
outclock
combout
DFF
dataout_h
Q
D
Input Register AI
LatchTCHLA
dataout_l
Q
D
DFF
neg_reg_out
Q
D
ENA
Latch C I
Input Register BI
inclock
Notes to Figure 3–10:
(1)
(2)
(3)
(4)
(5)
(6)
(7)
All control signals can be inverted at the IOE. No programmable delay chains are shown in this diagram.
The OE signal is active low, but the Quartus II software implements this as active high and automatically adds an
inverter before input to the AOE register during compilation.
The AOE register generates the enable signal for general-purpose DDR I/O applications.
This select line is to choose whether the OE signal should be delayed by half-a-clock cycle.
The BOE register generates the delayed enable signal for the write strobes and write clock for memory interfaces.
The tristate enable is active low by default. You can design it to be active high. The combinational control path for
the tristate is not shown in this diagram.
You can also have combinational output to the I/O pin; this path is not shown in the diagram.
Altera Corporation
June 2006
3–21
Stratix Device Handbook, Volume 2
DDR Memory Support Overview
Figure 3–11. Extending the OE Disable by Half-a-Clock Cycle for a Write Transaction
Note (1)
System clock
(outclock for DQS)
OE for DQS
(from logic array)
90˚
DQS
Delay
by Half
a Clock
Cycle
Preamble
Postamble
Write Clock
(outclock for DQ,
−90° phase shifted
from System Clock)
datain_h
(from logic array)
D0
D2
datain_l
(from logic array)
D1
D3
OE for DQ
(from logic array)
D0
DQ
D1
D2
D3
Note to Figure 3–11:
(1)
The waveform reflects the software simulation result. The OE signal is an active low on the device. However, the
Quartus II software implements this signal as an active high and automatically adds an inverter before the AOE
register D input.
Figures 3–12 and 3–13 summarize the IOE registers used for the DQ and
DQS signals.
3–22
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
External Memory Interfaces in Stratix & Stratix GX Devices
Figure 3–12. DQ Configuration in Stratix & Stratix GX IOE
Note (1)
DFF
(2)
D
OE
Q
OE Register AOE
DFF
D
datain_l
Q
0
1
Output Register AO
TRI
DQ Pin
DFF
Logic Array
D
datain_h
Q
Output Register BO
outclock (3)
DFF
Q
D
dataout_h
Input Register AI
Latch
TCH
LA
Q
dataout_l
D
DFF
neg_reg_out
Q
D
ENA
Latch C I
Input Register BI
inclock (from DQS bus)
(4)
Notes to Figure 3–12:
(1)
(2)
(3)
(4)
You can use the altdq megafunction to generate the DQ signals.
The OE signal is active low, but the Quartus II software implements this as active high and automatically adds an
inverter before the OE register AOE during compilation.
The outclock signal is phase shifted –90° from the system clock.
The shifted DQS signal must be inverted before going to the IOE. The inversion is automatic if you use the altdq
megafunction to generate the DQ signals.
Altera Corporation
June 2006
3–23
Stratix Device Handbook, Volume 2
DDR Memory Support Overview
Figure 3–13. DQS Configuration in Stratix & Stratix GX IOE
Note (1)
DFF
OE
(2)
D
Q
OE Register AOE
OR2
1
0
(3)
DFF
D
Q
OE Register BOE
DFF
Logic Array
datain_h (3)
D
Q
0
Output Register AO
TRI
DQS Pin (5)
1
DFF
datain_l (4)
system clock
D
Q
Output Register BO
undelayed DQS (6)
combout (7)
DQS Phase
Shift Circuitry
(8)
Notes to Figure 3–13:
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
You can use the altdqs megafunction to generate the DQS signals.
The OE signal is active low, but the Quartus II software implements this as active high and automatically adds an
inverter before OE register AOE during compilation.
The select line can be chosen in the altdqs MegaWizard Plug-In Manager.
The datain_l and datain_h pins are usually connected to VCC and ground, respectively.
DQS postamble handling is not shown in this diagram. For more information, see AN 342: Interfacing DDR SDRAM
with Stratix & Stratix GX Devices.
This undelayed DQS signal goes to the LE for the soft postamble circuitry.
You must invert this signal before it reaches the DQ IOE. This signal is automatically inverted if you use the altdq
megafunction to generate the DQ signals. Connect this port to the inclock port in the altdq megafunction.
DQS phase-shift circuitry is only available on DQS pins.
3–24
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
External Memory Interfaces in Stratix & Stratix GX Devices
The Stratix and Stratix GX DDR IOE structure requires you to invert the
incoming DQS signal by using a NOT gate to ensure proper data transfer.
The altdq megafunction automatically adds the inverter when it
generates the DQ signals. As shown in Figure 3–10, the inclock signal's
rising edge clocks the AI register, inclock signal's falling edge clocks
the BI register, and latch CI is opened when inclock is one. In a DDR
memory read operation, the last data coincides with DQS being low. If
you do not invert the DQS pin, you do not get this last data because the
latch does not open until the next rising edge of the DQS signal. The NOT
gate is inserted automatically if the altdg megafunction is used;
otherwise you need to add the NOT gate manually.
Figure 3–14 shows waveforms of the circuit shown in Figure 3–12. The
second set of waveforms in Figure 3–14 shows what happens if the
shifted DQS signal is not inverted; the last data, Dn, does not get latched
into the logic array as DQS goes to tristate after the read postamble time.
The third set of waveforms in Figure 3–14 shows a proper read operation
with the DQS signal inverted after the 90° shift; the last data Dn does get
latched. In this case the outputs of register AI and latch CI, which
correspond to dataout_h and dataout_l ports, are now switched
because of the DQS inversion.
Altera Corporation
June 2006
3–25
Stratix Device Handbook, Volume 2
DDR Memory Support Overview
Figure 3–14. DQ Captures with Non-Inverted & Inverted Shifted DQS
DQ & DQS Signals
DQ at the pin
Dn − 1
Dn
DQS at the pin
Shifted DQS Signal is Not Inverted
DQS shifted
by 90˚
Output of register A1
(dataout_h)
Output of register B1
Output of latch C1
(dataout_l)
Dn − 1
Dn − 2
Dn
Dn − 2
Shifted DQS Signal is Inverted
DQS inverted and
shifted by 90˚
Output of register A1
(dataout_h)
Output of register B1
Output of latch C1
(dataout_l)
3–26
Stratix Device Handbook, Volume 2
Dn − 2
Dn
Dn − 1
Dn − 3
Dn − 1
Altera Corporation
June 2006
External Memory Interfaces in Stratix & Stratix GX Devices
PLL
When using the Stratix and Stratix GX top and bottom I/O banks (I/O
banks 3, 4, 7, or 8) to interface with a DDR memory, at least one PLL with
two outputs is needed to generate the system clock and the write clock.
The system clock generates the DQS write signals, commands, and
addresses. The write clock is –90° shifted from the system clock and
generates the DQ signals during writes.
When using the Stratix and Stratix GX side I/O banks 1, 2, 5, or 6 to
interface with DDR SDRAM devices, two PLLs may be needed per I/O
bank for best performance. The side I/O banks do not have dedicated
circuitry, so one PLL captures data from the DDR SDRAM and another
PLL generates the write signals, commands, and addresses to the
DDR SDRAM device. Stratix and Stratix GX devices side I/O banks can
support DDR SDRAM up to 150 MHz.
f
Conclusion
Altera Corporation
June 2006
For more information, see AN 342: Interfacing DDR SDRAM with Stratix
& Stratix GX Devices.
Stratix and Stratix GX devices support SDR SDRAM, DDR SDRAM,
RLDRAM II, QDR SDRAM, QDRII SRAM, and ZBT SRAM external
memories. Stratix and Stratix GX devices feature high-speed interfaces
that transfer data between external memory devices at up to
200 MHz/400 Mbps. Phase-shift circuitry in the Stratix and Stratix GX
devices allows you to ensure that clock edges are properly aligned.
3–27
Stratix Device Handbook, Volume 2
Conclusion
3–28
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
Section III. I/O Standards
This section provides information on Stratix® single-ended, voltagereferenced, and differential I/O standards.
It contains the following chapters:
Revision History
■
Chapter 4, Selectable I/O Standards in Stratix & Stratix GX Devices
■
Chapter 5, High-Speed Differential I/O Interfaces in Stratix Devices
The table below shows the revision history for Chapters 4 and 5.
Chapter
Date/Version
4
June 2006, v3.4
●
Updated “AC Hot Socketing Specification” section.
July 2005, v3.3
●
●
Updated “Non-Voltage-Referenced Standards” section.
Minor change to Table 4–6.
●
Updated content throughout.
January 2005,
v3.2
Altera Corporation
Changes Made
Comments
Section III–1
I/O Standards
Chapter
Stratix Device Handbook, Volume 2
Date/Version
September 2004,
v3.1
Changes Made
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
April 2004, v3.0
●
●
●
November 2003,
v2.2
●
●
October 2003,
v2.1
●
●
July 2003, v2.0
●
●
●
●
●
●
Section III–2
Comments
Table 4–1 on page 4–1: renamed table, updated table, and
added Note 1.
Deleted Figure named “1.5-V Differential HSTL Class II
Termination.”
Updated text describing “SSTL-18 Class I & II - EIA/JEDEC
Preliminary Standard JC42.3” on page 4–11.
Updated HyperTransport data rates on page 4–17.
Changed HyperTransport device speed from 800 MHz to
400 MHz on page 4–17.
Added four rows to Table 4–2 on page 4–18: 1.5V HSTL Class I, 1.8-V HSTL Class I, 1.5-V HSTL Class II,
and 1.8-V HSTL Class II.
Changed title of Table 4–3 on page 4–21.
Updated Table 4–4 on page 4–22.
Updated Figure 4–20 on page 4–29.
Added description of which clock pins support differential
on-chip termination on page 4–30.
Updated description of flip-chip packages on page 4–31.
Changed title of Figure 4–21 on page 4–31.
Updated milliamps for non-thermally enhanced cavity up
and non-thermally enhanced FineLine BGA packages on
page 4–35.
Updated equation for FineLine BGA package on
page 4–35.
Updated milliamps in non-thermally enhanced cavity up and
non-thermally enhanced FineLine BGA packages
onpage 4–37.
Updated notes to Figure 4–18.
New information added to the “Hot Socketing” section.
New information added to the “Differential Pad Placement
Guidelines” section.
Removed support for series and parallel on-chip
termination.
Updated Figure 4–22.
Added the Output Enable Group Logic Option in Quartus II
and Toggle Rate Logic Option in Quartus II sections.
Updated notes to Table 4–10.
Renamed impedance matching to series termination
throughout Chapter.
Removed wide range specs for LVTTL and LVCMOS
standards pages 4-3 to 4-5.
Relaxed restriction of input pins next to differential pins for
flipchip packages (pages 4-20, 4-35, and 4-36).
Added Drive Strength section on page 4-26.
Removed text “for 10 ns or less” from AC Hot socketing
specification on page 4-27.
Added Series Termination column to Table 4-9.
Altera Corporation
I/O Standards
Chapter
Date/Version
5
July 2005, v3.2
September 2004,
v3.1
Changes Made
Updated Table 5–14 on page 5–58.
●
●
●
●
●
●
●
April 2004, v3.0
●
●
●
November 2003,
v2.2
●
●
Updated Note 3 in Table 5–10 on page 5–54.
Updated Table 5–7 on page 5–34.
Updated Table 5–8 on page 5–36.
Updated description of “RD Differential Termination” on
page 5–46.
Updated Note 5 in Table 5–14 on page 5–58.
Updated Notes 2, 5, and 7 in Table 5–11 on page 5–56
through Table 5–14 on page 5–58.
Added new text about spanning two I/O banks on
page 5–60.
Updated notes for Figure 5–17.
Updated Table 5–7, 5–8, and 5–10.
“Data Alignment with Clock” section, last sentence: change
made from 90 degrees to 180 degrees.
Removed support for series and parallel on-chip
termination.
Updated the number of channels per PLL in Tables 5-10
through 5-14.
October 2003,
v2.1
●
Added -8 speed grade device information, including Tables
5-7 and 5-8.
July 2003, v2.0
●
Format changes throughout Chapter.
Relaxed restriction of input pins next to differential pins for
flip chip packages in Figure 5-1, Note 5.
Wire bond package performance specification for “high”
speed channels was increased to 624 Mbps from 462 Mbps
throughout Chapter.
Updated high-speed I/O specification for J=2 in Tables 5-7
and 5-8.
Updated Tables 5-10 to 5-14 to reflect PLL cross-bank
support for high-speed differential channels at full speed.
Increased maximum output clock frequency to 462 to 500
MHz on page 5-66.
●
●
●
●
●
Altera Corporation
Comments
Section III–3
I/O Standards
Section III–4
Stratix Device Handbook, Volume 2
Altera Corporation
4. Selectable I/O Standards
in Stratix &
Stratix GX Devices
S52004-3.4
Introduction
The proliferation of I/O standards and the need for higher I/O
performance have made it critical that devices have flexible I/O
capabilities. Stratix® and Stratix GX programmable logic devices (PLDs)
feature programmable I/O pins that support a wide range of industry
I/O standards, permitting increased design flexibility. These I/O
capabilities enable fast time-to-market and high-performance solutions to
meet the demands of complex system designs. Additionally, Stratix and
Stratix GX devices simplify system board design and make it easy to
connect to microprocessors, peripherals, memories, gate arrays,
programmable logic circuits, and standard logic functions.
This chapter provides guidelines for using one or more industry I/O
standards in Stratix and Stratix GX devices, including:
■
■
■
■
■
■
■
■
Stratix & Stratix
GX I/O
Standards
Stratix and Stratix GX I/O standards
High-speed interfaces
Stratix and Stratix GX I/O banks
Programmable current drive strength
Hot socketing
Differential on-chip termination
I/O pad placement guidelines
Quartus® II software support
Stratix and Stratix GX devices support a wide range of industry I/O
standards as shown in the Stratix Device Family Data Sheet section in the
Stratix Device Handbook, Volume 1 and the Stratix GX Device Family Data
Sheet section of the Stratix GX Device Handbook, Volume 1. Several
applications that use these I/O standards are listed in Table 4–1.
Table 4–1. I/O Standard Applications & Performance (Part 1 of 2) Note (1)
I/O Standard
Altera Corporation
June 2006
Application
Performance
3.3-V LVTTL/LVCMOS
General purpose
350 MHz
2.5-V LVTTL/LVCMOS
General purpose
350 MHz
1.8-V LVTTL/LVCMOS
General purpose
250 MHz
1.5-V LVCMOS
General purpose
225 MHz
PCI/CompactPCI
PC/embedded systems
66 MHz
4–1
Stratix & Stratix GX I/O Standards
Table 4–1. I/O Standard Applications & Performance (Part 2 of 2) Note (1)
I/O Standard
Application
Performance
PCI-X 1.0
PC/embedded systems
133 MHz
AGP 1× and 2×
Graphics processors
66 to 133 MHz
SSTL-3 Class I and II
SDRAM
167 MHz
SSTL-2 Class I and II
DDR I SDRAM
160 to 400 Mbps
HSTL Class I
QDR SRAM/SRAM/CSIX
150 to 225 MHz
HSTL Class II
QDR SRAM/SRAM/CSIX
150 to 250 MHz
Differential HSTL
Clock interfaces
150 to 225 MHz
GTL
Backplane driver
200 MHz
GTL+
Pentium processor interface
133 to 200 MHz
LVDS
Communications
840 Mbps
HyperTransport
technology
Motherboard interfaces
800 Mbps
LVPECL
PHY interface
840 Mbps
PCML
Communications
840 Mbps
Differential SSTL-2
DDR I SDRAM
160 to 400 Mbps
CTT
Back planes and bus interfaces 200 MHz
Note to Table 4–1:
(1)
These performance values are dependent on device speed grade, package type
(flip-chip or wirebond) and location of I/Os (top/bottom or left/right). See the
DC & Switching Characteristics chapter of the Stratix Device Handbook, Volume 1.
3.3-V Low Voltage Transistor-Transistor Logic (LVTTL) EIA/JEDEC Standard JESD8-B
The 3.3-V LVTTL I/O standard is a general-purpose, single-ended
standard used for 3.3-V applications. The LVTTL standard defines the DC
interface parameters for digital circuits operating from a 3.0-V or 3.3-V
power supply and driving or being driven by LVTTL-compatible devices.
The LVTTL input standard specifies a wider input voltage range of
–0.5 V ≤VI ≤ 3.8 V. Altera allows an input voltage range of –0.5 V ≤VI ≤ 4.1
V. The LVTTL standard does not require input reference voltages or board
terminations.
Stratix and Stratix GX devices support both input and output levels for
3.3-V LVTTL operation.
4–2
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
Selectable I/O Standards in Stratix & Stratix GX Devices
3.3-V LVCMOS - EIA/JEDEC Standard JESD8-B
The 3.3-V low voltage complementary metal oxide semiconductor
(LVCMOS) I/O standard is a general-purpose, single-ended standard
used for 3.3-V applications. The LVCMOS standard defines the DC
interface parameters for digital circuits operating from a 3.0-V or 3.3-V
power supply and driving or being driven by LVCMOS-compatible
devices.
The LVCMOS standard specifies the same input voltage requirements as
LVTTL (–0.5 V ≤VI ≤ 3.8 V). The output buffer drives to the rail to meet the
minimum high-level output voltage requirements. The 3.3-V I/O
standard does not require input reference voltages or board terminations.
Stratix and Stratix GX devices support both input and output levels for
3.3-V LVCMOS operation.
2.5-V LVTTL Normal Voltage Range - EIA/JEDEC Standard
EIA/JESD8-5
The 2.5-V I/O standard is used for 2.5-V LVTTL applications. This
standard defines the DC interface parameters for high-speed, lowvoltage, non-terminated digital circuits driving or being driven by other
2.5-V devices. The input and output voltage ranges are:
■
■
The 2.5-V normal range input standards specify an input voltage
range of – 0.3 V ≤ VI ≤ 3.0 V.
The normal range minimum high-level output voltage requirement
(VOH) is 2.1 V.
Stratix and Stratix GX devices support both input and output levels for
2.5-V LVTTL operation.
2.5-V LVCMOS Normal Voltage Range - EIA/JEDEC Standard
EIA/JESD8-5
The 2.5-V I/O standard is used for 2.5-V LVCMOS applications. This
standard defines the DC interface parameters for high-speed, lowvoltage, non-terminated digital circuits driving or being driven by other
2.5-V parts. The input and output voltage ranges are:
■
■
Altera Corporation
June 2006
The 2.5-V normal range input standards specify an input voltage
range of – 0.5 V ≤ VI ≤ 3.0 V.
The normal range minimum VOH requirement is 2.1 V.
4–3
Stratix Device Handbook, Volume 2
Stratix & Stratix GX I/O Standards
Stratix and Stratix GX devices support both input and output levels for
2.5-V LVCMOS operation.
1.8-V LVTTL Normal Voltage Range - EIA/JEDEC Standard
EIA/JESD8-7
The 1.8-V I/O standard is used for 1.8-V LVTTL applications. This
standard defines the DC interface parameters for high-speed, lowvoltage, non-terminated digital circuits driving or being driven by other
1.8-V parts. The input and output voltage ranges are:
■
■
The 1.8-V normal range input standards specify an input voltage
range of – 0.5 V ≤ VI ≤ 2.3 V.
The normal range minimum VOH requirement is VCCIO – 0.45 V.
Stratix and Stratix GX devices support both input and output levels for
1.8-V LVTTL operation.
1.8-V LVCMOS Normal Voltage Range - EIA/JEDEC Standard
EIA/JESD8-7
The 1.8-V I/O standard is used for 1.8-V LVCMOS applications. This
standard defines the DC interface parameters for high-speed, lowvoltage, non-terminated digital circuits driving or being driven by other
1.8-V devices. The input and output voltage ranges are:
■
■
The 1.8-V normal range input standards specify an input voltage
range of – 0.5 V ≤ VI ≤ 2.5 V.
The normal range minimum VOH requirement is VCCIO – 0.45 V.
Stratix and Stratix GX devices support both input and output levels for
1.8-V LVCMOS operation.
1.5-V LVCMOS Normal Voltage Range - EIA/JEDEC Standard
JESD8-11
The 1.5-V I/O standard is used for 1.5-V applications. This standard
defines the DC interface parameters for high-speed, low-voltage, nonterminated digital circuits driving or being driven by other 1.5-V devices.
The input and output voltage ranges are:
■
■
The 1.5-V normal range input standards specify an input voltage
range of – 0.5 V ≤ VI ≤ 2.0 V.
The normal range minimum VOH requirement is 1.05 V.
4–4
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
Selectable I/O Standards in Stratix & Stratix GX Devices
Stratix and Stratix GX devices support both input and output levels for
1.5-V LVCMOS operation.
1.5-V HSTL Class I & II - EIA/JEDEC Standard EIA/JESD8-6
The high-speed transceiver logic (HSTL) I/O standard is used for
applications designed to operate in the 0.0- to 1.5-V HSTL logic switching
range. This standard defines single ended input and output specifications
for all HSTL-compliant digital integrated circuits. The single ended input
standard specifies an input voltage range of – 0.3 V ≤ VI ≤ VCCIO + 0.3 V.
Stratix and Stratix GX devices support both input and output levels
specified by the 1.5-V HSTL I/O standard. The input clock is
implemented using dedicated differential input buffers. Two singleended output buffers are automatically programmed to have opposite
polarity so as to implement a differential output clock. Additionally, the
1.5-V HSTL I/O standard in Stratix and Stratix GX devices is compatible
with the 1.8-V HSTL I/O standard in APEXTM 20KE and APEX 20KC
devices because the input and output voltage thresholds are compatible.
See Figures 4–1 and 4–2. Stratix and Stratix GX devices support both
input and output levels with VREF and VTT.
Figure 4–1. HSTL Class I Termination
VTT = 0.75 V
Output Buffer
50 Ω
Z = 50 Ω
Input Buffer
VREF = 0.75 V
Figure 4–2. HSTL Class II Termination
VTT = 0.75 V
VTT = 0.75 V
Output Buffer
50 Ω
50 Ω
Z = 50 Ω
Input Buffer
VREF = 0.75 V
Altera Corporation
June 2006
4–5
Stratix Device Handbook, Volume 2
Stratix & Stratix GX I/O Standards
1.5-V Differential HSTL - EIA/JEDEC Standard EIA/JESD8-6
The differential HSTL I/O standard is used for applications designed to
operate in the 0.0- to 1.5-V HSTL logic switching range such as quad data
rate (QDR) memory clock interfaces. The differential HSTL specification
is the same as the single ended HSTL specification. The standard specifies
an input voltage range of – 0.3 V ≤ VI ≤ VCCIO + 0.3 V. Differential HSTL
does not require an input reference voltage, however, it does require a
50 Ω resistor termination resistor to VTT at the input buffer (see
Figure 4–3). Stratix and Stratix GX devices support both input and output
clock levels for 1.5-V differential HSTL. The input clock is implemented
using dedicated differential input buffer. Two single-ended output
buffers are automatically programmed to have opposite polarity so as to
implement a differential output clock.
Figure 4–3. 1.5-V Differential HSTL Class I Termination
VTT = 0.75 V
Differential
Transmitter
50 Ω
VTT = 0.75 V
50 Ω
Differential
Receiver
Z0 = 50 Ω
Z0 = 50 Ω
3.3-V PCI Local Bus - PCI Special Interest Group PCI Local Bus
Specification Rev. 2.3
The PCI local bus specification is used for applications that interface to
the PCI local bus, which provides a processor-independent data path
between highly integrated peripheral controller components, peripheral
add-in boards, and processor/memory systems. The conventional PCI
specification revision 2.3 defines the PCI hardware environment
including the protocol, electrical, mechanical, and configuration
specifications for the PCI devices and expansion boards. This standard
requires 3.3-V VCCIO. Stratix and Stratix GX devices are fully compliant
with the 3.3-V PCI Local Bus Specification Revision 2.3 and meet
64-bit/66-MHz operating frequency and timing requirements. The 3.3-V
PCI standard does not require input reference voltages or board
terminations. Stratix and Stratix GX devices support both input and
output levels.
4–6
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
Selectable I/O Standards in Stratix & Stratix GX Devices
3.3-V PCI-X 1.0 Local Bus - PCI-SIG PCI-X Local Bus
Specification Revision 1.0a
The PCI-X 1.0 standard is used for applications that interface to the PCI
local bus. The standard enables the design of systems and devices that
operate at clock speeds up to 133 MHz, or 1 gigabit per second (Gbps) for
a 64-bit bus. The PCI-X 1.0 protocol enhancements enable devices to
operate much more efficiently, providing more usable bandwidth at any
clock frequency. By using the PCI-X 1.0 standard, devices can be designed
to meet PCI-X 1.0 requirements and operate as conventional 33- and
66-MHz PCI devices when installed in those systems. This standard
requires 3.3-V VCCIO. Stratix and Stratix GX devices are fully compliant
with the 3.3-V PCI-X Specification Revision 1.0a and meet the 133-MHz
operating frequency and timing requirements. The 3.3-V PCI standard
does not require input reference voltages or board terminations. Stratix
and Stratix GX devices support both input and output levels.
3.3-V Compact PCI Bus - PCI SIG PCI Local Bus Specification
Revision 2.3
The Compact PCI local bus specification is used for applications that
interface to the PCI local bus. It follows the PCI Local Bus Specification
Revision 2.3 plus additional requirements in PCI Industrial Computers
Manufacturing Group (PICMG) specifications PICMG 2.0 R3.0,
CompactPCI specification, and the hot swap requirements in PICMG 2.1
R2.0, CompactPCI Hot Swap Specification. This standard has similar
electrical requirements as LVTTL and requires 3.3-V VCCIO. Stratix and
Stratix GX devices are compliant with the Compact PCI electrical
requirements. The 3.3-V PCI standard does not require input reference
voltages or board terminations. Stratix and Stratix GX devices support
both input and output levels.
3.3-V 1× AGP - Intel Corporation Accelerated Graphics Port
Interface Specification 2.0
The AGP interface is a platform bus specification that enables highperformance graphics by providing a dedicated high-speed port for the
movement of large blocks of 3-dimensional texture data between a PC's
graphics controller and system memory. The 1× AGP I/O standard is a
single-ended standard used for 3.3-V graphics applications. The 1× AGP
input standard specifies an input voltage range of
– 0.5 V ≤ VI ≤ VCCIO + 0.5 V. The 1× AGP standard does not require input
reference voltages or board terminations. Stratix and Stratix GX devices
support both input and output levels.
Altera Corporation
June 2006
4–7
Stratix Device Handbook, Volume 2
Stratix & Stratix GX I/O Standards
3.3-V 2× AGP - Intel Corporation Accelerated Graphics Port
Interface Specification 2.0
The 2× AGP I/O standard is a voltage-referenced, single-ended standard
used for 3.3-V graphics applications. The 2× AGP input standard
specifies an input voltage range of – 0.5V ≤ VI ≤ VCCIO + 0.5V. The 2× AGP
standard does not require board terminations. Stratix and Stratix GX
devices support both input and output levels.
GTL - EIA/JEDEC Standard EIA/JESD8-3
The GTL I/O standard is a low-level, high-speed back plane standard
used for a wide range of applications from ASICs and processors to
interface logic devices. The GTL standard defines the DC interface
parameters for digital circuits operating from power supplies of 2.5, 3.3,
and 5.0 V. The GTL standard is an open-drain standard, and Stratix and
Stratix GX devices support a 2.5- or 3.3-V VCCIO to meet this standard.
GTL requires a 0.8-V VREF and open-drain outputs with a 1.2-V VTT (see
Figure 4–4). Stratix and Stratix GX devices support both input and output
levels.
Figure 4–4. GTL Termination
VTT = 1.2 V
Output Buffer
VTT = 1.2 V
50 Ω
Z = 50 Ω
50 Ω
Input Buffer
VREF = 0.8 V
GTL+
The GTL+ I/O standard is used for high-speed back plane drivers and
Pentium processor interfaces. The GTL+ standard defines the DC
interface parameters for digital circuits operating from power supplies of
2.5, 3.3, and 5.0 V. The GTL+ standard is an open-drain standard, and
Stratix and Stratix GX devices support a 2.5- or 3.3-V VCCIO to meet this
standard. GTL+ requires a 1.0-V VREF and open-drain outputs with a
1.5-V VTT (see Figure 4–5). Stratix and Stratix GX devices support both
input and output levels.
4–8
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
Selectable I/O Standards in Stratix & Stratix GX Devices
Figure 4–5. GTL+ Termination
VTT = 1.5 V
VTT = 1.5 V
Output Buffer
50 Ω
50 Ω
Z = 50 Ω
Input Buffer
VREF = 1.0 V
CTT - EIA/JEDEC Standard JESD8-4
The CTT I/O standard is used for backplanes and memory bus interfaces.
The CTT standard defines the DC interface parameters for digital circuits
operating from 2.5- and 3.3-V power supplies. The CTT standard does not
require special circuitry to interface with LVTTL or LVCMOS devices
when the CTT driver is not terminated. The CTT standard requires a 1.5-V
VREF and a 1.5-V VTT (see Figure 4–6). Stratix and Stratix GX devices
support both input and output levels.
Figure 4–6. CTT Termination
VTT = 1.5 V
Output Buffer
50 Ω
Z = 50 Ω
Input Buffer
VREF = 1.5 V
SSTL-3 Class I & II - EIA/JEDEC Standard JESD8-8
The SSTL-3 I/O standard is a 3.3-V memory bus standard used for
applications such as high-speed SDRAM interfaces. This standard
defines the input and output specifications for devices that operate in the
SSTL-3 logic switching range of 0.0 to 3.3 V. The SSTL-3 standard specifies
an input voltage range of – 0.3 V ≤ VI ≤ VCCIO + 0.3 V. SSTL-3 requires a 1.5V VREF and a 1.5-V VTT to which the series and termination resistors are
connected (see Figures 4–7 and 4–8). Stratix and Stratix GX devices
support both input and output levels.
Altera Corporation
June 2006
4–9
Stratix Device Handbook, Volume 2
Stratix & Stratix GX I/O Standards
Figure 4–7. SSTL-3 Class I Termination
VTT = 1.5 V
Output Buffer
50 Ω
25 Ω
Z = 50 Ω
Input Buffer
VREF = 1.5 V
Figure 4–8. SSTL-3 Class II Termination
VTT = 1.5 V
VTT = 1.5 V
50 Ω
50 Ω
Output Buffer
25 Ω
Z = 50 Ω
Input Buffer
VREF = 1.5 V
SSTL-2 Class I & II - EIA/JEDEC Standard JESD8-9A
The SSTL-2 I/O standard is a 2.5-V memory bus standard used for
applications such as high-speed DDR SDRAM interfaces. This standard
defines the input and output specifications for devices that operate in the
SSTL-2 logic switching range of 0.0 to 2.5 V. This standard improves
operation in conditions where a bus must be isolated from large stubs.
The SSTL-2 standard specifies an input voltage range of
– 0.3 V ≤ VI ≤ VCCIO + 0.3 V. SSTL-2 requires a 1.25-V VREF and a 1.25-V
VTT to which the series and termination resistors are connected (see
Figures 4–9 and 4–10). Stratix and Stratix GX devices support both input
and output levels.
Figure 4–9. SSTL-2 Class I Termination
VTT = 1.25 V
Output Buffer
50 Ω
25 Ω
Z = 50 Ω
Input Buffer
VREF = 1.25 V
4–10
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
Selectable I/O Standards in Stratix & Stratix GX Devices
Figure 4–10. SSTL-2 Class II Termination
VTT = 1.25 V
VTT = 1.25 V
Output Buffer
50 Ω
25 Ω
50 Ω
Z = 50 Ω
Input Buffer
VREF = 1.25 V
SSTL-18 Class I & II - EIA/JEDEC Preliminary Standard JC42.3
The SSTL-18 I/O standard is a 1.8-V memory bus standard. This standard
is similar to SSTL-2 and defines input and output specifications for
devices that are designed to operate in the SSTL-18 logic switching range
0.0 to 1.8 V. SSTL-18 requires a 0.9-V VREF and a 0.9-V VTT to which the
series and termination resistors are connected. See Figures 4–11 and 4–12
for details on SSTL-18 Class I and II termination. Stratix and Stratix GX
devices support both input and output levels.
Figure 4–11. SSTL-18 Class I Termination
VTT = 0.9 V
Output Buffer
50 Ω
25 Ω
Z = 50 Ω
Input Buffer
VREF = 0.9 V
Figure 4–12. SSTL-18 Class II Termination
VTT = 0.9 V
VTT = 0.9 V
50 Ω
50 Ω
Output Buffer
25 Ω
Z = 50 Ω
Input Buffer
VREF = 0.9 V
Differential SSTL-2 - EIA/JEDEC Standard JESD8-9A
The differential SSTL-2 I/O standard is a 2.5-V standard used for
applications such as high-speed DDR SDRAM clock interfaces. This
standard supports differential signals in systems using the SSTL-2
Altera Corporation
June 2006
4–11
Stratix Device Handbook, Volume 2
Stratix & Stratix GX I/O Standards
standard and supplements the SSTL-2 standard for differential clocks.
The differential SSTL-2 standard specifies an input voltage range of
– 0.3 V ≤ VI ≤ VCCIO + 0.3 V. The differential SSTL-2 standard does not
require an input reference voltage differential. See Figure 4–13 for details
on differential SSTL-2 termination. Stratix and Stratix GX devices support
output clock levels for differential SSTL-2 Class II operation. The output
clock is implemented using two single-ended output buffers which are
programmed to have opposite polarity.
Figure 4–13. Differential SSTL-2 Class II Termination
VTT = 1.25 V
Differential
Transmitter
50 Ω
VTT = 1.25 V
50 Ω
VTT = 1.25 V
50 Ω
VTT = 1.25 V
50 Ω
Differential
Receiver
25 Ω
Z0 = 50 Ω
25 Ω
Z0 = 50 Ω
LVDS - ANSI/TIA/EIA Standard ANSI/TIA/EIA-644
The LVDS I/O standard is a differential high-speed, low-voltage swing,
low-power, general-purpose I/O interface standard requiring a 3.3-V
VCCIO. This standard is used in applications requiring high-bandwidth
data transfer, backplane drivers, and clock distribution. The
ANSI/TIA/EIA-644 standard specifies LVDS transmitters and receivers
capable of operating at recommended maximum data signaling rates of
655 Mbps. However, devices can operate at slower speeds if needed, and
there is a theoretical maximum of 1.923 Gbps. Stratix and Stratix GX
devices meet the ANSI/TIA/EIA-644 standard.
Due to the low voltage swing of the LVDS I/O standard, the
electromagnetic interference (EMI) effects are much smaller than CMOS,
TTL, and PECL. This low EMI makes LVDS ideal for applications with
low EMI requirements or noise immunity requirements. The LVDS
standard does not require an input reference voltage, however, it does
require a 100 Ω termination resistor between the two signals at the input
buffer. Stratix and Stratix GX devices include an optional differential
LVDS termination resistor within the device using differential on-chip
termination. Stratix and Stratix GX devices support both input and
output levels.
4–12
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
Selectable I/O Standards in Stratix & Stratix GX Devices
f
For more information on the LVDS I/O standard in Stratix devices, see
the High-Speed Differential I/O Interfaces in Stratix Devices chapter.
LVPECL
The LVPECL I/O standard is a differential interface standard requiring a
3.3-V VCCIO. The standard is used in applications involving video
graphics, telecommunications, data communications, and clock
distribution. The high-speed, low-voltage swing LVPECL I/O standard
uses a positive power supply and is similar to LVDS, however, LVPECL
has a larger differential output voltage swing than LVDS. The LVPECL
standard does not require an input reference voltage, but it does require
a 100-Ω termination resistor between the two signals at the input buffer.
See Figures 4–14 and 4–15 for two alternate termination schemes for
LVPECL. Stratix and Stratix GX devices support both input and output
levels.
Figure 4–14. LVPECL DC Coupled Termination
Output Buffer
Input Buffer
Z = 50 Ω
100 Ω
Z = 50 Ω
Figure 4–15. LVPECL AC Coupled Termination
VCCIO
VCCIO
Output Buffer
10 to 100 nF
Z = 50 Ω
R1
R1
R2
R2
Input Buffer
100 Ω
10 to 100 nF
Z = 50 Ω
Pseudo Current Mode Logic (PCML)
The PCML I/O standard is a differential high-speed, low-power I/O
interface standard used in applications such as networking and
telecommunications. The standard requires a 3.3-V VCCIO. The PCML I/O
standard consumes less power than the LVPECL I/O standard. The
Altera Corporation
June 2006
4–13
Stratix Device Handbook, Volume 2
Stratix & Stratix GX I/O Standards
PCML standard is similar to LVPECL, but PCML has a reduced voltage
swing, which allows for a faster switching time and lower power
consumption. The PCML standard uses open drain outputs and requires
a differential output signal. See Figure 4–16 for details on PCML
termination. Stratix and Stratix GX devices support both input and
output levels.
Additionally, Stratix GX devices support 1.5-V PCML as described in the
Stratix GX Device Handbook, Volume 1.
Figure 4–16. PCML Termination
VTT
Output Buffer
50 Ω
50 Ω
Z = 50 Ω
50 Ω
50 Ω
Input Buffer
Z = 50 Ω
HyperTransport Technology - HyperTransport Consortium
The HyperTransport technology I/O standard is a differential highspeed, high-performance I/O interface standard requiring a 2.5-V VCCIO.
This standard is used in applications such as high-performance
networking, telecommunications, embedded systems, consumer
electronics, and Internet connectivity devices. The HyperTransport
technology I/O standard is a point-to-point standard in which each
HyperTransport technology bus consists of two point-to-point
unidirectional links. Each link is 2 to 32 bits. The HyperTransport
technology standard does not require an input reference voltage.
However, it does require a 100-Ω termination resistor between the two
signals at the input buffer. See Figure 4–17 for details on HyperTransport
technology termination. Stratix and Stratix GX devices support both
input and output levels.
Figure 4–17. HyperTransport Technology Termination
Output Buffer
Input Buffer
Z = 50 Ω
100 Ω
Z = 50 Ω
4–14
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
Selectable I/O Standards in Stratix & Stratix GX Devices
f
High-Speed
Interfaces
See the Stratix Device Family Data Sheet section in the Stratix Device
Handbook, Volume 1; the Stratix GX Device Family Data Sheet section of the
Stratix GX Device Handbook, Volume 1; and the High-Speed Differential I/O
Interfaces in Stratix Devices chapter for more information on differential
I/O standards.
In addition to current industry physical I/O standards, Stratix and
Stratix GX devices also support a variety of emerging high-speed
interfaces. This section provides an overview of these interfaces.
OIF-SPI4.2
This implementation agreement is widely used in the industry for
OC-192 and 10-Gbps multi-service system interfaces. SONET and SDH
are synchronous transmission systems over which data packets are
transferred. POS-PHY Level 4 is a standard interface for switches and
routers, and defines the operation between a physical layer (PHY) device
and link layer devices (ATM, Internet protocol, and Gigabit Ethernet) for
bandwidths of OC-192 ATM, POS, and 10-Gigabit Ethernet applications.
Some key POS-PHY Level 4 system features include:
■
■
■
■
■
■
■
■
■
Large selection of POS-PHY Level 4-based PHYs
Independent of data protocol
Wide industry support
LVDS I/O standard to improve signal integrity
Inband addressing/control
Out of band flow control
Scalable architecture
Over 622-Mbps operation
Dynamic interface timing mode
POS-PHY Level 4 operates at a wide range of frequencies.
OIF-SFI4.1
This implementation agreement is widely used in the industry for
interfacing physical layer (PHY) to the serializer-deserializer (SERDES)
devices in OC-192 and 10 Gbps multi-service systems. The POS-PHY
Level 4 interface standard defines the SFI-4 standard. POS-PHY
Level 4: SFI-4 is a standardized 16-bit × 622-Mbps line-side interface for
10-Gbps applications. Internet LAN and WAN architectures use
telecommunication SONET protocols for data transferring data over the
PHY layer. SFI-4 interfaces between OC-192 SERDES and SONET
framers.
Altera Corporation
June 2006
4–15
Stratix Device Handbook, Volume 2
High-Speed Interfaces
10 Gigabit Ethernet Sixteen Bit Interface (XSBI) - IEEE Draft
Standard P802.3ae/D2.0
10 Gigabit Ethernet XSBI is an interface standard for LANs, metropolitan
area networks (MANs), storage area networks (SANs), and WANs.
10 Gigabit Ethernet XSBI provides many features for efficient, effective
high-speed networking, including easy migration to higher performance
levels without disruption, lower cost of ownership including acquisition
and support versus other alternatives, familiar management tools and
common skills, ability to support new applications and data protocols,
flexibility in network design, and multiple vendor sourcing and
interoperability.
Under the ISO Open Systems Interconnection (OSI) model, Ethernet is a
Layer 2 protocol. 10 Gigabit Ethernet XSBI uses the IEEE 802.3 Ethernet
media access control (MAC) protocol, Ethernet frame format, and the
minimum/maximum frame size. An Ethernet PHY corresponding to OSI
layer 1 connects the media to the MAC layer that corresponds to OSI
layer 2. The PHY is divided into a physical media dependent (PMD)
element, such as optical transceivers, and a physical coding sub-layer
(PCS), which has coding and a serializer/multiplexor. This standard
defines two PHY types, including the LAN PHY and the WAN PHY,
which are distinguished by the PCS. The 10 Gigabit Ethernet XSBI
standard is a full-duplex technology standard that can increase the speed
and distance of Ethernet.
RapidIO Interconnect Specification Revision 1.1
The RapidIO interface is a communications standard used to connect
devices on a circuit board and circuit boards on a backplane. RapidIO is a
packet-switched interconnect standard designed for embedded systems
such as those used in networking and communications. The RapidIO
interface standard is a high-performance interconnect interface used for
transferring data and control information between microprocessors,
DSPs, system memory, communications and network processors, and
peripheral devices in a system.
RapidIO replaces existing peripheral bus and processor technologies
such as PCI. Some features of RapidIO include multiprocessing support,
an open standard, flexible topologies, higher bandwidth, low latency,
error management support in hardware, small silicon footprint, widely
available process and I/O technologies, and transparency to existing
applications and operating system software. The RapidIO standard
provides 10-Gbps device bandwidth using 8-bit-wide input and output
data ports. RapidIO uses LVDS technology, has the capability to be scaled
to multi-GHz frequencies, and features a 10-bit interface.
4–16
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
Selectable I/O Standards in Stratix & Stratix GX Devices
HyperTransport Technology - HyperTransport Consortium
The HyperTransport technology I/O standard is a differential
high-speed, high performance I/O interface standard developed for
communications and networking chip-to-chip communications.
HyperTransport technology is used in applications such as highperformance networking, telecommunications, embedded systems,
consumer electronics, and Internet connectivity devices. The
HyperTransport technology I/O standard is a point-to-point (one source
connected to exactly one destination) standard that provides a highperformance interconnect between integrated circuits in a system, such as
on a motherboard.
Stratix devices support HyperTransport technology at data rates up to
800 Mbps and 32 bits in each direction. HyperTransport technology uses
an enhanced differential signaling technology to improve performance.
HyperTransport technology supports data widths of 2, 4, 8, 16, or 32 bits
in each direction. HyperTransport technology in Stratix and Stratix GX
devices operates at multiple clock speeds up to 400 MHz.
UTOPIA Level 4 – ATM Forum Technical Committee Standard AFPHY-0144.001
The UTOPIA Level 4 frame-based interface standard allows device
manufacturers and network developers to develop components that can
operate at data rates up to 10 Gbps. This standard increases interface
speeds using LVDS I/O and advanced silicon technologies for fast data
transfers.
UTOPIA Level 4 provides new control techniques and a 32-, 16-, or 8-bit
LVDS bus, a symmetric transmit/receive bus structure for easier
application design and testability, nominal data rates of 10 Gbps, in-band
control of cell delimiters and flow control to minimize pin count, sourcesynchronous clocking, and supports variable length packet systems.
UTOPIA Level 4 handles sustained data rates for OC-192 and supports
ATM cells. UTOPIA Level 4 also supports interconnections across
motherboards, daughtercards, and backplane interfaces.
Stratix & Stratix
GX I/O Banks
Altera Corporation
June 2006
Stratix devices have eight I/O banks in addition to the four enhanced PLL
external clock output banks, as shown in Table 4–2 and Figure 4–18. I/O
banks 3, 4, 7, and 8 support all single-ended I/O standards. I/O banks 1,
2, 5, and 6 support differential HSTL (on input clocks), LVDS, LVPECL,
PCML, and HyperTransport technology, as well as all single-ended I/O
standards except HSTL Class II, GTL, SSTL-18 Class II, PCI/PCI-X 1.0,
and 1× /2× AGP. The four enhanced PLL external clock output banks
(I/O banks 9, 10, 11, and 12) support clock outputs all single-ended I/O
4–17
Stratix Device Handbook, Volume 2
Stratix & Stratix GX I/O Banks
standards in addition to differential SSTL-2 and HSTL (both on the output
clock only). Since Stratix devices support both non-voltage-referenced
and voltage-referenced I/O standards, there are different guidelines
when working with either separately or when working with both.
Table 4–2. I/O Standards Supported in Stratix I/O Banks (Part 1 of 2)
Enhanced PLL External
Clock Output Banks
I/O Bank
I/O Standard
1
2
3
4
5
6
7
8
9
10
11
12
3.3-V LVTTL/LVCMOS
v
v
v
v
v
v
v
v
v
v
v
v
2.5-V LVTTL/LVCMOS
v
v
v
v
v
v
v
v
v
v
v
v
1.8-V LVTTL/LVCMOS
v
v
v
v
v
v
v
v
v
v
v
v
1.5-V LVCMOS
v
v
v
v
v
v
v
v
v
v
v
v
PCI/PCIX//Compact PCI
v
v
v
v
v
v
v
v
AGP 1×
v
v
v
v
v
v
v
v
AGP 2×
v
v
v
v
v
v
v
v
SSTL-3 Class I
v
v
v
v
v
v
v
v
v
v
v
v
SSTL-3 Class II
v
v
v
v
v
v
v
v
v
v
v
v
SSTL-2 Class I
v
v
v
v
v
v
v
v
v
v
v
v
SSTL-2 Class II
v
v
v
v
v
v
v
v
v
v
v
v
SSTL-18 Class I
v
v
v
v
v
v
v
v
v
v
v
v
v
v
v
v
v
v
v
v
v
v
v
v
SSTL-18 Class II
Differential SSTL-2
(output clocks)
HSTL Class I
v
v
v
v
v
v
v
v
v
v
v
v
1.5-V HSTL Class I
v
v
v
v
v
v
v
v
v
v
v
v
1.8-V HSTL Class I
v
v
v
v
v
v
v
v
v
v
v
v
HSTL Class II
v
v
v
v
v
v
v
v
1.5-V HSTL Class II
v
v
v
v
v
v
v
v
1.8-V HSTL Class II
v
v
v
v
v
v
v
v
v
v
v
v
v
v
v
v
v
v
v
v
Differential HSTL (input
clocks)
v
v
v
v
Differential HSTL (output
clocks)
GTL
4–18
Stratix Device Handbook, Volume 2
v
v
v
v
Altera Corporation
June 2006
Selectable I/O Standards in Stratix & Stratix GX Devices
Table 4–2. I/O Standards Supported in Stratix I/O Banks (Part 2 of 2)
Enhanced PLL External
Clock Output Banks
I/O Bank
I/O Standard
1
2
3
4
5
6
7
8
9
10
11
12
GTL+
v
v
v
v
v
v
v
v
v
v
v
v
CTT
v
v
v
v
v
v
v
v
v
v
v
v
LVDS
v
v
(1)
(1)
v
v
(1)
(1)
(2)
(2)
(2)
(2)
HyperTransport
technology
v
v
(1)
(1)
v
v
(1)
(1)
(2)
(2)
(2)
(2)
LVPECL
v
v
(1)
(1)
v
v
(1)
(1)
(2)
(2)
(2)
(2)
PCML
v
v
(1)
(1)
v
v
(1)
(1)
(2)
(2)
(2)
(2)
Notes to Table 4–2:
(1)
(2)
This I/O standard is only supported on input clocks in this I/O bank.
This I/O standard is only supported on output clocks in this I/O bank.
Altera Corporation
June 2006
4–19
Stratix Device Handbook, Volume 2
Stratix & Stratix GX I/O Banks
Figure 4–18. Stratix I/O Banks Notes (1), (2), (3)
DQS5T
9
DQS4T
PLL11
(5)
DQS1T
DQS0T
10
Bank 4
LVDS, LVPECL, 3.3-V PCML,
and HyperTransport I/O Block
and Regular I/O Pins (4)
(5)
I/O Banks 1, 2, 5, and 6 Support All
Single-Ended I/O Standards Except
Differential HSTL Output Clocks,
Differential SSTL-2 Output Clocks,
HSTL Class II, GTL, SSTL-18 Class II,
PCI, PCI-X 1.0, and AGP 1×/2×
PLL2
Bank 1
DQS2T
I/O Banks 3, 4, 9 & 10 Support
All Single-Ended I/O Standards
PLL1
Bank 8
PLL3
DQS8B
DQS7B
DQS6B
DQS5B
(5)
LVDS, LVPECL, 3.3-V PCML,
and HyperTransport I/O Block
and Regular I/O Pins (4)
11
VREF5B8 VREF4B8 VREF3B8 VREF2B8 VREF1B8
DQS9B
PLL4
I/O Banks 7, 8, 11 & 12 Support
All Single-Ended I/O Standards
(5)
LVDS, LVPECL, 3.3-V PCML,
and HyperTransport I/O Block
and Regular I/O Pins (4)
PLL8
DQS3T
VREF1B4 VREF2B4 VREF3B4 VREF4B4 VREF5B4 PLL10
LVDS, LVPECL, 3.3-V PCML,
and HyperTransport I/O Block
and Regular I/O Pins (4)
Bank 2
VREF1B2 VREF2B2 VREF3B2 VREF4B2
Bank 3
VREF1B1 VREF2B1 VREF3B1 VREF4B1
PLL5
12
PLL6
Bank 5
DQS6T
VREF4B5 VREF3B5 VREF2B5 VREF1B5
DQS7T
Bank 6
DQS8T
VREF1B3 VREF2B3 VREF3B3 VREF4B3 VREF5B3
VREF4B6 VREF3B6 VREF2B6 VREF1B6
DQS9T
PLL7
Bank 7
PLL12
VREF5B7 VREF4B7 VREF3B7 VREF2B7 VREF1B7
DQS4B
DQS3B
DQS2B
DQS1B
PLL9
DQS0B
Notes to Figure 4–18:
(1)
(2)
(3)
(4)
(5)
Figure 4–18 is a top view of the silicon die. This corresponds to a top-down view for non-flip-chip packages, but is
a reverse view for flip-chip packages.
Figure 4–18 is a graphic representation only. See the pin list and the Quartus II software for exact locations.
Banks 9 through 12 are enhanced PLL external clock output banks.
If the high-speed differential I/O pins are not used for high-speed differential signaling, they can support all of the
I/O standards except HSTL Class II, GTL, SSTL-18 Class II, PCI, PCI-X 1.0, and AGP 1×/2×.
For guidelines on placing single-ended I/O pads next to differential I/O pads, see “I/O Pad Placement Guidelines”
on page 4–30.
4–20
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
Selectable I/O Standards in Stratix & Stratix GX Devices
Tables 4–3 and 4–4 list the I/O standards that Stratix GX enhanced and
fast PLL pins support. Figure 4–19 shows the I/O standards that each
Stratix GX I/O bank supports.
Table 4–3. I/O Standards Supported in Stratix & Stratix GX Enhanced PLL Pins
Input
Output
I/O Standard
INCLK
FBIN
PLLENABLE
EXTCLK
LVTTL
v
v
v
v
LVCMOS
v
v
v
v
2.5 V
v
v
v
1.8 V
v
v
v
1.5 V
v
v
v
3.3-V PCI
v
v
v
3.3-V PCI-X 1.0
v
v
v
LVPECL
v
v
v
3.3-V PCML
v
v
v
LVDS
v
v
v
HyperTransport technology
v
v
v
Differential HSTL
v
v
v
Differential SSTL
3.3-V GTL
v
v
v
3.3-V GTL+
v
v
v
1.5-V HSTL Class I
v
v
v
1.5-V HSTL Class II
v
v
v
SSTL-18 Class I
v
v
v
SSTL-18 Class II
v
v
v
SSTL-2 Class I
v
v
v
SSTL-2 Class II
v
v
v
SSTL-3 Class I
v
v
v
SSTL-3 Class II
v
v
v
AGP (1× and 2×)
v
v
v
CTT
v
v
v
Altera Corporation
June 2006
4–21
Stratix Device Handbook, Volume 2
Stratix & Stratix GX I/O Banks
Table 4–4. I/O Standards Supported in Stratix & Stratix GX Fast PLL Pins
Input
I/O Standard
INCLK
PLLENABLE
LVTTL
v
v
LVCMOS
v
v
2.5 V
v
1.8 V
v
1.5 V
v
3.3-V PCI
3.3-V PCI-X 1.0
LVPECL
v
3.3-V PCML
v
LVDS
v
HyperTransport technology
v
Differential HSTL
v
Differential SSTL
3.3-V GTL
3.3-V GTL+
1.5V HSTL Class I
v
1.5V HSTL Class II
SSTL-18 Class I
v
SSTL-18 Class II
SSTL-2 Class I
v
SSTL-2 Class II
v
SSTL-3 Class I
v
SSTL-3 Class II
v
AGP (1× and 2×)
CTT
4–22
Stratix Device Handbook, Volume 2
v
Altera Corporation
June 2006
Selectable I/O Standards in Stratix & Stratix GX Devices
Figure 4–19. Stratix GX I/O Banks
I/O Bank 2
I/O Banks 1 & 2 Support:
■ Differential I/O Standards
- True LVDS
- LVPECL
- 3.3-V PCML
- HyperTransport Technology
■ Single-Ended I/O Standard
- 3.3 -, 2.5 -, 1.8 -V LVTTL
- GTL+
- CTT
- SSTL-18 Class I
- SSTL-2 Class I and II
- SSTL-3 Class I and II
- 1.5 -, 1.8 -V HSTL Class I
I/O Bank 1
I/O Bank 3
I/O Bank 4
I/O Banks 3, 4, 6 & 7 Support:
■ 3.3-, 2.5-, 1.8-V LVTTL
■ 3.3-V PCI, PCI-X 1.0
■ GTL
■ GTL+
■ AGP
■ CTT
■ SSTL-18 Class I and II
■ SSTL-2 Class I and II
■ SSTL-3 Class I and II
■ HSTL Class I and II
Individual
Power Bus
I/O Bank 7
I/O Bank 6
I/O Bank 5
I/O Bank 5 Contains Transceiver Blocks
There is some flexibility with the number of I/O standards each Stratix
I/O bank can simultaneously support. The following sections provide
guidelines for mixing non-voltage-referenced and voltage-referenced
I/O standards in Stratix devices.
Altera Corporation
June 2006
4–23
Stratix Device Handbook, Volume 2
Stratix & Stratix GX I/O Banks
Non-Voltage-Referenced Standards
Each Stratix I/O bank has its own VCCIO pins and supports only one
VCCIO, either 1.5, 1.8, 2.5 or 3.3 V. A Stratix I/O bank can simultaneously
support any number of input signals with different I/O standard
assignments, as shown in Table 4–5.
Table 4–5. Acceptable Input Levels for LVTTL/LVCMOS
Acceptable Input Levels
Bank VCCIO
3.3 V
2.5 V
1.8 V
1.5 V
3.3 V
v
v
2.5 V
v
v
1.8 V
v (2)
v (2)
v
v (1)
1.5 V
v (2)
v (2)
v
v
Notes to Table 4–5:
(1)
(2)
Because the input signal will not drive to the rail, the input buffer does not
completely shut off, and the I/O current will be slightly higher than the default
value.
These input values overdrive the input buffer, so the pin leakage current will be
slightly higher than the default value.
For output signals, a single I/O bank can only support non-voltagereferenced output signals driving at the same voltage as VCCIO. A Stratix
I/O bank can only have one VCCIO value, so it can only drive out that one
value for non-voltage referenced signals. For example, an I/O bank with
a 2.5-V VCCIO setting can support 2.5-V LVTTL inputs and outputs,
HyperTransport technology inputs and outputs, and 3.3-V LVCMOS
inputs (not output or bidirectional pins).
1
If the output buffer overdrives the input buffer, you must turn
on the Allow voltage overdrive for LVTTL/LVCMOS option in
the Quartus II software. To see this option, click the Device &
Pin Options button in the Device page of the Settings dialog
box (Assignments menu). Then click the Pin Placement tab in
the Device & Pin Options dialog box.
Voltage-Referenced Standards
To accommodate voltage-referenced I/O standards, each Stratix I/O
bank supports multiple VREF pins feeding a common VREF bus. The
number of available VREF pins increases as device density increases. If
these pins are not used as VREF pins, they can not be used as generic I/O
pins.
4–24
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
Selectable I/O Standards in Stratix & Stratix GX Devices
An I/O bank featuring single-ended or differential standards can support
voltage-referenced standards as long as all voltage-referenced standards
use the same VREF setting. For example, although one I/O bank can
implement both SSTL-3 and SSTL-2 I/O standards, I/O pins using these
standards must be in different banks since they require different VREF
values
For voltage-referenced inputs, the receiver compares the input voltage to
the voltage reference and does not take into account the VCCIO setting.
Therefore, the VCCIO setting is irrelevant for voltage referenced inputs.
Voltage-referenced bidirectional and output signals must be the same as
the I/O bank’s VCCIO voltage. For example, although you can place an
SSTL-2 input pin in any I/O bank with a 1.25-V VREF level, you can only
place SSTL-2 output pins in an I/O bank with a 2.5-V VCCIO.
Mixing Voltage Referenced & Non-Voltage Referenced
Standards
Non-voltage referenced and voltage referenced pins can safely be mixed
in a bank by applying each of the rule-sets individually. For example, on
I/O bank can support SSTL-3 inputs and 1.8-V LVCMOS inputs and
outputs with a 1.8-V VCCIO and a 1.5-V VREF. Similarly, an I/O bank can
support 1.5-V LVCMOS, 3.3-V LVTTL (inputs, but not outputs), and
HSTL I/O standards with a 1.5-V VCCIO and 0.75-V VREF.
For the voltage-referenced examples, see the “I/O Pad Placement
Guidelines” section. For details on how the Quartus II software supports
I/O standards, see the “Quartus II Software Support”section.
Altera Corporation
June 2006
4–25
Stratix Device Handbook, Volume 2
Drive Strength
Drive Strength
Each I/O standard supported by Stratix and Stratix GX devices drives out
a minimum drive strength. When an I/O is configured as LVTTL or
LVCMOS I/O standards, you can specify the current drive strength, as
summarized in Table 4–7.
Standard Current Drive Strength
Each I/O standard supported by Stratix and Stratix GX devices drives out
a minimum drive strength. Table 4–6 summarizes the minimum drive
strength of each I/O standard.
Table 4–6. Minimum Current Drive Strength of Each I/O Standard
I/O Standard
Current Strength, IOL/IOH (mA)
GTL
40 (1)
GTL+
34 (1)
SSTL-3 Class I
8
SSTL-3 Class II
16
SSTL-2 Class I
8.1
SSTL-2 Class II
16.4
SSTL-18 Class I
6.7
SSTL-18 Class II
13.4
1.5-V HSTL Class I
8
1.5-V HSTL Class II
16
CTT
8
AGP 1×
IOL = 1.5, IOH = –0.5
Note to Table 4–6:
(1)
Because this I/O standard uses an open drain buffer, this value refers to IOL.
When the SSTL-2 Class I and II I/O standards are implemented on top or
bottom I/O pins, the drive strength is designed to be higher than the
drive strength of the buffer when implemented on side I/O pins. This
allows the top or bottom I/O pins to support 200-MHz operation with the
standard 35-pF load. At the same time, the current consumption when
using top or bottom I/O pins is higher than the side I/O pins. The high
current strength may not be necessary for certain applications where the
value of the load is less than the standard test load (e.g., DDR interface).
The Quartus II software allows you to reduce the drive strength when the
I/O pins are used for the SSTL-2 Class I or Class II I/O standard and
being implemented on the top or bottom I/O through the Current
Strength setting. Select the minimum strength for lower drive strength.
4–26
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
Selectable I/O Standards in Stratix & Stratix GX Devices
Programmable Current Drive Strength
The Stratix and Stratix GX device I/O pins support various output
current drive settings as shown in Table 4–7. These programmable drive
strength settings help decrease the effects of simultaneously switching
outputs (SSO) in conjunction with reducing system noise. The supported
settings ensure that the device driver meets the IOH and IOL specifications
for the corresponding I/O standard.
Table 4–7. Programmable Drive Strength
I/O Standard
IOH / IOL Current Strength Setting (mA)
3.3-V LVTTL
24 (1), 16, 12, 8, 4
3.3-V LVCMOS
24 (2), 12 (1), 8, 4, 2
2.5-V LVTTL/LVCMOS
16 (1), 12, 8, 2
1.8-V LVTTL/LVCMOS
12 (1), 8, 2
1.5-V LVCMOS
8 (1), 4, 2
Notes to Table 4–7:
(1)
(2)
This is the Quartus II software default current setting.
I/O banks 1, 2, 5, and 6 do not support this setting.
These drive-strength settings are programmable on a per-pin basis (for
output and bidirectional pins only) using the Quartus II software. To
modify the current strength of a particular pin, see “Programmable Drive
Strength Settings” on page 4–40.
Hot Socketing
Stratix devices support hot socketing without any external components.
In a hot socketing situation, a device’s output buffers are turned off
during system power-up or power-down. Stratix and Stratix GX devices
support any power-up or power-down sequence (VCCIO and VCCINT) to
simplify designs. For mixed-voltage environments, you can drive signals
into the device before or during power-up or power-down without
damaging the device. Stratix and Stratix GX devices do not drive out until
the device is configured and has attained proper operating conditions.
Even though you can power up or down the VCCIO and VCCINT power
supplies in any sequence you should not power down any I/O bank(s)
that contains the configuration pins while leaving other I/O banks
powered on. For power up and power down, all supplies (VCCINT and all
VCCIO power planes) must be powered up and down within 100 ms of one
another. This prevents I/O pins from driving out.
Altera Corporation
June 2006
4–27
Stratix Device Handbook, Volume 2
I/O Termination
You can power up or power down the VCCIO and VCCINT pins in any
sequence. The power supply ramp rates can range from 100 ns to 100 ms.
During hot socketing, the I/O pin capacitance is less than 15 pF and the
clock pin capacitance is less than 20 pF.
DC Hot Socketing Specification
The hot socketing DC specification is | IIOPIN | < 300 μ A.
AC Hot Socketing Specification
The hot socketing AC specification is | IIOPIN | < 8 mA for 10 ns or less.
This specification takes into account the pin capacitance, but not board
trace and external loading capacitance. Additional capacitance for trace,
connector, and loading must be considered separately.
IIOPIN is the current at any user I/O pin on the device. The DC
specification applies when all VCC supplies to the device are stable in the
powered-up or powered-down conditions. For the AC specification, the
peak current duration because of power-up transients is 10 ns or less. For
more information, refer to the Hot-Socketing & Power-Sequencing Feature &
Testing for Altera Devices white paper.
I/O Termination
Although single-ended, non-voltage-referenced I/O standards do not
require termination, Altera recommends using external termination to
improve signal integrity where required.
The following I/O standards do not require termination:
■
■
■
■
■
■
■
■
LVTTL
LVCMOS
2.5 V
1.8 V
1.5 V
3.3-V PCI/Compact PCI
3.3-V PCI-X 1.0
3.3-V AGP 1×
Voltage-Referenced I/O Standards
Voltage-referenced I/O standards require both an input reference
voltage, VREF, and a termination voltage, VTT. Off-chip termination on the
board should be used for series and parallel termination.
4–28
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
Selectable I/O Standards in Stratix & Stratix GX Devices
For more information on termination for voltage-referenced I/O
standards, see the Selectable I/O Standards in Stratix & Stratix GX Devices
chapter in the Stratix Device Handbook, Volume 2; or the Stratix GX Device
Handbook, Volume 2.
Differential I/O Standards
Differential I/O standards typically require a termination resistor
between the two signals at the receiver. The termination resistor must
match the differential load impedance of the bus. Stratix and Stratix GX
devices provide an optional differential termination on-chip resistor
when using LVDS.
See the High-Speed Differential I/O Interfaces in Stratix Devices chapter for
more information on differential I/O standards and their interfaces.
For differential I/O standards, I/O banks support differential
termination when VCCIO equals 3.3 V.
Differential Termination (RD)
Stratix devices support differential on-chip termination for sourcesynchronous LVDS signaling. The differential termination resistors are
adjacent to the differential input buffers on the device. This placement
eliminates stub effects, improving the signal integrity of the serial link.
Using differential on-chip termination resistors also saves board space.
Figure 4–20 shows the differential termination connections for Stratix and
Stratix GX devices.
Figure 4–20. Differential Termination
Differential
Transmitter
Stratix LVDS
Receiver Buffer with
Differential On-Chip Termination
Z0
RD
Z0
Altera Corporation
June 2006
4–29
Stratix Device Handbook, Volume 2
I/O Pad Placement Guidelines
Differential termination for Stratix devices is supported for the left and
right I/O banks. Differential termination for Stratix GX devices is
supported for the left, source-synchronous I/O bank. Some of the clock
input pins are in the top and bottom I/O banks, which do not support
differential termination. Clock pins CLK[1,3,8,10] support differential
on-chip termination. Clock pins CLK[0,2,9,11], CLK[4-7], and CLK[12-15]
do not support differential on-chip termination.
Transceiver Termination
Stratix GX devices feature built-in on-chip termination within the
transceiver at both the transmit and receive buffers. This termination
improves signal integrity and provides support for the 1.5-V PCML I/O
standard.
I/O Pad
Placement
Guidelines
This section provides pad placement guidelines for the programmable
I/O standards supported by Stratix and Stratix GX devices and includes
essential information for designing systems using the devices' selectable
I/O capabilities. These guidelines will reduce noise problems so that
FPGA devices can maintain an acceptable noise level on the line from the
VCCIO supply. Since Altera FPGAs require that a separate VCCIO power
each bank, these noise issues do not have any effect when crossing bank
boundaries and these guidelines do not apply. Although pad placement
rules need not be considered between I/O banks, some rules must be
considered if you are using a VREF signal in a PLLOUT bank. Note that the
signals in the PLLOUT banks share the VREF supply with neighboring I/O
banks and, therefore, must adhere to the VREF rules discussed in “VREF
Pad Placement Guidelines”.
Differential Pad Placement Guidelines
To avoid cross coupling and maintain an acceptable noise level on the
VCCIO supply, there are restrictions on the placement of single-ended I/O
pads in relation to differential pads. Use the following guidelines for
placing single-ended pads with respect to differential pads in Stratix
devices. These guidelines apply for LVDS, HyperTransport technology,
LVPECL, and PCML I/O standards. The differential pad placement
guidelines do not apply for differential HSTL and differential SSTL
output clocks since each differential output clock is essentially
implemented using two single-ended output buffers. These rules do not
apply to differential HSTL input clocks either even though the dedicated
input buffers are used. However, both differential HSTL and differential
SSTL output standards must adhere to the single-ended (VREF) pad
placement restrictions discussed in “VREF Pad Placement Guidelines”.
4–30
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
Selectable I/O Standards in Stratix & Stratix GX Devices
■
■
For flip-chip packages, there are no restrictions for placement of
single-ended input signals with respect to differential signals (see
Figure 4–21). For wire-bond packages, single ended input pads may
only be placed four or more pads away from a differential pad.
Single-ended outputs and bidirectional pads may only be placed five
or more pads away from a differential pad (see Figure 4–21),
regardless of package type.
Figure 4–21. Legal Pin Placement Note (1)
Wirebond
Input
Input, Output,
Bidirectional
Differential Pin
FlipChip
Input
Input
Input, Output,
Bidirectional
Note to Figure 4–21:
(1)
Input pads on a flip-chip packages have no restrictions.
VREF Pad Placement Guidelines
Restrictions on the placement of single-ended voltage-referenced I/O
pads with respect to VREF pads help maintain an acceptable noise level
on the VCCIO supply and to prevent output switching noise from shifting
the VREF rail. The following guidelines are for placing single-ended pads
in Stratix devices.
Input Pins
Each VREF pad supports a maximum of 40 input pads with up to 20 on
each side of the VREF pad.
Output Pins
When a voltage referenced input or bidirectional pad does not exist in a
bank, there is no limit to the number of output pads that can be
implemented in that bank. When a voltage referenced input exists, each
VREF pad supports 20 outputs for thermally enhanced FineLine BGA®
and thermally enhanced BGA cavity up packages or 15 outputs for Nonthermally enhanced cavity up and non-thermally enhanced
FineLine BGA packages.
Altera Corporation
June 2006
4–31
Stratix Device Handbook, Volume 2
I/O Pad Placement Guidelines
Bidirectional Pins
Bidirectional pads must satisfy input and output guidelines
simultaneously. If the bidirectional pads are all controlled by the same OE
and there are no other outputs or voltage referenced inputs in the bank,
then there is no case where there is a voltage referenced input active at the
same time as an output. Therefore, the output limitation does not apply.
However, since the bidirectional pads are linked to the same OE, the
bidirectional pads act as inputs at the same time. Therefore, the input
limitation of 40 input pads (20 on each side of the VREF pad) applies.
If any of the bidirectional pads are controlled by different output enables
(OE) and there are no other outputs or voltage referenced inputs in the
bank, then there may be a case where one group of bidirectional pads is
acting as inputs while another group is acting as outputs. In such cases,
apply the formulas shown in Table 4–8.
Table 4–8. Input-Only Bidirectional Pin Limitation Formulas
Package Type
Formula
Thermally enhanced FineLine BGA and <Total number of bidirectional pads> – <Total number of pads from the
thermally enhanced BGA cavity up
smallest group of pads controlled by an OE> ≤20 (per VREF pad)
Non-thermally enhanced cavity up and <Total number of bidirectional pads> – <Total number of pads from the
non-thermally enhanced FineLine BGA smallest group of pads controlled by an OE> ≤15 (per VREF pad).
Consider a thermally enhanced FineLine BGA package with eight
bidirectional pads controlled by OE1, eight bidirectional pads controlled
by OE2, and six bidirectional pads controlled by OE3. While this totals 22
bidirectional pads, it is safely allowable because there would be a
maximum of 16 outputs per VREF pad possible assuming the worst case
where OE1 and OE2 are active and OE3 is inactive. This is particularly
relevant in DDR SDRAM applications.
When at least one additional voltage referenced input and no other
outputs exist in the same VREF bank, then the bidirectional pad limitation
must simultaneously adhere to the input and output limitations. See the
following equation.
<Total number of bidirectional pads> + <Total number of input pads> ≤40 (20 on
each side of the VREF pad)
4–32
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
Selectable I/O Standards in Stratix & Stratix GX Devices
The previous equation accounts for the input limitations, but you must
apply the appropriate equation from Table 4–9 to determine the output
limitations.
Table 4–9. Bidirectional pad Limitation Formulas (Where VREF Inputs Exist)
Package Type
Formula
Thermally enhanced FineLine BGA and <Total number of bidirectional pads> ≤20 (per VREF pad)
thermally enhanced BGA cavity up
Non-thermally enhanced cavity up and <Total number of bidirectional pads> ≤15 (per VREF pad)
non-thermally enhanced FineLine BGA
When at least one additional output exists but no voltage referenced
inputs exist, apply the appropriate formula from Table 4–10.
Table 4–10. Bidirectional Pad Limitation Formulas (Where VREF Outputs Exist)
Package Type
Formula
Thermally enhanced FineLine BGA and <Total number of bidirectional pads> + <Total number of additional
thermally enhanced BGA cavity up
output pads> – <Total number of pads from the smallest group of pads
controlled by an OE> ≤20 (per VREF pad)
Non-thermally enhanced cavity up and <Total number of bidirectional pads> + <Total number of additional
non-thermally enhanced FineLine BGA output pads> – <Total number of pads from the smallest group of pads
controlled by an OE> ≤15 (per VREF pad)
When additional voltage referenced inputs and other outputs exist in the
same VREF bank, then the bidirectional pad limitation must again
simultaneously adhere to the input and output limitations. See the
following equation.
<Total number of bidirectional pads> + <Total number of input pads> ≤40 (20 on
each side of the VREF pad)
Altera Corporation
June 2006
4–33
Stratix Device Handbook, Volume 2
I/O Pad Placement Guidelines
The previous equation accounts for the input limitations, but you must
apply the appropriate equation from Table 4–9 to determine the output
limitations.
Table 4–11. Bidirectional Pad Limitation Formulas (Multiple VREF Inputs & Outputs)
Package Type
Formula
Thermally enhanced FineLine BGA and <Total number of bidirectional pads> + <Total number of additional
thermally enhanced BGA cavity up
output pads> ≤20 (per VREF pad)
non-thermally enhanced cavity up and <Total number of bidirectional pads> + <Total number of additional
non-thermally enhanced FineLine BGA output pads> ≤15 (per VREF pad)
In addition to the pad placement guidelines, use the following guidelines
when working with VREF standards:
■
■
Each bank can only have a single VCCIO voltage level and a single
VREF voltage level at a given time. Pins of different I/O standards can
share the bank if they have compatible VCCIO values (see Table 4–12
for more details).
In all cases listed above, the Quartus II software generates an error
message for illegally placed pads.
Output Enable Group Logic Option in Quartus II
The Quartus II software can check a design to make sure that the pad
placement does not violate the rules mentioned above. When the
software checks the design, if the design contains more bidirectional pins
than what is allowed, the Quartus II software returns a fitting error. When
all the bidirectional pins are either input or output but not both (for
example, in a DDR memory interface), you can use the Output Enable
Group Logic option. Turning on this option directs the Quartus II Fitter
to view the specified nodes as an output enable group. This way, the Fitter
does not violate the requirements for the maximum number of pins
driving out of a VREF bank when a voltaged-referenced input pin or
bidirectional pin is present.
In a design that implements DDR memory interface with dq, dqs and dm
pins utilized, there are two ways to enable the above logic options. You
can enable the logic options through the Assignment Editor or by adding
the following assignments to your project’s ESF file:
OPTIONS_FOR_INDIVIDUAL_NODES_ONLY
{
dq : OUTPUT_ENABLE_GROUP 1;
dqs : OUTPUT_ENABLE_GROUP 1;
4–34
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
Selectable I/O Standards in Stratix & Stratix GX Devices
dm : OUTPUT_ENABLE_GROUP 1;
}
As a result, the Quartus II Fitter does not count the bidirectional pin
potential outputs, and the number of VREF bank outputs remains in the
legal range.
Toggle Rate Logic Option in Quartus II
You should specify the pin’s output toggling rate in order to perform a
stricter pad placement check in the Quartus II software. Specify the
frequency at which a pin toggles in the Quartus II Assignment Editor.
This option is useful for adjusting the pin toggle rate in order to place
them closer to differential pins. The option directs the Quartus II Fitter
toggle-rate checking while allowing you to place a single-ended pin
closer to a differential pin.
DC Guidelines
Variables affecting the DC current draw include package type and desired
termination methods. This section provides information on each of these
variables and also shows how to calculate the DC current for pin
placement.
1
The Quartus II software automatically takes these variables into
account during compilation.
For any 10 consecutive output pads in an I/O bank, Altera recommends
a maximum current of 200 mA for thermally enhanced FineLine BGA and
thermally enhanced BGA cavity up packages and 164 mA for
non-thermally enhanced cavity up and non-thermally enhanced FineLine
BGA packages. The following equation shows the current density
limitation equation for thermally enhanced FineLine BGA and thermally
enhanced BGA cavity up packages:
pin + 9
Σ
Ipin < 200 mA
pin
The following equation shows the current density limitation equation for
non-thermally enhanced cavity up and non-thermally enhanced
FineLine BGA packages:
Altera Corporation
June 2006
4–35
Stratix Device Handbook, Volume 2
I/O Pad Placement Guidelines
pin + 9
Σ
Ipin < 164 mA
pin
Table 4–12 shows the DC current specification per pin for each I/O
standard. I/O standards not shown in the table do not exceed these
current limitations.
Table 4–12. I/O Standard DC Specification Note (1)
IPIN (mA)
Pin I/O Standard
3.3-V VCCIO
2.5-V VCCIO
1.5-V VCCIO
GTL
40
40
-
GTL+
34
34
-
SSTL-3 Class I
8
-
-
SSTL-3 Class II
16
-
-
CTT
8
-
-
SSTL-2 Class I
-
8.1
-
SSTL-2 Class II
-
16.4
-
HSTL Class I
-
-
8
HSTL Class II
-
-
16
Note to Table 4–12:
(1)
f
The current rating on a VREF pin is less than 10μA.
For more information on Altera device packaging, see the Package
Information for Stratix Devices chapter in the Stratix Device Handbook,
Volume 2.
4–36
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
Selectable I/O Standards in Stratix & Stratix GX Devices
Figure 4–22. Current Draw Limitation Guidelines
I/O Pin Sequence
of an I/O Bank
Any 10 Consecutive I/O Pins,
VCC
GND
Any 10 consecutive I/O pads cannot exceed 200 mA in thermally
enhanced FineLine BGA and thermally enhanced BGA cavity up
packages or 164 mA in non-thermally enhanced cavity up and nonthermally enhanced FineLine BGA packages.
For example, consider a case where a group of 10 consecutive pads are
configured as follows for a thermally enhanced FineLine BGA and
thermally enhanced BGA cavity up package:
■
■
■
Number of SSTL-3 Class I output pads = 3
Number of GTL+ output pads = 4
The rest of the surrounding I/O pads in the consecutive group of 10
are unused
In this case, the total current draw for these 10 consecutive I/O pads
would be:
(# of SSTL-3 Class I pads × 8 mA) +
(# of GTL+ output pads × 34 mA) = (3 × 8 mA) + (4 × 34 mA) = 160 mA
In the above example, the total current draw for all 10 consecutive I/O
pads is less than 200 mA.
Altera Corporation
June 2006
4–37
Stratix Device Handbook, Volume 2
Power Source of Various I/O Standards
Power Source of
Various I/O
Standards
For Stratix and Stratix GX devices, the I/O standards are powered by
different power sources. To determine which source powers the input
buffers, see Table 4–13. All output buffers are powered by VCCIO.
Table 4–13. The Relationships Between Various I/O Standards and the
Power Sources
I/O Standard
Quartus II
Software
Support
Power Source
2.5V/3.3V LVTTL
VCCIO
PCI/PCI-X 1.0
VCCIO
AGP
VCCIO
1.5V/1.8V
VCCIO
GTL
VCCINT
GTL+
VCCINT
SSTL
VCCINT
HSTL
VCCINT
CTT
VCCINT
LVDS
VCCINT
LVPECL
VCCINT
PCML
VCCINT
HyperTransport
VCCINT
You specify which programmable I/O standards to use for Stratix and
Stratix GX devices with the Quartus II software. This section describes
Quartus II implementation, placement, and assignment guidelines,
including
■
■
■
■
■
■
Compiler Settings
Device & Pin Options
Assign Pins
Programmable Drive Strength Settings
I/O Banks in the Floorplan View
Auto Placement & Verification
Compiler Settings
You make Compiler settings in the Compiler Settings dialog box
(Processing menu). Click the Chips & Devices tab to specify the device
family, specific device, package, pin count, and speed grade to use for
your design.
4–38
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
Selectable I/O Standards in Stratix & Stratix GX Devices
Device & Pin Options
Click Device & Pin Options in the Compiler Settings dialog box to
access the I/O pin settings. For example, in the Voltage tab you can select
a default I/O standard for all pins for the targeted device. I/O pins that
do not have a specific I/O standard assignment default this standard.
Click OK when you are done setting I/O pin options to return to the
Compiler Settings dialog box.
Assign Pins
Click Assign Pins in the Compiler Settings dialog box to view the
device’s pin settings and pin assignments (see Figure 4–23). You can view
the pin settings under Available Pins & Existing Assignments. The
listing does not include VREF pins because they are dedicated pins. The
information for each pin includes:
■
■
■
■
■
■
■
■
Number
Name
I/O Bank
I/O Standard
Type (e.g., row or column I/O and differential or control)
SignalProbe Source Name
Enabled (that is, whether SignalProbe routing is enabled or disabled
Status
Figure 4–23. Assign Pins
Altera Corporation
June 2006
4–39
Stratix Device Handbook, Volume 2
Quartus II Software Support
When you assign an I/O standard that requires a reference voltage to an
I/O pin, the Quartus II software automatically assigns VREF pins. See the
Quartus II Help for instructions on how to use an I/O standard for a pin.
Programmable Drive Strength Settings
To make programmable drive strength settings, perform the following
steps:
1.
In the Tools menu, choose Assignment Organizer.
2.
Choose the Edit specific entity & node settings for: setting, then
select the output or bidirectional pin to specify the current strength
for.
3.
In the Assignment Categories dialog box, select Options for
Individual Nodes Only.
4.
Select Click here to add a new assignment.
5.
In the Assignment dialog box, set the Name field to Current
Strength and set the Setting field to the desired, allowable value.
6.
Click Add.
7.
Click Apply, then OK.
I/O Banks in the Floorplan View
You can view the arrangement of the device I/O banks in the Floorplan
View (View menu) as shown in Figure 4–24. You can assign multiple I/O
standards to the I/O pins in any given I/O bank as long as the VCCIO of
the standards is the same. Pins that belong to the same I/O bank must use
the same VCCIO signal.
Each device I/O pin belongs to a specific, numbered I/O bank. The
Quartus II software color codes the I/O bank to which each I/O pin and
VCCIO pin belong. Turn on the Show I/O Banks option to display the I/O
bank color and the bank numbers for each pin.
4–40
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
Selectable I/O Standards in Stratix & Stratix GX Devices
Figure 4–24. Floorplan View Window
Auto Placement & Verification of Selectable I/O Standards
The Quartus II software automatically verifies the placement for all I/O
and VREF pins and performs the following actions.
■
■
■
■
■
Altera Corporation
June 2006
Automatically places I/O pins of different VREF standards without
pin assignments in separate I/O banks and enables the VREF pins of
these I/O banks.
Verifies that voltage-referenced I/O pins requiring different VREF
levels are not placed in the same bank.
Reports an error message if the current limit is exceeded for a Stratix
or Stratix GX power bank, as determined by the equation
documented in “DC Guidelines” on page 4–35.
Reserves the unused high-speed differential I/O channels and
regular user I/O pins in the high-speed differential I/O banks when
any of the high-speed differential I/O channels are being used.
Automatically assigns VREF pins and I/O pins such that the current
requirements are met and I/O standards are placed properly.
4–41
Stratix Device Handbook, Volume 2
Conclusion
Conclusion
Stratix and Stratix GX devices provide the I/O capabilities to allow you
to work with current and emerging I/O standards and requirements.
Today’s complex designs demand increased flexibility to work with the
wide variety of available I/O standards and to simplify board design.
With Stratix and Stratix GX device features, such as hot socketing and
differential on-chip termination, you can reduce board design interface
costs and increase your development flexibility.
More
Information
For more information, see the following sources:
■
■
■
■
References
The Stratix Device Family Data Sheet section in the Stratix Device
Handbook, Volume 1
The Stratix GX Device Family Data Sheet section of the Stratix GX
Device Handbook, Volume 1
The High-Speed Differential I/O Interfaces in Stratix Devices chapter
AN 224: High-Speed Board Layout Guidelines
For more information, see the following references:
■
■
■
■
■
■
■
■
Stub Series Terminated Logic for 2.5-V (SSTL-2), JESD8-9B,
Electronic Industries Association, December 2000.
High-Speed Transceiver Logic (HSTL) – A 1.5-V Output Buffer
Supply Voltage Based Interface Standard for Digital Integrated
Circuits, EIA/JESD8-6, Electronic Industries Association, August
1995.
1.5-V +/- 0.1 V (Normal Range) and 0.9 V – 1.6 V (Wide Range)
Power Supply Voltage and Interface Standard for Non-terminated
Digital Integrated Circuits, JESD8-11, Electronic Industries
Association, October 2000.
1.8-V +/- 0.15 V (Normal Range) and 1.2 V – 1.95 V (Wide Range)
Power Supply Voltage and Interface Standard for Non-terminated
Digital Integrated Circuits, JESD8-7, Electronic Industries
Association, February 1997.
Center-Tap-Terminated (CTT) Low-Level, High-Speed Interface
Standard for Digital Integrated Circuits, JESD8-9A, Electronic
Industries Association, November 1993.
2.5-V +/- 0.2V (Normal Range) and 1.8-V to 2.7V (Wide Range)
Power Supply Voltage and Interface Standard for Non-terminated
Digital Integrated Circuits, JESD8-5, Electronic Industries
Association, October 1995.
Interface Standard for Nominal 3V/ 3.3-V Supply Digital Integrated
Circuits, JESD8-B, Electronic Industries Association, September 1999.
Gunning Transceiver Logic (GTL) Low-Level, High-Speed Interface
Standard for Digital Integrated Circuits, JESD8-3, Electronic
Industries Association, November 1993.
4–42
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
Selectable I/O Standards in Stratix & Stratix GX Devices
■
■
■
■
■
■
■
■
Altera Corporation
June 2006
Accelerated Graphics Port Interface Specification 2.0, Intel
Corporation.
Stub Series Terminated Logic for 1.8-V (SSTL-18), Preliminary JC42.3,
Electronic Industries Association.
PCI Local Bus Specification, Revision 2.2, PCI Special Interest Group,
December 1998.
PCI-X Local Bus Specification, Revision 1.0a, PCI Special Interest
Group.
UTOPIA Level 4, AF-PHY-0144.001, ATM Technical Committee.
POS-PHY Level 4: SPI-4, OIF-SPI4-02.0, Optical Internetworking
Forum.
POS-PHY Level 4: SFI-4, OIF-SFI4-01.0, Optical Internetworking
Forum.
Electrical Characteristics of Low Voltage Differential Signaling
(LVDS) Interface Circuits, ANSI/TIA/EIA-644, American National
Standards Institute/Telecommunications Industry/Electronic
Industries Association, October 1995.
4–43
Stratix Device Handbook, Volume 2
References
4–44
Stratix Device Handbook, Volume 2
Altera Corporation
June 2006
5. High-Speed Differential I/O
Interfaces in Stratix Devices
S52005-3.2
Introduction
To achieve high data transfer rates, Stratix® devices support TrueLVDSTM differential I/O interfaces which have dedicated
serializer/deserializer (SERDES) circuitry for each differential I/O pair.
Stratix SERDES circuitry transmits and receives up to 840 megabits per
second (Mbps) per channel. The differential I/O interfaces in Stratix
devices support many high-speed I/O standards, such as LVDS,
LVPECL, PCML, and HyperTransportTM technology. Stratix device highspeed modules are designed to provide solutions for many leading
protocols such as SPI-4 Phase 2, SFI-4, 10G Ethernet XSBI, RapidIO,
HyperTransport technology, and UTOPIA-4.
The SERDES transmitter is designed to serialize 4-, 7-, 8-, or 10-bit wide
words and transmit them across either a cable or printed circuit board
(PCB). The SERDES receiver takes the serialized data and reconstructs the
bits into a 4-, 7-, 8-, or 10-bit-wide parallel word. The SERDES contains the
necessary high-frequency circuitry, multiplexer, demultiplexer, clock,
and data manipulation circuitry. You can use double data rate I/O
(DDRIO) circuitry to transmit or receive differential data in by-one (×1)
or by-two (×2) modes.
1
Contact Altera Applications for more information on other B
values that the Stratix devices support and using ×7-mode in the
Quartus® II software. Stratix devices currently only support
B = 1 and B = 7 in ×7 mode.
This chapter describes the high-speed differential I/O capabilities of
Stratix programmable logic devices (PLDs) and provides guidelines for
their optimal use. You should use this document in conjunction with the
Stratix Device Family Data Sheet section of the Stratix Device Handbook,
Volume 1. Consideration of the critical issues of controlled impedance of
traces and connectors, differential routing, termination techniques, and
DC balance gets the best performance from the device. Therefore, an
elementary knowledge of high-speed clock-forwarding techniques is also
helpful.
Stratix I/O Banks
Altera Corporation
July 2005
Stratix devices contain eight I/O banks, as shown in Figure 5–1. The two
I/O banks on each side contain circuitry to support high-speed LVDS,
LVPECL, PCML, HSTL Class I and II, SSTL-2 Class I and II, and
HyperTransport inputs and outputs.
5–1
Stratix I/O Banks
Figure 5–1. Stratix I/O Banks Notes (1), (2), (3)
DQS5T
9
DQS4T
PLL11
(5)
DQS1T
DQS0T
10
Bank 4
LVDS, LVPECL, 3.3-V PCML,
and HyperTransport I/O Block
and Regular I/O Pins (4)
(5)
I/O Banks 1, 2, 5, and 6 Support All
Single-Ended I/O Standards Except
Differential HSTL Output Clocks,
Differential SSTL-2 Output Clocks,
HSTL Class II, GTL, SSTL-18 Class II,
PCI, PCI-X 1.0, and AGP 1×/2×
PLL2
Bank 1
DQS2T
I/O Banks 3, 4, 9 & 10 Support
All Single-Ended I/O Standards
PLL1
Bank 8
PLL3
DQS8B
DQS7B
DQS6B
DQS5B
(5)
LVDS, LVPECL, 3.3-V PCML,
and HyperTransport I/O Block
and Regular I/O Pins (4)
11
VREF5B8 VREF4B8 VREF3B8 VREF2B8 VREF1B8
DQS9B
PLL4
I/O Banks 7, 8, 11 & 12 Support
All Single-Ended I/O Standards
(5)
LVDS, LVPECL, 3.3-V PCML,
and HyperTransport I/O Block
and Regular I/O Pins (4)
PLL8
DQS3T
VREF1B4 VREF2B4 VREF3B4 VREF4B4 VREF5B4 PLL10
LVDS, LVPECL, 3.3-V PCML,
and HyperTransport I/O Block
and Regular I/O Pins (4)
Bank 2
VREF1B2 VREF2B2 VREF3B2 VREF4B2
Bank 3
VREF1B1 VREF2B1 VREF3B1 VREF4B1
PLL5
12
PLL6
Bank 5
DQS6T
VREF4B5 VREF3B5 VREF2B5 VREF1B5
DQS7T
Bank 6
DQS8T
VREF1B3 VREF2B3 VREF3B3 VREF4B3 VREF5B3
VREF4B6 VREF3B6 VREF2B6 VREF1B6
DQS9T
PLL7
Bank 7
PLL12
VREF5B7 VREF4B7 VREF3B7 VREF2B7 VREF1B7
DQS4B
DQS3B
DQS2B
DQS1B
PLL9
DQS0B
Notes to Figure 5–1:
(1)
(2)
(3)
(4)
(5)
Figure 5–1 is a top view of the Stratix silicon die, which corresponds to a top-down view of non-flip-chip packages
and a bottom-up view of flip-chip packages.
Figure 5–1 is a graphic representation only. See the pin list and the Quartus II software for exact locations.
Banks 9 through 12 are enhanced PLL external clock output banks.
If the high-speed differential I/O pins are not used for high-speed differential signaling, they can support all of the
I/O standards except HSTL Class I and II, GTL, SSTL-18 Class II, PCI, PCI-X 1.0, and AGP 1× /2× .
See “Differential Pad Placement Guidelines” on page 4–30. You can only place single-ended output/bidirectional
pads five or more pads away from a differential pad. Use the Show Pads view in the Quartus II Floorplan Editor to
locate these pads. The Quartus II software gives an error message for illegal output or bidirectional pin placement
next to a high-speed differential I/O pin.
Stratix Differential I/O Standards
Stratix devices provide a multi-protocol interface that allows
communication between a variety of I/O standards, including LVDS,
HyperTransport technology, LVPECL, PCML, HSTL Class I and II, and
5–2
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
SSTL-2 Class I and II. This feature makes the Stratix device family ideal
for applications that require multiple I/O standards, such as a protocol
translator.
f
For more information on termination for Stratix I/O standards, see
“Differential I/O Termination” on page 5–46.
Figure 5–2 compares the voltage levels between differential I/O
standards supported in all the Stratix devices.
Figure 5–2. Differential I/O Standards Supported by Stratix Devices
4.0
3.3 V
PCML
3.0 V
3.0
2.1 V
Voltage
(V)
2.0
LVPECL
1.7 V
1.4 V
LVDS
1.0
1.0 V
0.9 V
HyperTransport
0.3 V
0.0
Technology
Altera Corporation
July 2005
5–3
Stratix Device Handbook, Volume 2
Stratix I/O Banks
LVDS
The LVDS I/O standard is a differential high-speed, low-voltage swing,
low-power, general-purpose I/O interface standard requiring a 3.3-V
VCCIO. This standard is used in applications requiring high-bandwidth
data transfer, backplane drivers, and clock distribution. The
ANSI/TIA/EIA-644 standard specifies LVDS transmitters and receivers
capable of operating at recommended maximum data signaling rates of
655 Mbps. However, devices can operate at slower speeds if needed, and
there is a theoretical maximum of 1.923 Gbps. Stratix devices meet the
ANSI/TIA/EIA-644 standard.
Due to the low voltage swing of the LVDS I/O standard, the
electromagnetic interference (EMI) effects are much smaller than CMOS,
transistor-to-transistor logic (TTL), and PECL. This low EMI makes LVDS
ideal for applications with low EMI requirements or noise immunity
requirements. The LVDS standard specifies a differential output voltage
range of 0.25 V × VOD ≤ 0.45 V. The LVDS standard does not require an
input reference voltage, however, it does require a 100-Ω termination
resistor between the two signals at the input buffer. Stratix devices
include an optional differential termination resistor within the device. See
Section I, Stratix Device Family Data Sheet of the Stratix Device Handbook,
Volume 1 for the LVDS parameters.
HyperTransport Technology
The HyperTransport technology I/O standard is a differential highspeed, high-performance I/O interface standard requiring a 2.5-V
VCCIO. This standard is used in applications such as high-performance
networking, telecommunications, embedded systems, consumer
electronics, and Internet connectivity devices. The HyperTransport
technology I/O standard is a point-to-point standard in which each
HyperTransport technology bus consists of two point-to-point
unidirectional links. Each link is 2 to 32 bits. See the Stratix Device Family
Data Sheet section of the Stratix Device Handbook, Volume 1 for the
HyperTransport parameters.
LVPECL
The LVPECL I/O standard is a differential interface standard requiring a
3.3-V VCCIO. The standard is used in applications involving video
graphics, telecommunications, data communications, and clock
distribution. The high-speed, low-voltage swing LVPECL I/O standard
uses a positive power supply and is similar to LVDS, however, LVPECL
has a larger differential output voltage swing than LVDS. See the Stratix
Device Family Data Sheet section of the Stratix Device Handbook, Volume 1
for the LVPECL signaling characteristics.
5–4
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
PCML
The PCML I/O standard is a differential high-speed, low-power I/O
interface standard used in applications such as networking and
telecommunications. The standard requires a 3.3-V VCCIO. The PCML I/O
standard achieves better performance and consumes less power than the
LVPECL I/O standard. The PCML standard is similar to LVPECL, but
PCML has a reduced voltage swing, which allows for a faster switching
time and lower power consumption.See the Stratix Device Family Data
Sheet section of the Stratix Device Handbook, Volume 1 for the PCML
signaling characteristics.
Differential HSTL (Class I & II)
The differential HSTL I/O standard is used for applications designed to
operate in the 0.0- to 1.5-V HSTL logic switching range such as quad data
rate (QDR) memory clock interfaces. The differential HSTL specification
is the same as the single ended HSTL specification. The standard specifies
an input voltage range of – 0.3 V ≤ VI ≤ VCCIO + 0.3 V. The differential
HSTL I/O standard is only available on the input and output clocks. See
the Stratix Device Family Data Sheet section of the Stratix Device Handbook,
Volume 1 for the HSTL signaling characteristics
Differential SSTL-2 (Class I & II)
The differential SSTL-2 I/O standard is a 2.5-V memory bus standard
used for applications such as high-speed double data rate (DDR) SDRAM
interfaces. This standard defines the input and output specifications for
devices that operate in the SSTL-2 logic switching range of 0.0 to 2.5 V.
This standard improves operation in conditions where a bus must be
isolated from large stubs. The SSTL-2 standard specifies an input voltage
range of – 0.3 V ≤ VI ≤ VCCIO + 0.3 V. Stratix devices support both input
and output levels. The differential SSTL-2 I/O standard is only available
on output clocks. See the Stratix Device Family Data Sheet section of the
Stratix Device Handbook, Volume 1 for the SSTL-2 signaling characteristics.
Stratix Differential I/O Pin Location
The differential I/O pins are located on the I/O banks on the right and
left side of the Stratix device. Table 5–1 shows the location of the Stratix
device high-speed differential I/O buffers. When the I/O pins in the I/O
banks that support differential I/O standards are not used for high-speed
Altera Corporation
July 2005
5–5
Stratix Device Handbook, Volume 2
Principles of SERDES Operation
signaling, you can configure them as any of the other supported I/O
standards. DDRIO capabilities are detailed in “SERDES Bypass DDR
Differential Signaling” on page 5–42.
Table 5–1. I/O Pin Locations on Each Side of Stratix Devices
Differential Input
Differential Output
DDRIO
Left
Device Side (1)
v
v
v
Right
v
v
v
Top
v
Bottom
v
Note to Table 5–1:
(1)
Principles of
SERDES
Operation
Device sides are relative to pin A1 in the upper left corner of the device (top view
of the package).
Stratix devices support source-synchronous differential signaling up to
840 Mbps. Serial data is transmitted and received along with a lowfrequency clock. The PLL can multiply the incoming low-frequency clock
by a factor of 1 to 10. The SERDES factor J can be 4, 7, 8, or 10 and does not
have to equal the clock multiplication value. ×1 and ×2 operation is also
possible by bypassing the SERDES; it is explained in “SERDES Bypass
DDR Differential Interface Review” on page 5–42.
On the receiver side, the high-frequency clock generated by the PLL shifts
the serial data through a shift register (also called deserializer). The
parallel data is clocked out to the logic array synchronized with the lowfrequency clock. On the transmitter side, the parallel data from the logic
array is first clocked into a parallel-in, serial-out shift register
synchronized with the low-frequency clock and then transmitted out by
the output buffers.
There are four dedicated fast PLLs in EP1S10 to EP1S25 devices, and eight
in EP1S30 to EP1S80 devices. These PLLs are used for the SERDES
operations as well as general-purpose use.
The differential channels and the high-speed PLL layout in Stratix
devices are described in the “Differential I/O Interface & Fast PLLs”
section on page 5–16.
5–6
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Stratix Differential I/O Receiver Operation
You can configure any of the Stratix differential input channels as a
receiver channel (see Figure 5–3). The differential receiver deserializes
the incoming high-speed data. The input shift register continuously
clocks the incoming data on the negative transition of the high-frequency
clock generated by the PLL clock (×W).
The data in the serial shift register is shifted into a parallel register by the
RXLOADEN signal generated by the fast PLL counter circuitry on the third
falling edge of the high-frequency clock. However, you can select which
falling edge of the high frequency clock loads the data into the parallel
register, using the data-realignment circuit. For more information on the
data-realignment circuit, see “Data Realignment Principles of Operation”
on page 5–25.
In normal mode, the enable signal RXLOADEN loads the parallel data into
the next parallel register on the second rising edge of the low-frequency
clock. You can also load data to the parallel register through the
TXLOADEN signal when using the data-realignment circuit.
Figure 5–3 shows the block diagram of a single SERDES receiver channel.
Figure 5–4 shows the timing relationship between the data and clocks in
Stratix devices in ×10 mode. W is the low-frequency multiplier and J is
data parallelization division factor.
Altera Corporation
July 2005
5–7
Stratix Device Handbook, Volume 2
Principles of SERDES Operation
Figure 5–3. Stratix High-Speed Interface Deserialized in ×10 Mode
Receiver Circuit
Serial Shift
Registers
RXIN+
RXIN−
Parallel
Registers
PD0
PD1
PD2
PD3
PD4
PD5
PD6
PD7
PD8
PD9
PD0
PD1
PD2
PD3
PD4
PD5
PD6
PD7
PD8
PD9
×W
RXCLKIN+
RXCLKIN−
Parallel
Registers
PD0
PD1
PD2
PD3
PD4
PD5
PD6
PD7
PD8
PD9
Stratix
Logic Array
×W/J (1)
Fast RXLOADEN
PLL (2)
TXLOADEN
Notes to Figure 5–3:
(1)
(2)
W = 1, 2, 4, 7, 8, or 10.
J = 4, 7, 8, or 10.
W does not have to equal J. When J = 1 or 2, the deserializer is bypassed. When J = 2, the device uses DDRIO registers.
This figure does not show additional circuitry for clock or data manipulation.
Figure 5–4. Receiver Timing Diagram
Internal ×1 clock
Internal ×10 clock
RXLOADEN
Receiver
data input
n–1
n–0
9
8
7
6
5
4
3
2
1
0
n–1
n–0
9
8
7
6
5
4
3
2
1
0
Internal ×1 clock
Internal ×10 clock
RXLOADEN
Receiver
data input
5–8
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Stratix Differential I/O Transmitter Operation
You can configure any of the Stratix differential output channels as a
transmitter channel. The differential transmitter is used to serialize
outbound parallel data.
The logic array sends parallel data to the SERDES transmitter circuit
when the TXLOADEN signal is asserted. This signal is generated by the
high-speed counter circuitry of the logic array low-frequency clock’s
rising edge. The data is then transferred from the parallel register into the
serial shift register by the TXLOADEN signal on the third rising edge of the
high-frequency clock.
Figure 5–5 shows the block diagram of a single SERDES transmitter
channel and Figure 5–6 shows the timing relationship between the data
and clocks in Stratix devices in ×10 mode. W is the low-frequency
multiplier and J is the data parallelization division factor.
Figure 5–5. Stratix High-Speed Interface Serialized in ×10 Mode
Transmitter Circuit
Stratix
Logic Array
PD9
PD8
PD7
PD6
PD5
PD4
PD3
PD2
PD1
PD0
Parallel
Register
PD9
PD8
PD7
PD6
PD5
PD4
PD3
PD2
PD1
PD0
Serial
Register
TXOUT+
TXOUT−
×W
Fast
PLL
Altera Corporation
July 2005
TXLOADEN
5–9
Stratix Device Handbook, Volume 2
Principles of SERDES Operation
Figure 5–6. Transmitter Timing Diagram
Internal ×1 clock
Internal ×10 clock
TXLOADEN
Receiver
data input
n–1
n–0
9
8
7
6
5
4
3
2
1
0
Transmitter Clock Output
Different applications and protocols call for various clocking schemes.
Some applications require you to center-align the rising or falling clock
edge with the data. Other applications require a divide version of the
transmitted clock, or the clock and data to be at the same high-speed
frequency. The Stratix device transmitter clock output is versatile and
easily programmed for all such applications.
Stratix devices transmit data using the source-synchronous scheme,
where the clock is transmitted along with the serialized data to the
receiving device. Unlike APEXTM 20KE and APEX II devices, Stratix
devices do not have a fixed transmitter clock output pin. The Altera®
Quartus II software generates the transmitter clock output by using a fast
clock to drive a transmitter dataout channel. Therefore, you can place
the transmitter clock pair close to the data channels, reducing clock-todata skew and increasing system margins. This approach is more flexible,
as any channel can drive a clock, not just specially designated clock pins.
Divided-Down Transmitter Clock Output
You can divide down the high-frequency clock by 2, 4, 8, or 10, depending
on the system requirements. The various options allow Stratix devices to
accommodate many different types of protocols. The divided-down clock
is generated by an additional transmitting data channel.
5–10
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Table 5–2 shows the divided-down version of the high-frequency clock
and the selected serialization factor J (described in pervious sections). The
Quartus II software automatically generates the data input to the
additional transmitter data channel.
Table 5–2. Differential Transmitter Output Clock Division
J
Data Input
Output Clock Divided By (1)
4
1010
2
4
0011
4
8
10101010
2
8
00110011
4
8
11000011
8
10
1010101010
2
10
1110000011
10
Note to Table 5–2:
(1)
This value is usually referred to as B.
Center-Aligned Transmitter Clock Output
A negative-edge-triggered D flipflop (DFF) register is located between
the serial register of each data channel and its output buffer, as show in
Figure 5–7. The negative-edge-triggered DFF register is used when
center-aligned data is required. For center alignment, the DFF only shifts
the output from the channel used as the transmitter clock out. The
transmitter data channels bypass the negative-edge DFF. When you use
the DFF register, the data is transmitted at the negative edge of the
multiplied clock. This delays the transmitted clock output relative to the
data channels by half the multiplied clock cycle. This is used for
HyperTransport technology, but can also be used for any interface
requiring center alignment.
Altera Corporation
July 2005
5–11
Stratix Device Handbook, Volume 2
Principles of SERDES Operation
Figure 5–7. Stratix Programmable Transmitter Clock
Transmitter Circuit
Stratix
Logic Array
PD9
PD8
PD7
PD6
PD5
PD4
PD3
PD2
PD1
PD0
Parallel
Register
PD9
PD8
PD7
PD6
PD5
PD4
PD3
PD2
PD1
PD0
Serial
Register
TXOUT+
TXOUT−
×W
Fast
PLL
TXLOADEN
SDR Transmitter Clock Output
You can route the high-frequency clock internally generated by the PLL
out as a transmitter clock output on any of the differential channels. The
high-frequency clock output allows Stratix devices to support
applications that require a 1-to-1 relationship between the clock and data.
The path of the high-speed clock is shown in Figure 5–8. A programmable
inverter allows you to drive the signal out on either the negative edge of
the clock or 180º out of phase with the streaming data.
5–12
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Figure 5–8. High-Speed 1-to-1 Transmitter Clock Output
Transmitter Circuit
Stratix
Logic Array
PD9
PD8
PD7
PD6
PD5
PD4
PD3
PD2
PD1
PD0
Parallel
Register
PD9
PD8
PD7
PD6
PD5
PD4
PD3
PD2
PD1
PD0
Serial
Register
Inverter
TXOUT+
TXOUT−
×W
Fast
PLL (1)
TXLOADEN
Note to Figure 5–8:
(1)
This figure does not show additional circuitry for clock or data manipulation.
Using SERDES
to Implement
DDR
Some designs require a 2-to-1 data-to-clock ratio. These systems are
usually based on Rapid I/O, SPI-4 Phase 2 (POS_PHY Level 4), or
HyperTransport interfaces, and support various data rates. Stratix
devices meet this requirement for such applications by providing a
variable clock division factor. The SERDES clock division factor is set to 2
for double data rate (DDR).
An additional differential channel (as described in “Transmitter Clock
Output” on page 5–10) is automatically configured to produce the
transmitter clock output signal with half the frequency of the data.
For example, when a system is required to transmit 6.4 Gbps with a
2-to-1 clock-to-data ratio, program the SERDES with eight high-speed
channels running at 800 Mbps each. When you set the output clock
division factor (2 for this example), the Quartus II software automatically
assigns a ninth channel as the transmitter clock output. You can edge- or
center-align the transmitter clock by selecting the default PLL phase or
selecting the negative-edge transmitter clock output. On the receiver side,
the clock signal is connected to the receiver PLL's clock.
The multiplication factor W is also calculated automatically. The data rate
divides by the input clock frequency to calculate the W factor. The
deserialization factor (J) may be 4, 7, 8, or 10.
Altera Corporation
July 2005
5–13
Stratix Device Handbook, Volume 2
Using SERDES to Implement SDR
Figure 5–9 shows a DDR clock-to-data timing relationship with the clock
center-aligned with respect to data. Figure 5–10 shows the connection
between the receiver and transmitter circuits.
Figure 5–9. DDR Clock-to-Data Relationship
inclock
DDR
XX
B0
A0
B1
A1
B2
A2
B3
A3
Figure 5–10. DDR Receiver & Transmitter Circuit Connection
Stratix SERDES DDR Receiver
rx_d[0]
Channel
0
Serial-to-Parallel
Register
Stratix SERDES DDR Transmitter
Parallel
Register
8
8
Parallel
Register
Parallel-to-Serial
Register
Parallel
Register
Parallel-to-Serial
Register
Parallel
Register
Parallel-to-Serial
Register
Channel
0
tx_d[0]
data rate = 800 Mbps
Stratix
Logic
Array
rx_d[15]
data rate = 800 Mbps
Channel
15
Serial-to-Parallel
Register
Parallel
Register
rxclk
LVDS PLL
8
8
input clock × W
400 MHz
8
rxloadena
Channel
15
Channel
16
÷2
txclk_out
800 Mbps
txclk_out
400 MHz
txloaden
LVDS PLL
input clock × W
txclk_in
100 MHz
Using SERDES
to Implement
SDR
Stratix devices support systems based on single data rate (SDR)
operations applications by allowing you to directly transmit out the
multiplied clock (as described in “SDR Transmitter Clock Output” on
page 5–12). These systems are usually based on Utopia-4, SFI-4, or XSBI
interfaces, and support various data rates.
An additional differential channel is automatically configured to produce
the transmitter clock output signal and is transmitted along with the data.
For example, when a system is required to transmit 10 Gbps with a 1-to1 clock-to-data ratio, program the SERDES with sixteen high-speed
channels running at 624 Mbps each. The Quartus II software
5–14
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
automatically assigns a seventeenth channel as the transmitter clock
output. You can edge- or center-align the transmitter clock output by
selecting the default PLL phase or selecting the 90° phase of the PLL
output. On the receiver side, the clock signal is connected to the receiver
PLL's clock input, and you can assign identical clock-to-data alignment.
The multiplication factor W is calculated automatically. The data rate is
dividing by the input clock frequency to calculate the W factor. The
deserialization factor J may be 4, 7, 8, or 10.
Figure 5–11 shows an SDR clock-to-data timing relationship, with clock
center aligned with respect to data. Figure 5–12 shows the connection
between the receiver and transmitter circuits.
Figure 5–11. SDR Clock-to-Data Relationship
inclock
SDR
XX
B0
B1
B2
B3
Figure 5–12. SDR Receiver & Transmitter Circuit Connection
Stratix SERDES SDR Receiver
rx_d[0]
Channel
0
Serial-to-Parallel
Register
Stratix SERDES SDR Transmitter
Parallel
Register
8
8
Parallel
Register
Parallel-to-Serial
Register
Parallel
Register
Parallel-to-Serial
Register
Channel
0
tx_d[0]
data rate = 624 Mbps
Stratix
Logic
Array
rx_d[15]
data rate = 624 Mbps
Channel
15
Serial-to-Parallel
Register
Parallel
Register
8
8
input clock × W
Channel
15
Channel
16
txloaden
tx_d[15]
txclk_out
624 MHz
rxclk
624 MHz
LVDS PLL
rxloaden
LVDS PLL
input clock × W
txclk_in
624 MHz
Altera Corporation
July 2005
5–15
Stratix Device Handbook, Volume 2
Differential I/O Interface & Fast PLLs
Differential I/O
Interface & Fast
PLLs
Stratix devices provide 16 dedicated global clocks, 8 dedicated fast
regional I/O pins, and up to 16 regional clocks (four per device quadrant)
that are fed from the dedicated global clock pins or PLL outputs. The 16
dedicated global clocks are driven either by global clock input pins that
support all I/O standards or from enhanced and fast PLL outputs.
Stratix devices use the fast PLLs to implement clock multiplication and
division to support the SERDES circuitry. The input clock is either
multiplied by the W feedback factor and/or divided by the J factor. The
resulting clocks are distributed to SERDES, local, or global clock lines.
Fast PLLs are placed in the center of the left and right sides for EP1S10 to
EP1S25 devices. For EP1S30 to EP1S80 devices, fast PLLs are placed in the
center of the left and right sides, as well as the device corners (see
Figure 5–13). These fast PLLs drive a dedicated clock network to the
SERDES in the rows above and below or top and bottom of the device as
shown in Figure 5–13.
5–16
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Figure 5–13. Stratix Fast PLL Positions & Clock Naming Convention Note (1)
CLK[15..12]
5
11
FPLLCLK0
7
10
FPLLCLK3
CLK[3..0]
1
2
4
3
CLK[11..8]
8
9
FPLLCLK2
PLLs
FPLLCLK1
6
12
CLK[7..4]
Notes to Figure 5–13:
(1)
(2)
Dedicated clock input pins on the right and left sides do not support PCI or PCI-X 1.0.
PLLs 7, 8, 9, and 10 are not available on the EP1S30 device in the 780-pin FineLine BGA® package.
Altera Corporation
July 2005
5–17
Stratix Device Handbook, Volume 2
Differential I/O Interface & Fast PLLs
Clock Input & Fast PLL Output Relationship
Table 5–3 summarizes the PLL interface to the input clocks and the enable
signal (ENA). Table 5–4 summarizes the clock networks each fast PLL can
connect to across all Stratix family devices.
Table 5–3. Fast PLL Clock Inputs (Including Feedback Clocks) & Enables Note (1)
All Stratix Devices
EP1S30 to EP1S80 Devices Only
Input Pin
PLL 1
CLK0 (2)
v
CLK1
v
PLL 2
PLL 3
PLL 4
PLL 7
PLL 8
PLL 9
PLL 10
v (3)
CLK2 (2)
v
CLK3
v
v (3)
CLK4
CLK5
CLK6
CLK7
CLK8
v
CLK9 (2)
v
v (3)
CLK10
v
CLK11 (2)
v
v (3)
CLK12
CLK13
CLK14
CLK15
v
ENA
v
FPLL7CLK
FPLL8CLK
FPLL9CLK
FPLL10CLK
v
v
v
v
v
v
v
v
v
v
Notes to Table 5–3:
(1)
(2)
(3)
PLLs 5, 6, 11, and 12 are not fast PLLs.
Clock pins CLK0, CLK2, CLK9, CLK11, and pins FPLL[7..10]CLK do not support differential on-chip
termination.
Either a FPLLCLK pin or a CLK pin can drive the corner fast PLLs (PLL7, PLL8, PLL9, and PLL10) when used for
general purpose. CLK pins cannot drive these fast PLLs in high-speed differential I/O mode.
5–18
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Table 5–4. Fast PLL Relationship with Stratix Clock Networks (Part 1 of 2) Notes (1), (2)
All Stratix Devices
EP1S30 to EP1S80 Devices Only
Output Signal
PLL 1
GCLK0
v
GCLK1
v
PLL 2
GCLK2
v
GCLK3
v
PLL 3
GCLK4
v
GCLK9
v
PLL 4
GCLK10
v
GCLK11
v
PLL 7
PLL 8
RCLK1
v
v
v
RCLK2
v
v
v
RCLK3
v
v
v
RCLK4
v
v
v
PLL 9
PLL 10
RCLK9
v
v
v
RCLK10
v
v
v
RCLK11
v
v
v
RCLK12
v
v
v
DIFFIOCLK1
v
DIFFIOCLK2
v
DIFFIOCLK3
v
DIFFIOCLK4
v
DIFFIOCLK5
v
DIFFIOCLK6
v
DIFFIOCLK7
v
DIFFIOCLK8
v
DIFFIOCLK9
v
DIFFIOCLK10
v
DIFFIOCLK11
v
DIFFIOCLK12
v
DIFFIOCLK13
Altera Corporation
July 2005
v
5–19
Stratix Device Handbook, Volume 2
Differential I/O Interface & Fast PLLs
Table 5–4. Fast PLL Relationship with Stratix Clock Networks (Part 2 of 2) Notes (1), (2)
All Stratix Devices
EP1S30 to EP1S80 Devices Only
Output Signal
PLL 1
PLL 2
PLL 3
PLL 4
PLL 7
PLL 8
PLL 9
PLL 10
v
DIFFIOCLK14
DIFFIOCLK15
v
DIFFIOCLK16
v
Notes to Table 5–4:
(1)
(2)
PLLs 5, 6, 11, and 12 are not fast PLLs.
The input clock for PLLs used to clock receiver the rx_inclock port on the altlvds_rx megafunction must be
driven by a dedicated clock pin (CLK[3..0] and CLK[8..11]) or the corner pins that clock the corner PLLs
(FPLL[10..7]CLK).
Fast PLL Specifications
You can drive the fast PLLs by an external pin or any one of the sectional
clocks [21..0]. You can connect the clock input directly to local or global
clock lines, as shown in Figure 5–14. You cannot use the sectional-clock
inputs to the fast PLL’s input multiplexer for the receiver PLL. You can
only use the sectional clock inputs in the transmitter only mode or as a
general purpose PLL.
5–20
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Figure 5–14. Fast PLL Block Diagram
Post-Scale
Counters
DIFFIOCLK1 (1)
÷k
Regional clock
TXLOADEN (2)
VCO Phase Selection
Selectable at each PLL
Output Port
RXLOADEN (2)
÷v
Global or
regional clock
Clock Input
Phase
Frequency
Detector
Charge
Pump
8
Loop
Filter
Regional clock
DIFFIOCLK2 (1)
VCO
÷l
Global or
regional clock
rxclkin
÷m
Notes to Figure 5–14:
(1)
(2)
In high-speed differential I/O mode, the high-speed PLL clock feeds the SERDES. Stratix devices only support one
rate of data transfer per fast PLL in high-speed differential I/O mode.
Control signal for high-speed differential I/O SERDES.
You can multiply the input clock by a factor of 1 to 16. The multiplied
clock is used for high-speed serialization or deserialization operations.
Fast PLL specifications are shown in the Stratix Device Family Data Sheet
section of the Stratix Device Handbook, Volume 1. The voltage controlled
oscillators (VCOs) are designed to operate within the frequency range of
300 to 840 MHz, to provide data rates of up to 840 Mbps.
High-Speed Phase Adjust
There are eight phases of the multiplied clock at the PLL output, each
delayed by 45° from the previous clock and synchronized with the
original clock. The three multiplexers (shown in Figure 5–14) select one of
the delayed, multiplied clocks. The PLL output drives the three counters
k, v, and l. You can program the three individual post scale counters (k, v,
and l) independently for division ratio or phase. The selected PLL output
is used for the serialization or deserialization process in SERDES.
Altera Corporation
July 2005
5–21
Stratix Device Handbook, Volume 2
Differential I/O Interface & Fast PLLs
Counter Circuitry
The multiplied clocks bypass the counter taps k and v to directly feed the
SERDES serial registers. These two taps also feed to the quadrant local
clock network and the dedicated RXLOADENA or TXLOADENA pins, as
shown in Figure 5–15. Both k and v are utilized simultaneously during the
data-realignment procedure. When the design does not use the data
realignment, both TXLOADEN and RXLOADEN pins use a single counter.
Figure 5–15. Fast PLL Connection to Logic Array
Counter Circuitry
Post-Scale
Counters
VCO Phase Selection
Selectable at each PLL
Output Port
÷k
CLK1 SERDES
Circuitry
×1 CLK1 to logic array
or local clocks
TXLOADEN
RXLOADEN
8
÷v
PLL Output
Clock
Distribution
Circuitry
×1 CLK2 to logic array
or local clocks
CLK2 SERDES
Circuitry
÷l
Regional clock
clkin
The Stratix device fast PLL has another GCLK connection for generalpurpose applications. The third tap l feeds the quadrant local clock as
well as the global clock network. You can use the l counter's multiplexer
for applications requiring the device to connect the incoming clock
directly to the local or global clocks. You can program the multiplexer to
connect the RXCLKIN signal directly to the local or global clock lines.
Figure 5–15 shows the connection between the incoming clock, the l tap,
and the local or global clock lines.
The differential clock selection is made per differential bank. Since the
length of the clock tree limits the performance, each fast PLL should drive
only one differential bank.
5–22
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Fast PLL SERDES Channel Support
The Quartus II MegaWizard Plug-In Manager only allows you to
implement up to 20 receiver or 20 transmitter channels for each fast PLL.
These channels operate at up to 840 Mbps. For more information on
implementing more than 20 channels, see “Fast PLLs” on page 5–52. The
receiver and transmitter channels are interleaved such that each I/O bank
on the left and right side of the device has one receiver channel and one
transmitter channel per row. Figure 5–16 shows the fast PLL and channel
layout in EP1S10, EP1S20, and EP1S25 devices. Figure 5–17 shows the fast
PLL and channel layout in EP1S30 to EP1S80 devices.
f
For more the number of channels in each device, see Tables 5–10 through
5–14.
Figure 5–16. Fast PLL & Channel Layout in EP1S10, EP1S20 & EP1S25 Devices Note (1)
Up to 20 Receiver and
Transmitter Channels (2)
Transmitter
Up to 20 Receiver and
Transmitter Channels (2)
Transmitter
Receiver
Receiver
CLKIN
Fast
PLL 1
CLKIN
Fast
PLL 2
(3)
Transmitter
Receiver
Up to 20 Receiver and
Transmitter Channels (2)
Fast
PLL 4
CLKIN
Fast
PLL 3
CLKIN
(3)
Transmitter
Receiver
Up to 20 Receiver and
Transmitter Channels (2)
Notes to Figure 5–16:
(1)
(2)
(3)
Wire-bond packages only support up to 624 Mbps until characterization shows otherwise.
See Tables 5–10 through 5–14 for the exact number of channels each package and device density supports.
There is a multiplexer here to select the PLL clock source. If a PLL uses this multiplexer to clock channels outside of
its bank quadrant (e.g., if PLL 2 clocks PLL 1’s channel region), those clocked channels support up to 840 Mbps.
Altera Corporation
July 2005
5–23
Stratix Device Handbook, Volume 2
Differential I/O Interface & Fast PLLs
Figure 5–17. Fast PLL & Channel Layout in EP1S30 to EP1S80 Devices Note (1)
FPLL7CLK
Fast
PLL 7
Fast
PLL 10
Up to 20 Receiver and
20 Transmitter
Channels in 20 Rows (2)
Transmitter
FPLL10CLK
Transmitter
Receiver
Receiver
Up to 20 Receiver and
20 Transmitter
Channels in 20 Rows (2)
Transmitter
Transmitter
Receiver
CLKIN
CLKIN
Receiver
Fast
PLL 1
(3)
(3)
Fast
PLL 2
Fast
PLL 4
CLKIN
Fast
PLL 3
CLKIN
Transmitter
Transmitter
Receiver
Receiver
Up to 20 Receiver and
20 Transmitter
Channels in 20 Rows (2)
Transmitter
Transmitter
Receiver
FPLL8CLK
Receiver
Fast
PLL 8
Up to 20 Receiver and
20 Transmitter
Channels in 20 Rows (2)
Fast
PLL 9
FPLL9CLK
Notes to Figure 5–17:
(1)
(2)
(3)
Wire-bond packages only support up to 624-Mbps until characterization shows otherwise.
See Tables 5–10 through 5–14 for the exact number of channels each package and device density supports.
There is a multiplexer here to select the PLL clock source. If a PLL uses this multiplexer to clock channels outside of
its bank quadrant (e.g., if PLL 2 clocks PLL 1’s channel region), those clocked channels support up to 840 Mbps.
5–24
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Advanced Clear & Enable Control
There are several control signals for clearing and enabling PLLs and their
outputs. You can use these signals to control PLL resynchronization and
to gate PLL output clocks for low-power applications.
The PLLENABLE pin is a dedicated pin that enables and disables Stratix
device enhanced and fast PLLs. When the PLLENABLE pin is low, the
clock output ports are driven by GND and all the PLLs go out of lock.
When the PLLENABLE pin goes high again, the PLLs relock and
resynchronize to the input clocks.
The reset signals are reset/resynchronization inputs for each enhanced
PLL. Stratix devices can drive these input signals from an input pin or
from LEs. When driven high, the PLL counters reset, clearing the PLL
output and placing the PLL out of lock. When driven low again, the PLL
resynchronizes to its input as it relocks.
Receiver Data
Realignment
Most systems using serial differential I/O data transmission require a
certain data-realignment circuit. Stratix devices contain embedded datarealignment circuitry. While normal I/O operation guarantees that data
is captured, it does not guarantee the parallelization boundary, as this
point is randomly determined based on the power-up of both
communicating devices. The data-realignment circuitry corrects for bit
misalignments by shifting, or delaying, data bits.
Data Realignment Principles of Operation
Stratix devices use a realignment and clock distribution circuitry
(described in “Counter Circuitry” on page 5–22) for data realignment.
Set the internal rx_data_align node end high to assert the datarealignment circuitry. When this node is switched from a low to a high
state, the realignment circuitry is activated and the data is delayed by one
bit. To ensure the rising edge of the rx_data_align node end is latched
into the PLL, the rx_data_align node end should stay high for at least
two low-frequency clock cycles.
An external circuit or an internal custom-made state machine using LEs
can generate the signal to pull the rx_data_align node end to a high
state.
When the data realignment circuitry is activated, it generates an internal
pulse Sync S1 or Sync S2 that disables one of the two counters used for the
SERDES operation (described in “Counter Circuitry” on page 5–22). One
counter is disabled for one high-frequency clock cycle, delaying the
Altera Corporation
July 2005
5–25
Stratix Device Handbook, Volume 2
Receiver Data Realignment
RXLOADEN signal and dropping the first incoming bit of the serial input
data stream located in the first serial register of the SERDES circuitry
(shown in Figure 5–3 on page 5–8).
Figure 5–18 shows the function-timing diagram of a Stratix SERDES in
normal ×8 mode, and Figure 5–19 shows the function-timing diagrams of
a Stratix SERDES when data realignment is used.
Figure 5–18. SERDES Function Timing Diagram in Normal Operation
×8 clock
Serial data
D7
D0
D1
D2
D3
D4
D5
D6
D7
D0
D1
D2
D3
D4
D5
D6
D7
D0
D1
D2
×1 clock
PD7
D2
D2
D2
PD6
D3
D3
D3
PD5
D4
D4
D4
PD4
D5
D5
D5
PD3
D6
D6
D6
PD2
D7
D7
D7
PD1
D0
D0
D0
PD0
D1
D1
D1
5–26
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Figure 5–19. SERDES Function Timing Diagram with Data-Realignment Operation
×8 clock
Serial data
D7
D0
D1
D2
D3
D4
D5
D6
D7
D0
D1
D2
D3
D4
D5
D6
D7
D0
D1
D2
×1 clock
PD7
D2
D3
D3
PD6
D3
D4
D4
PD5
D4
D5
D5
PD4
D5
D6
D6
PD3
D6
D7
D7
PD2
D7
D0
D0
PD1
D0
D1
D1
PD0
D1
D2
D2
Generating the TXLOADEN Signal
The TXLOADEN signal controls the transfer of data between the SERDES
circuitry and the logic array when data realignment is used. To prevent
the interruption of the TXLOADEN signal during data realignment, both k
and v counter are used.
In normal operation the TXLOADEN signal is generated by the k counter.
However, during the data-realignment operation this signal is generated
by either counter. When the k counter is used for realignment, the
Altera Corporation
July 2005
5–27
Stratix Device Handbook, Volume 2
Receiver Data Realignment
TXLOADEN signal is generated by the v counter, and when the v counter
is used for realignment, the TXLOADEN signal is generated by the k
counter, as shown in Figure 5–20.
Figure 5–20. Realignment Circuit TXLOADEN Signal Control Note (1)
Counter Circuitry
Clock
Distribution
Circuitry
CLK1 LVDS
Circuitry
×1 CLK1 to logic array
÷k
TXLOADEN
Sync S1
Data
Realignment
Circuit
Realignment CLK
8
PLL Output
SYNC
Realignment CLK
Data
Realignment
Circuit
Sync S2
RXLOADEN
÷v
×1 CLK2 to logic array
CLK2 LVDS
Circuitry
÷l
GCLK/LCLK
Note to Figure 5–20:
(1)
This figure does not show additional realignment circuitry.
Realignment Implementation
The realignment signal (SYNC) is used for data realignment and
reframing. An external pin (RX_DATA_ALIGN) or an internal signal
controls the rx_data_align node end. When the rx_data_align
node end is asserted high for at least two low-frequency clock cycles, the
RXLOADEN signal is delayed by one high-frequency clock period and the
parallel bits shift by one bit. Figure 5–21 shows the timing relationship
between the high-frequency clock, the RXLOADEN signal, and the parallel
data.
5–28
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Figure 5–21. Realignment by rx_data_align Node End
10× clock
1× clock
SYNC
rxloaden
6
datain
receiver A
receiver B
5
7
6
8
7
9
8
0
9
1
0
2
1
3
2
0123456789
4
3
5
4
6
5
7
6
8
7
9
8
0
9
1
0
2
1
0123456789
3
2
4
3
5
4
6
5
7
6
8
7
9
8
0
9
1
0
1234567890
2
1
3
2
4
3
4
1234567890
A state machine can generate the realignment signal to control the
alignment procedure. Figure 5–22 shows the connection between the
realignment signal and the rx_data_align node end.
Figure 5–22. SYNC Signal Path to Realignment Circuit
Stratix Logic Array
Receiver Circuit
PD0
PD1
PD2
PD3
PD4
PD5
PD6
PD7
PD8
PD9
Parallel
Register
PD0
PD1
PD2
PD3
PD4
PD5
PD6
PD7
PD8
PD9
Register
Array
SYNC Out
10
Pattern
Detection
State Machine
Hold
Register
TXLOADEN
×1
×W/J
Realignment
Circuit
SYNC
To guarantee that the rx_data_align signal generated by a user state
machine is latched correctly by the counters, the user circuit must meet
certain requirements.
■
Altera Corporation
July 2005
The design must include an input synchronizing register to ensure
that data is synchronized to the ×1 clock.
5–29
Stratix Device Handbook, Volume 2
Source-Synchronous Timing Budget
■
■
■
■
SourceSynchronous
Timing Budget
After the pattern detection state machine, use another synchronizing
register to capture the generated SYNC signal and synchronize it to
the ×1 clock.
Since the skew in the path from the output of this synchronizing
register to the PLL is undefined, the state machine must generate a
pulse that is high for two ×1 clock periods.
Since the SYNC generator circuitry only generates a single fast clock
period pulse for each SYNC pulse, you cannot generate additional
SYNC pulses until the comparator signal is reset low.
To guarantee the pattern detection state machine does not incorrectly
generate multiple SYNC pulses to shift a single bit, the state machine
must hold the SYNC signal low for at least three ×1 clock periods
between pulses.
This section discusses the timing budget, waveforms, and specifications
for source-synchronous signaling in Stratix devices. LVDS, LVPECL,
PCML, and HyperTransport I/O standards enable high-speed data
transmission. This high data-transmission rate results in better overall
system performance. To take advantage of fast system performance, you
must understand how to analyze timing for these high-speed signals.
Timing analysis for the differential block is different from traditional
synchronous timing analysis techniques.
Rather than focusing on clock-to-output and setup times, sourcesynchronous timing analysis is based on the skew between the data and
the clock signals. High-speed differential data transmission requires you
to use timing parameters provided by IC vendors and to consider board
skew, cable skew, and clock jitter. This section defines the sourcesynchronous differential data orientation timing parameters, and timing
budget definitions for Stratix devices, and explains how to use these
timing parameters to determine a design's maximum performance.
Differential Data Orientation
There is a set relationship between an external clock and the incoming
data. For operation at 840 Mbps and W = 10, the external clock is
multiplied by 10 and phase-aligned by the PLL to coincide with the
sampling window of each data bit. The third falling edge of highfrequency clock is used to strobe the incoming high-speed data.
Therefore, the first two bits belong to the previous cycle. Figure 5–23
shows the data bit orientation of the ×10 mode as defined in the
Quartus II software.
5–30
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Figure 5–23. Bit Orientation in the Quartus II Software
inclock/outclock
10 LVDS Bits
MSB
data in
n-1
n-0
9
8
7
6
5
4
LSB
3
2
1
0
high-frequency clock
Differential I/O Bit Position
Data synchronization is necessary for successful data transmission at
high frequencies. Figure 5–24 shows the data bit orientation for a receiver
channel operating in ×8 mode. Similar positioning exists for the most
significant bits (MSBs) and least significant bits (LSBs) after
deserialization, as listed in Table 5–5.
Figure 5–24. Bit Order for One Channel of Differential Data
inclock/outclock
Previous Cycle
Data in/
Data out
Current Cycle
D7
MSB
D6
D5
0
0
D4
D3
Next Cycle
D2
D1
D0
LSB
1
0
Example: Sending the Data 10010110
Previous Cycle
Data in/
Data out
Current Cycle
1
MSB
Altera Corporation
July 2005
1
0
Next Cycle
1
LSB
5–31
Stratix Device Handbook, Volume 2
Source-Synchronous Timing Budget
Table 5–5 shows the conventions for differential bit naming for
18 differential channels. However, the MSB and LSB are increased with
the number of channels used in a system.
Table 5–5. LVDS Bit Naming
Receiver Data Channel
Number
Internal 8-Bit Parallel Data
MSB Position
LSB Position
1
7
0
2
15
8
3
23
16
4
31
24
5
39
32
6
47
40
7
55
48
8
63
56
9
71
64
10
79
72
11
87
80
12
95
88
13
103
96
14
111
104
15
119
112
16
127
120
17
135
128
18
143
136
Timing Definition
The specifications used to define high-speed timing are described in
Table 5–6.
Table 5–6. High-Speed Timing Specifications & Terminology (Part 1 of 2)
High-Speed Timing Specification
Terminology
tC
High-speed receiver/transmitter input and output clock period.
fHSCLK
High-speed receiver/transmitter input and output clock frequency.
tRISE
Low-to-high transmission time.
5–32
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Table 5–6. High-Speed Timing Specifications & Terminology (Part 2 of 2)
High-Speed Timing Specification
Terminology
tFALL
High-to-low transmission time.
Timing unit interval (TUI)
The timing budget allowed for skew, propagation delays, and data
sampling window. (TUI = 1/(Receiver Input Clock Frequency ×
Multiplication Factor) = tC/w).
fHSDR
Maximum LVDS data transfer rate (fHSDR = 1/TUI).
Channel-to-channel skew (TCCS)
The timing difference between the fastest and slowest output edges,
including tCO variation and clock skew. The clock is included in the TCCS
measurement.
Sampling window (SW)
The period of time during which the data must be valid in order for you to
capture it correctly. The setup and hold times determine the ideal strobe
position within the sampling window.
SW = tSW (max) – tSW (min).
Input jitter (peak-to-peak)
Peak-to-peak input jitter on high-speed PLLs.
Output jitter (peak-to-peak)
Peak-to-peak output jitter on high-speed PLLs.
tDUTY
Duty cycle on high-speed transmitter output clock.
tLOCK
Lock time for high-speed transmitter and receiver PLLs.
Altera Corporation
July 2005
5–33
Stratix Device Handbook, Volume 2
Table 5–7. High-Speed I/O Specifications for Flip-Chip Packages (Part 1 of 3) Notes (1), (2)
-5 Speed Grade
Symbol
fHSDR Device
operation
(LVDS, LVPECL,
HyperTransport
technology)
5–34
Stratix Device Handbook, Volume 2
fHSCLK (Clock
frequency)
(PCML)
fHSCLK = fHSDR / W
-7 Speed Grade
-8 Speed Grade
Unit
Min
fHSCLK (Clock
frequency)
(LVDS, LVPECL,
HyperTransport
technology)
fHSCLK = fHSDR / W
-6 Speed Grade
Conditions
Typ
Max
Min
Typ
Max
Min
Typ
Max
Min
Typ
Max
W = 4 to 30
10
210
10
210
10
156
10
115.5
MHz
W = 2 (Serdes
bypass)
50
231
50
231
50
231
50
231
MHz
W = 2 (Serdes used)
150
420
150
420
150
312
150
231
MHz
W = 1 (Serdes
bypass)
100
462
100
462
100
462
100
462
MHz
W = 1 (Serdes used)
300
717
300
717
300
624
300
462
MHz
J = 10
300
840
300
840
300
640
300
462
Mbps
J=8
300
840
300
840
300
640
300
462
Mbps
J=7
300
840
300
840
300
640
300
462
Mbps
J=4
300
840
300
840
300
640
300
462
Mbps
J=2
100
462
100
462
100
640
100
462
Mbps
J = 1 (LVDS and
LVPECL only)
100
462
100
462
100
640
100
462
Mbps
W = 4 to 30 (Serdes
used)
10
100
10
100
10
77.75
10
77.75
MHz
W = 2 (Serdes
bypass)
50
200
50
200
50
150
50
150
MHz
W = 2 (Serdes used)
150
200
150
200
150
155.5
150
155.5
MHz
W = 1 (Serdes
bypass)
100
250
100
250
100
200
100
200
MHz
W = 1 (Serdes used)
300
400
300
400
300
311
300
311
MHz
Source-Synchronous Timing Budget
Altera Corporation
July 2005
Tables 5–7 and 5–8 show the high-speed I/O timing for Stratix devices
-5 Speed Grade
Symbol
-7 Speed Grade
-8 Speed Grade
Unit
Min
fHSDR Device
operation (PCML)
-6 Speed Grade
Conditions
Typ
Max
Min
Typ
Max
Min
Typ
Max
Min
Typ
Max
J = 10
300
400
300
400
300
311
300
311
Mbps
J=8
300
400
300
400
300
311
300
311
Mbps
J=7
300
400
300
400
300
311
300
311
Mbps
J=4
300
400
300
400
300
311
300
311
Mbps
J=2
100
400
100
400
100
300
100
300
Mbps
J=1
100
250
100
250
100
200
100
200
Mbps
300
ps
TCCS
All
SW
PCML (J = 4, 7, 8,
10)
200
750
200
300
750
800
800
ps
5–35
Stratix Device Handbook, Volume 2
PCML (J = 2)
900
900
1,200
1,200
ps
PCML (J = 1)
1,500
1,500
1,700
1,700
ps
LVDS and LVPECL
(J = 1)
500
500
550
550
ps
LVDS, LVPECL,
HyperTransport
technology (J = 2
through 10)
440
440
500
500
ps
Input jitter tolerance All
(peak-to-peak)
250
250
250
250
ps
Output jitter (peakto-peak)
All
160
160
200
200
ps
Output tRISE
LVDS
80
110
120
80
110
120
80
110
120
80
110
120
ps
HyperTransport
technology
110
170
200
110
170
200
120
170
200
120
170
200
ps
LVPECL
90
130
150
90
130
150
100
135
150
100
135
150
ps
PCML
80
110
135
80
110
135
80
110
135
80
110
135
ps
Source-Synchronous Timing Budget
Altera Corporation
July 2005
Table 5–7. High-Speed I/O Specifications for Flip-Chip Packages (Part 2 of 3) Notes (1), (2)
-5 Speed Grade
Symbol
Output tFALL
tDUTY
-7 Speed Grade
-8 Speed Grade
Unit
Min
Typ
Max
Min
Typ
Max
Min
Typ
Max
Min
Typ
Max
LVDS
80
110
120
80
110
120
80
110
120
80
110
120
ps
HyperTransport
technology
110
170
200
110
170
200
110
170
200
110
170
200
ps
LVPECL
90
130
160
90
130
160
100
135
160
100
135
160
ps
PCML
105
140
175
105
140
175
110
145
175
110
145
175
ps
LVDS (J = 2 through
10)
47.5
50
52.5
47.5
50
52.5
47.5
50
52.5
47.5
50
52.5
%
45
50
55
45
50
55
45
50
55
45
50
55
%
100
μs
LVDS (J =1) and
LVPECL, PCML,
HyperTransport
technology
tLOCK
-6 Speed Grade
Conditions
All
100
100
100
Notes to Table 5–7:
(1)
(2)
When J = 4, 7, 8, and 10, the SERDES block is used.
When J = 2 or J = 1, the SERDES is bypassed.
Table 5–8. High-Speed I/O Specifications for Wire-Bond Packages (Part 1 of 3)
-6 Speed Grade
Symbol
-8 Speed Grade
Unit
Min
Altera Corporation
July 2005
fHSCLK (Clock frequency)
(LVDS,LVPECL, HyperTransport
technology)
fHSCLK = fHSDR / W
-7 Speed Grade
Conditions
W = 4 to 30 (Serdes used)
10
Typ
Max
Min
156
Typ
Max
Min
Typ
Max
10
115.5
10
115.5
MHz
W = 2 (Serdes bypass)
50
231
50
231
50
231
MHz
W = 2 (Serdes used)
150
312
150
231
150
231
MHz
W = 1 (Serdes bypass)
100
311
100
270
100
270
MHz
W = 1 (Serdes used)
300
624
300
462
300
462
MHz
Source-Synchronous Timing Budget
5–36
Stratix Device Handbook, Volume 2
Table 5–7. High-Speed I/O Specifications for Flip-Chip Packages (Part 3 of 3) Notes (1), (2)
-6 Speed Grade
Symbol
fH S C L K (Clock frequency)
(PCML)
fHSCLK = fHSDR / W
Device operation, fH S D R
(PCML)
-8 Speed Grade
Unit
Min
fHSDR Device operation,
(LVDS,LVPECL, HyperTransport
technology)
-7 Speed Grade
Conditions
Typ
Max
Min
Typ
Max
Min
Typ
Max
J = 10
300
624
300
462
300
462
Mbps
J=8
300
624
300
462
300
462
Mbps
J=7
300
624
300
462
300
462
Mbps
J=4
300
624
300
462
300
462
Mbps
J=2
100
462
100
462
100
462
Mbps
J = 1 (LVDS and LVPECL only)
100
311
100
270
100
270
Mbps
W = 4 to 30 (Serdes used)
10
77.75
W = 2 (Serdes bypass)
50
150
W = 2 (Serdes used)
150
155.5
W = 1 (Serdes bypass)
100
200
W = 1 (Serdes used)
300
311
MHz
50
77.5
50
77.5
MHz
MHz
100
155
100
155
MHz
MHz
5–37
Stratix Device Handbook, Volume 2
J = 10
300
311
Mbps
J=8
300
311
Mbps
J=7
300
311
Mbps
J=4
300
311
Mbps
J=2
100
300
100
155
100
155
Mbps
J=1
100
200
100
155
100
155
Mbps
TCCS
All
SW
PCML (J = 4, 7, 8, 10) only
400
400
400
ps
800
800
800
ps
PCML (J = 2) only
1,200
1,200
1,200
ps
PCML (J = 1) only
1,700
1,700
1,700
ps
LVDS and LVPECL (J = 1) only
550
550
550
ps
LVDS, LVPECL, HyperTransport
technology (J = 2 through 10) only
500
500
500
ps
Source-Synchronous Timing Budget
Altera Corporation
July 2005
Table 5–8. High-Speed I/O Specifications for Wire-Bond Packages (Part 2 of 3)
-6 Speed Grade
Symbol
-7 Speed Grade
-8 Speed Grade
Conditions
Unit
Min
Typ
Max
Min
Typ
Max
Min
Typ
Max
Input jitter tolerance (peak-topeak)
All
250
250
250
ps
Output jitter (peak-to-peak)
All
200
200
200
ps
Output tR I S E
LVDS
80
110
120
80
110
120
80
110
120
ps
HyperTransport technology
120
170
200
120
170
200
120
170
200
ps
LVPECL
100
135
150
100
135
150
100
135
150
ps
PCML
80
110
135
80
110
135
80
110
135
ps
Output tFA L L
tD U T Y
LVDS
80
110
120
80
110
120
80
110
120
ps
HyperTransport
110
170
200
110
170
200
110
170
200
ps
LVPECL
100
135
160
100
135
160
100
135
160
ps
PCML
110
145
175
110
145
175
110
145
175
ps
LVDS (J =2..10) only
47.5
50
52.5
47.5
50
52.5
47.5
50
52.5
%
45
50
55
45
50
55
45
50
55
%
100
μs
LVDS (J =1) and LVPECL, PCML,
HyperTransport technology
tL O C K
All
100
100
Source-Synchronous Timing Budget
Altera Corporation
July 2005
Table 5–8. High-Speed I/O Specifications for Wire-Bond Packages (Part 3 of 3)
5–38
Stratix Device Handbook, Volume 2
High-Speed Differential I/O Interfaces in Stratix Devices
Input Timing Waveform
Figure 5–25 illustrates the essential operations and the timing
relationship between the clock cycle and the incoming serial data. For a
functional description of the SERDES, see “Principles of SERDES
Operation” on page 5–6.
Figure 5–25. Input Timing Waveform Note (1)
Input Clock
(Differential
Signal)
Previous Cycle
bit 0
Input Data
Next
Cycle
Current Cycle
bit 1
bit 2
bit 3
bit 4
bit 5
bit 6
tsw0 (min)
MSB
tsw0 (max)
tsw1 (min)
tsw1 (max)
tsw2 (min)
tsw2 (max)
tsw3 (min)
tsw3 (max)
tsw4 (min)
tsw4 (max)
tsw5 (min)
tsw5 (max)
tsw6 (min)
tsw6 (max)
tsw7 (min)
tsw7 (max)
bit 7
LSB
Note to Figure 5–25:
(1)
The timing specifications are referenced at a 100-mV differential voltage.
Altera Corporation
July 2005
5–39
Stratix Device Handbook, Volume 2
Source-Synchronous Timing Budget
Output Timing
The output timing waveform in Figure 5–26 illustrates the relationship
between the output clock and the serial output data stream.
Figure 5–26. Output Timing Waveform Note (1)
Output Clock
(Differential
Signal)
Previous Cycle
Output Data
bit 0
Next
Cycle
Current Cycle
bit 1
bit 2
bit 3
bit 4
bit 5
bit 6
bit 7
TPPos0 (min)
MSB
TPPos0 (max)
LSB
TPPos1 (min)
TPPos1 (max)
TPPos2 (min)
TPPos2 (max)
TPPos3 (min)
TPPos3 (max)
TPPos4 (min)
TPPos4 (max)
TPPos5 (min)
TPPos5 (max)
TPPos6 (min)
TPPos6 (max)
TPPos7 (min)
TPPos7 (max)
Note to Figure 5–26:
(1)
The timing specifications are referenced at a 250-mV differential voltage.
Receiver Skew Margin
Change in system environment, such as temperature, media (cable,
connector, or PCB) loading effect, a receiver's inherent setup and hold,
and internal skew, reduces the sampling window for the receiver. The
timing margin between receiver’s clock input and the data input
sampling window is known as RSKM. Figure 5–27 illustrates the
relationship between the parameter and the receiver’s sampling window.
5–40
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Figure 5–27. Differential High-Speed Timing Diagram & Timing Budget
Timing Diagram
External
Input Clock
Time Unit Interval (TUI)
Internal
Clock
TCCS
TCCS
Receiver
Input Clock
Sampling
Window (SW)
RSKM
RSKM
TPPos (max)
Bit n + 1
TPPos (max)
Bit n
tSW (min)
Bit n
TPPos (min)
Bit n
Timing Budget
Internal tSW (max)
Clock
Bit n
Falling Edge
TPPos (min)
Bit n + 1
TUI
External
Clock
Clock Placement
Internal
Clock
Synchronization
Transmitter
Output Data
RSKM
RSKM
TCCS
TCCS
2
TSWEND
Receiver
Input Data
TSWBEGIN
Altera Corporation
July 2005
Sampling
Window
5–41
Stratix Device Handbook, Volume 2
SERDES Bypass DDR Differential Signaling
Switching Characteristics
Timing specifications for Stratix devices are listed in Tables 5–7 and 5–8.
You can also find Stratix device timing information in the Stratix Device
Family Data Sheet section of the Stratix Device Handbook, Volume 1.
Timing Analysis
Differential timing analysis is based on skew between data and the clock
signals. For static timing analysis, the timing characteristics of the
differential I/O standards are guaranteed by design and depend on the
frequency at which they are operated. Use the values in the Stratix Device
Family Data Sheet section of the Stratix Device Handbook, Volume 1 to
calculate system timing margins for various I/O protocols. For detailed
descriptions and implementations of these protocols, see the Altera web
site at www.altera.com.
SERDES Bypass
DDR Differential
Signaling
Each Stratix device high-speed differential I/O channel can transmit or
receive data in by-two (×2) mode at up to 624 Mbps using PLLs. These
pins do not require dedicated SERDES circuitry and they implement
serialization and deserialization with minimal logic.
SERDES Bypass DDR Differential Interface Review
Stratix devices use dedicated DDR circuitry to implement ×2 differential
signaling. Although SDR circuitry samples data only at the positive edge
of the clock, DDR captures data on both the rising and falling edges for
twice the transfer rate of SDR. Stratix device shift registers, internal global
PLLs, and I/O cells can perform serial-to-parallel conversions on
incoming data and parallel-to-serial conversion on outgoing data.
SERDES Clock Domains
The SERDES bypass differential signaling can use any of the many clock
domains available in Stratix devices. These clock domains fall into four
categories: global, regional, fast regional, and internally generated.
General-purpose PLLs generate the global clock domains. The fast PLLs
can generate additional global clocks domains. Each PLL features two
taps that directly drive two unique global clock networks. A dedicated
clock pin drives each general-purpose PLL. These clock lines are utilized
when designing for speeds up to 420 Mbps. Tables 5–3 and 5–4 on
page 5–19, respectively, show the available clocks in Stratix devices.
5–42
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
SERDES Bypass DDR Differential Signaling Receiver Operation
The SERDES bypass differential signaling receiver uses the Stratix device
DDR input circuitry to receive high-speed serial data. The DDR input
circuitry consists of a pair of shift registers used to capture the high-speed
serial data, and a latch.
One register captures the data on the positive edge of the clock (generated
by PLL) and the other register captures the data on the negative edge of
the clock. Because the data captured on the negative edge is delayed by
one-half of the clock cycle, it is latched before it interfaces with the system
logic.
Figure 5–28 shows the DDR timing relationship between the incoming
serial data and the clock. In this example, the inclock signal is running
at half the speed of the incoming data. However, other combinations are
also possible. Figure 5–29 shows the DDR input and the other modules
used in a Flexible-LVDS receiver design to interface with the system logic.
Figure 5–28. ×2 Timing Relation between Incoming Serial Data & Clock
clock
datain
B0
neg_reg_out XX
Altera Corporation
July 2005
A0
B1
B0
A1
B2
B1
A2
B3
B2
A3
B3
dataout_l
XX
B0
B1
B2
dataout_h
XX
A0
A1
A2
5–43
Stratix Device Handbook, Volume 2
SERDES Bypass DDR Differential Signaling
Figure 5–29. ×2 Data Rate Receiver Channel with Deserialization Factor of 8
DDR IOE
datain
Shift
Register
DFF
D0, D2, D4, D6
Register
D1, D3, D5, D7
Stratix
Logic
Array
Latch
DFF
Shift
Register
inclock
PLL
×4
Clock
×1
SERDES Bypass DDR Differential Signaling Transmitter
Operation
The ×2 differential signaling transmitter uses the Stratix device DDR
output circuitry to transmit high-speed serial data. The DDR output
circuitry consists of a pair of shift registers and a multiplexer. The shift
registers capture the parallel data on the clock’s rising edge (generated by
the PLL), and a multiplexer transmits the data in sync with the clock.
Figure 5–30 shows the DDR timing relation between the parallel data and
the clock. In this example, the inclock signal is running at half the speed
of the data. However, other combinations are possible. Figure 5–31 shows
the DDR output and the other modules used in a ×2 transmitter design to
interface with the system logic.
5–44
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Figure 5–30. ×2 Timing Relation between Parallel Data & Clock
outclock
datain_l XX
B0
B1
B2
B3
datain_h XX
A0
A1
A2
A3
dataout
XX
A0
B0
A1
B1
A2
B2
A3
Figure 5–31. ×2 Data Rate Transmitter Channel with Serialization Factor of 8
DDR IOE
DFF
D0, D2,
D4, D6
Stratix
Logic
Array
Shift
Register
dataout
DFF
D1, D3,
D5, D7
Shift
Register
×1
PLL
×4
×1
inclock
High-Speed
Interface Pin
Locations
Stratix high-speed interface pins are located at the edge of the package to
limit the possible mismatch between a pair of high-speed signals. Stratix
devices have eight programmable I/O banks. Figure 5–32 shows the I/O
pins and their location relative to the package.
Altera Corporation
July 2005
5–45
Stratix Device Handbook, Volume 2
Differential I/O Termination
Figure 5–32. Differential I/O Pin Locations
Differential I/O Pins
(LVDS, LVPECL,
PCML, HyperTransport)
Regular I/O Pins
Differential I/O Pins
(LVDS, LVPECL,
PCML, HyperTransport)
A
B
C
D
E
F
G
H
J
K
L
M
N
P
R
T
U
V
W
Y
AA
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
Regular I/O Pins
Differential I/O
Termination
Stratix devices implement differential on-chip termination to reduce
reflections and maintain signal integrity. On-chip termination also
minimizes the number of external resistors required. This simplifies
board design and places the resistors closer to the package, eliminating
small stubs that can still lead to reflections.
RD Differential Termination
Stratix devices support differential on-chip termination for the LVDS I/O
standard. External termination is required on output pins for PCML
transmitters. HyperTransport, LVPECL, and LVDS receivers require
100 ohm termination at the input pins. Figure 5–33 shows the device with
differential termination for the LVDS I/O standard.
f
For more information on differential on-chip termination technology, see
the Selectable I/O Standards in Stratix & Stratix GX Devices chapter.
5–46
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Figure 5–33. LVDS Differential On-Chip Termination
LVDS Receiver with
On-Chip 100-Ω Termination
LVDS Transmitter
Z0 = 50 Ω
RD
Z0 = 50 Ω
HyperTransport & LVPECL Differential Termination
HyperTransport and LVPECL I/O standards are terminated by an
external 100-Ω resistor on the input pin. Figure 5–34 shows the device
with differential termination for the HyperTransport or LVPECL I/O
standard.
Figure 5–34. HyperTransport & LVPECL Differential Termination
Differential
Transmitter
Differential Receiver
Z0 = 50 Ω
RD
Z0 = 50 Ω
PCML Differential Termination
The PCML I/O technology is an alternative to the LVDS I/O technology,
and use an external voltage source (VTT), a pair of 100-Ω resistors on the
input side and a pair of 50-Ω resistors on the output side. Figure 5–35
shows the device with differential termination for PCML I/O standard.
Altera Corporation
July 2005
5–47
Stratix Device Handbook, Volume 2
Differential I/O Termination
Figure 5–35. PCML Differential Termination
VTT
Differential
Transmitter
50 Ω
50 Ω
50 Ω
Differential
Receiver
50 Ω
Z0 = 50 Ω
Z0 = 50 Ω
Differential HSTL Termination
The HSTL Class I and II I/O standards require a 0.75-V VREF and a 0.75V VTT. Figures 5–36 and 5–37 show the device with differential
termination for HSTL Class I and II I/O standard.
Figure 5–36. Differential HSTL Class I Termination
VTT = 0.75 V
Differential
Transmitter
50 Ω
VTT = 0.75 V
50 Ω
Differential
Receiver
Z0 = 50 Ω
Z0 = 50 Ω
5–48
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Figure 5–37. Differential HSTL Class II Termination
VTT = 0.75 V
Differential
Transmitter
50 Ω
VTT = 0.75 V
VTT = 0.75 V
50 Ω
50 Ω
VTT = 0.75 V
Differential
Receiver
50 Ω
Z0 = 50 Ω
Z0 = 50 Ω
Differential SSTL-2 Termination
The SSTL-2 Class I and II I/O standards require a 1.25-V VREF and a
1.25-V VTT. Figures 5–37 and 5–38 show the device with differential
termination for SSTL-2 Class I and II I/O standard.
Figure 5–38. Differential SSTL-2 Class I Termination
VTT = 1.25 V
Differential
Transmitter
50 Ω
VTT = 1.25 V
50 Ω
Differential
Receiver
25 Ω
Z0 = 50 Ω
25 Ω
Z0 = 50 Ω
Altera Corporation
July 2005
5–49
Stratix Device Handbook, Volume 2
Board Design Consideration
Figure 5–39. Differential SSTL-2 Class II Termination
VTT = 1.25 V
Differential
Transmitter
50 Ω
VTT = 1.25 V
50 Ω
VTT = 1.25 V
50 Ω
VTT = 1.25 V
50 Ω
Differential
Receiver
25 Ω
Z0 = 50 Ω
25 Ω
Z0 = 50 Ω
Board Design
Consideration
This section is a brief explanation of how to get the optimal performance
from the Stratix high-speed I/O block and ensure first-time success in
implementing a functional design with optimal signal quality. For more
information on detailed board layout recommendation and I/O pin
terminations see AN 224: High-Speed Board Layout Guidelines.
You must consider the critical issues of controlled impedance of traces
and connectors, differential routing, and termination techniques to get
the best performance from the IC. For more information, use this chapter
and the Stratix Device Family Data Sheet section of the Stratix Device
Handbook, Volume 1.
The Stratix high-speed module generates signals that travel over the
media at frequencies as high as 840 Mbps. Board designers should use the
following general guidelines:
■
■
■
■
■
■
■
Baseboard designs on controlled differential impedance. Calculate
and compare all parameters such as trace width, trace thickness, and
the distance between two differential traces.
Place external reference resistors as close to receiver input pins as
possible.
Use surface mount components.
Avoid 90° or 45° corners.
Use high-performance connectors such as HS-3 connectors for
backplane designs. High-performance connectors are provided by
Teradyne Corp (www.teradyne.com) or Tyco International Ltd.
(www.tyco.com).
Design backplane and card traces so that trace impedance matches
the connector’s and/or the termination’s impedance.
Keep equal number of vias for both signal traces.
5–50
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
■
■
■
■
■
Software
Support
Create equal trace lengths to avoid skew between signals. Unequal
trace lengths also result in misplaced crossing points and system
margins as the TCCS value increases.
Limit vias because they cause discontinuities.
Use the common bypass capacitor values such as 0.001 µF, 0.01 µF,
and 0.1 µF to decouple the fast PLL power and ground planes.
Keep switching TTL signals away from differential signals to avoid
possible noise coupling.
Do not route TTL clock signals to areas under or above the
differential signals.
This section provides information on using the Quartus II software to
create Stratix designs with LVDS transmitters or receivers. You can use
the altlvds megafunction in the Quartus II software to implement the
SERDES circuitry. You must bypass the SERDES circuitry in ×1 and ×2
mode designs and use the altddio megafunction to implement the
deserialization instead. You can use either the logic array or the M512
RAM blocks closest to the differential pins for deserialization in SERDES
bypass mode.
Differential Pins in Stratix
Stratix device differential pins are located in I/O banks 1, 2, 5, and 6 (see
Figure 5–1 on page 5–2). Each bank has differential transmitter and
differential receiver pin pairs. You can use each differential transmitter
pin pair as either a differential data pin pair or a differential clock pin pair
because Stratix devices do not have dedicated LVDS tx_outclock pin
pairs. The differential receiver pin pairs can only function as differential
data pin pairs. You can use these differential pins as regular user I/O pins
when not used as differential pins. When using differential signaling in
an I/O bank, you cannot place non-differential output or bidirectional
pads within five I/O pads of either side of the differential pins to avoid a
decrease in performance on the LVDS signals.
You only need to make assignments to the positive pin of the pin-pair.
The Quartus II software automatically reserves and makes the same
assignment to the negative pin. If you do not assign any differential I/O
standard to the differential pins, the Quartus II software sets them as
LVDS differential pins during fitting, if the design uses the SERDES
circuitry. Additionally, if you bypass the SERDES circuitry, you can still
use the differential pins by assigning a differential I/O standard to the
pins in the Quartus II software. However, when you bypass the SERDES
circuitry in the ×1 and ×2 mode, you must assign the correct differential
I/O standard to the associated pins in the Assignment Organizer. For
more information on how to use the Assignment Organizer, see the
Quartus II On-Line Help.
Altera Corporation
July 2005
5–51
Stratix Device Handbook, Volume 2
Software Support
Stratix devices can drive the PLL_LOCK signal to both output pins and
internal logic. As a result, you do not need a dedicated LOCK pin for your
PLLs. In addition, there is only one PLL_ENABLE pin that enables all the
PLLs on the device, including the fast PLLs. You must use either the
LVTTL or LVCMOS I/O standard with this pin.
Table 5–9 displays the LVDS pins in Stratix devices.
Table 5–9. LVDS Pin Names
Pin Names
Functions
DIFFIO_TX#p
Transmitter positive data or output clock pin
DIFFIO_TX#n
Transmitter negative data or output clock pin
DIFFIO_RX#p
Receiver positive data pin
DIFFIO_RX#n
Receiver negative data pin
FPLLCLK#p
Positive input clock pin to the corner fast PLLs (1), (2)
FPLLCLK#n
Negative input clock pin to the corner fast PLLs (1), (2)
CLK#p
Positive input clock pin (2)
CLK#n
Negative input clock pin (2)
Notes to Table 5–9:
(1)
(2)
The FPLLCLK pin-pair is only available in EP1S30, EP1S40, EP1S60, EP1S80
devices.
Either a FPLLCLK pin or a CLK pin can drive the corner fast PLLs (PLL7, PLL8,
PLL9, and PLL10) when used for general purpose. CLK pins cannot drive these
fast PLLs in high-speed differential I/O mode.
Fast PLLs
Each fast PLL features a multiplexed input path from a global or regional
clock net. A clock pin or an output from another PLL in the device can
drive the input path. The input clock for PLLs used to clock receiver the
rx_inclock port on the altlvds_rx megafunction must be driven by
a dedicated clock pin (CLK[3..0,8..11]) or the corner pins that clock the
corner PLLs (FPLL[10..7]CLK). EP1S10, EP1S20, and EP1S25 devices have
a total of four fast PLLs located in the center of both sides of the device
(see Figure 5–16 on page 5–23). EP1S30 and larger devices have two
additional fast PLLs per side at the top and bottom corners of the device.
As shown in Figure 5–17 on page 5–24, the corner fast PLL shares an I/O
bank with the closest center fast PLL (e.g., PLLs 1 and 7 share an I/O
bank). The maximum input clock frequency for enhanced PLLs is 684
MHz and 717 MHz for fast PLLs.
f
For more information on Stratix PLLs, see the General-Purpose PLLs in
Stratix & Stratix GX Devices chapter.
5–52
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
One fast PLL can drive the 20 transmitter channels and 20 receiver
channels closest to it with data rates of up to 840 Mbps. Wire-bond
packages support a data rate of 624 Mbps. The corner fast PLLs in EP1S80
devices support data rates of up to 840 Mbps. See Tables 5–10 through
5–14 for the number of high-speed differential channels in a particular
Stratix device density and package.
Since the fast PLL drives the 20 closest differential channels, there are
coverage overlaps in the EP1S30 and larger devices that have two fast
PLLs per I/O bank. In these devices, either the center fast PLL or the
corner fast PLL can drive the differential channels in the middle of the
I/O bank.
Fast PLLs can drive more than 20 transmitter and 20 receiver channels
(see Tables 5–10 through 5–14 and Figures 5–16, and 5–17 for the number
of channels each PLL can drive). In addition, the center fast PLLs can
drive either one I/O bank or both I/O banks on the same side (left or
right) of the device, while the corner fast PLLs can only drive the
differential channels in its I/O bank. Neither fast PLL can drive the
differential channels in the opposite side of the device.
The center fast PLLs can only drive two I/O at 840 Mbps. For example,
EP1S20 device fast PLL 1 can drive all 33 differential channels on its side
(17 channels from I/O bank 2 and 16 channels from I/O bank 1) running
at 840 Mbps in 4× mode. When a center fast PLL drives the opposite bank
on the same side of the device, the other center fast PLL cannot drive any
differential channels on the device.
See Tables 5–10 through 5–14 for the maximum number of channels that
one fast PLL can drive. The number of channels is also listed in the
Quartus II software. The Quartus II software gives an error message if
you try to compile a design exceeding the maximum number of channels.
f
Altera Corporation
July 2005
Additional high-speed DIFFIO pin information for Stratix devices is
available in Volume 3 of the Stratix Device Handbook.
5–53
Stratix Device Handbook, Volume 2
Software Support
Table 5–10 shows the number of channels and fast PLLs in EP1S10,
EP1S20, and EP1S25 devices. Tables 5–11 through 5–14 show this
information for EP1S30, EP1S40, EP1S60, and EP1S80 devices.
Table 5–10. EP1S10, EP1S20 & EP1S25 Device Differential Channels (Part 1 of 2) Note (1)
Device
EP1S10
Package
Transmitter/
Receiver
484-pin FineLine BGA Transmitter
(2)
Receiver
672-pin FineLine BGA Transmitter
672-pin BGA
(2)
Receiver
780-pin FineLine BGA Transmitter
(2)
Receiver
EP1S20
484-pin FineLine BGA Transmitter
(2)
Receiver
672-pin FineLine BGA Transmitter
672-pin BGA
(2)
Receiver
780-pin FineLine BGA Transmitter
(2)
Receiver
5–54
Stratix Device Handbook, Volume 2
Total
Channels
20
20
36
36
44
44
24
20
48
50
66
66
Maximum
Speed
(Mbps)
Center Fast PLLs
PLL 1
PLL 2
PLL 3
PLL 4
840
5
5
5
5
840 (3)
10
10
10
10
840
5
5
5
5
840 (3)
10
10
10
10
624 (4)
9
9
9
9
624 (3)
18
18
18
18
624 (4)
9
9
9
9
624 (3)
18
18
18
18
840
11
11
11
11
840 (3)
22
22
22
22
840
11
11
11
11
840 (3)
22
22
22
22
840
6
6
6
6
840 (3)
12
12
12
12
840
5
5
5
5
840 (3)
10
10
10
10
624 (4)
12
12
12
12
624 (3)
24
24
24
24
624 (4)
13
12
12
13
624 (3)
25
25
25
25
840
17
16
16
17
840 (3)
33
33
33
33
840
17
16
16
17
840 (3)
33
33
33
33
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Table 5–10. EP1S10, EP1S20 & EP1S25 Device Differential Channels (Part 2 of 2) Note (1)
Device
EP1S25
Package
Transmitter/
Receiver
672-pin FineLine BGA Transmitter
672-pin BGA
(2)
Receiver
780-pin FineLine BGA Transmitter
(2)
Receiver
1,020-pin FineLine
BGA
Total
Channels
Maximum
Speed
(Mbps)
PLL 1
PLL 2
PLL 3
PLL 4
56
624 (4)
14
14
14
14
624 (3)
28
28
28
28
624 (4)
14
15
15
14
624 (3)
29
29
29
29
58
70
66
Transmitter
(2)
78
Receiver
78
Center Fast PLLs
840
18
17
17
18
840 (3)
35
35
35
35
840
17
16
16
17
840 (3)
33
33
33
33
840
19
20
20
19
840 (3)
39
39
39
39
840
19
20
20
19
840 (3)
39
39
39
39
Notes to Table 5–10:
(1)
(2)
(3)
(4)
The first row for each transmitter or receiver reports the number of channels driven directly by the PLL. The second
row below it shows the maximum channels a PLL can drive if cross bank channels are used from the adjacent center
PLL. For example, in the 484-pin FineLine BGA EP1S10 device, PLL 1 can drive a maximum of five channels at
840 Mbps or a maximum of 10 channels at 840 Mbps. The Quartus II software may also merge receiver and
transmitter PLLs when a receiver is driving a transmitter. In this case, one fast PLL can drive both the maximum
numbers of receiver and transmitter channels.
The number of channels listed includes the transmitter clock output (tx_outclock) channel. If the design
requires a DDR clock, it can use an extra data channel.
These channels span across two I/O banks per side of the device. When a center PLL clocks channels in the
opposite bank on the same side of the device it is called cross-bank PLL support. Both center PLLs can clock crossbank channels simultaneously if, for example, PLL_1 is clocking all RX channels and PLL_2 is clocking all TX
channels. You cannot have two adjacent PLLs simultaneously clocking cross-bank RX channels or two adjacent
PLLs simultaneously clocking TX channels. Cross-bank allows for all receiver channels on one side of the device to
be clocked on one clock while all transmitter channels on the device are clocked on the other center PLL. Crossbank
PLLs are supported at full-speed, 840 Mbps. For wire-bond devices, the full-speed is 624 Mbps.
These values show the channels available for each PLL without crossing another bank.
Altera Corporation
July 2005
5–55
Stratix Device Handbook, Volume 2
Software Support
Table 5–11. EP1S30 Differential Channels Note (1)
Package
780-pin
FineLine
BGA
956-pin
FineLine
BGA
1,020-pin
FineLine
BGA
Transmitter
Total
/Receiver Channels
Transmitter
(4)
70
Receiver
66
Transmitter
(4)
80 (7)
Receiver
80 (7)
Transmitter
(4)
80 (2) (7)
Receiver
80 (2) (7)
Maximum
Center Fast PLLs
Corner Fast PLLs (2), (3)
Speed
(Mbps) PLL1 PLL2 PLL3 PLL4 PLL7 PLL8 PLL9 PLL10
840
18
17
17
18
(6)
(6)
(6)
(6)
840 (5)
35
35
35
35
(6)
(6)
(6)
(6)
840
17
16
16
17
(6)
(6)
(6)
(6)
840 (5)
33
33
33
33
(6)
(6)
(6)
(6)
840
19
20
20
19
20
20
20
20
840 (5)
39
39
39
39
20
20
20
20
840
20
20
20
20
19
20
20
19
840 (5)
40
40
40
40
19
20
20
19
840
19
(1)
20
20
19
(1)
20
20
20
20
840 (5),(8)
39
(1)
39
(1)
39
(1)
39
(1)
20
20
20
20
840
20
20
20
20
19 (1)
20
20
19 (1)
840 (5),(8)
40
40
40
40
19 (1)
20
20
19 (1)
Table 5–12. EP1S40 Differential Channels (Part 1 of 2) Note (1)
Package
780-pin
FineLine
BGA
956-pin
FineLine
BGA
Transmitter
Total
/Receiver Channels
Transmitter
(4)
68
Receiver
66
Transmitter
(4)
80
Receiver
80
5–56
Stratix Device Handbook, Volume 2
Maximum
Speed
(Mbps)
Center Fast PLLs
Corner Fast PLLs (2), (3)
PLL1 PLL2 PLL3 PLL4 PLL7 PLL8 PLL9 PLL10
840
18
16
16
18
(6)
(6)
(6)
(6)
840 (5)
34
34
34
34
(6)
(6)
(6)
(6)
840
17
16
16
17
(6)
(6)
(6)
(6)
840 (5)
33
33
33
33
(6)
(6)
(6)
(6)
840
18
17
17
18
20
20
20
20
840 (5)
35
35
35
35
20
20
20
20
840
20
20
20
20
18
17
17
18
840 (5)
40
40
40
40
18
17
17
18
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Table 5–12. EP1S40 Differential Channels (Part 2 of 2) Note (1)
Package
1,020-pin
FineLine
BGA
Transmitter
Total
/Receiver Channels
Transmitter
(4)
Receiver
1,508-pin
FineLine
BGA
Transmitter
(4)
Receiver
80 (10)
(7)
80 (10)
(7)
80 (10)
(7)
80 (10)
(7)
Maximum
Speed
(Mbps)
Center Fast PLLs
Corner Fast PLLs (2), (3)
PLL1 PLL2 PLL3 PLL4 PLL7 PLL8 PLL9 PLL10
840
18
(2)
17
(3)
17
(3)
18
(2)
20
20
20
20
840 (5), (8)
35
(5)
35
(5)
35
(5)
35
(5)
20
20
20
20
840
20
20
20
20
18
(2)
17
(3)
17
(3)
18 (2)
840 (5), (8)
40
40
40
40
18
(2)
17
(3)
17
(3)
18 (2)
840
18
(2)
17
(3)
17
(3)
18
(2)
20
20
20
20
840 (5), (8)
35
(5)
35
(5)
35
(5)
35
(5)
20
20
20
20
840
20
20
20
20
18
(2)
17
(3)
17
(3)
18 (2)
840 (5), (8)
40
40
40
40
18
(2)
17
(3)
17
(3)
18 (2)
Table 5–13. EP1S60 Differential Channels (Part 1 of 2) Note (1)
Package
956-pin
FineLine
BGA
1,020-pin
FineLine
BGA
Transmitter
Total
/Receiver Channels
Transmitter
(4)
80
Receiver
80
Transmitter
(4)
80 (12)
(7)
Receiver
Altera Corporation
July 2005
80 (10)
(7)
Maximum
Center Fast PLLs
Corner Fast PLLs (2), (3)
Speed
PLL1 PLL2 PLL3 PLL4 PLL7 PLL8 PLL9 PLL10
(Mbps)
840
12
10
10
12
20
20
20
20
840 (5), (8)
22
22
22
22
20
20
20
20
840
20
20
20
20
12
10
10
12
840 (5), (8)
40
40
40
40
12
10
10
12
840
12
(2)
10
(4)
10
(4)
12
(2)
20
20
20
20
840 (5), (8)
22
(6)
22
(6)
22
(6)
22
(6)
20
20
20
20
840
20
20
20
20
12
(8)
10
(10)
10
(10)
12 (8)
840 (5), (8)
40
40
40
40
12
(8)
10
(10)
10
(10)
12 (8)
5–57
Stratix Device Handbook, Volume 2
Software Support
Table 5–13. EP1S60 Differential Channels (Part 2 of 2) Note (1)
Package
1,508-pin
FineLine
BGA
Transmitter
Total
/Receiver Channels
Transmitter
(4)
Receiver
80 (36)
(7)
80 (36)
(7)
Maximum
Center Fast PLLs
Corner Fast PLLs (2), (3)
Speed
PLL1 PLL2 PLL3 PLL4 PLL7 PLL8 PLL9 PLL10
(Mbps)
840
12
(8)
10
(10)
10
(10)
12
(8)
20
20
20
20
840 (5),(8)
22
(18)
22
(18)
22
(18)
22
(18)
20
20
20
20
840
20
20
20
20
12
(8)
10
(10)
10
(10)
12 (8)
840 (5),(8)
40
40
40
40
12
(8)
10
(10)
10
(10)
12 (8)
Table 5–14. EP1S80 Differential Channels (Part 1 of 2) Note (1)
Package
956-pin
FineLine
BGA
1,020-pin
FineLine
BGA
Transmitter
Total
/Receiver Channels
Transmitter
(4)
80 (40)
(7)
Receiver
80
Transmitter
(4)
80 (12)
(7)
Receiver
80 (10)
(7)
5–58
Stratix Device Handbook, Volume 2
Maximum
Center Fast PLLs
Corner Fast PLLs (2)
Speed
(Mbps) PLL1 PLL2 PLL3 PLL4 PLL7 PLL8 PLL9 PLL10
840
10
10
10
10
20
20
20
20
840 (5),(8)
20
20
20
20
20
20
20
20
840
20
20
20
20
10
10
10
10
840 (5),(8)
40
40
40
40
10
10
10
10
840
10
(2)
10
(4)
10
(4)
10
(2)
20
20
20
20
840 (5),(8)
20
(6)
20
(6)
20
(6)
20
(6)
20
20
20
20
840
20
20
20
20
10
(2)
10
(3)
10 (3)
10 (2)
840 (5),(8)
40
40
40
40
10
(2)
10
(3)
10 (3)
10 (2)
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Table 5–14. EP1S80 Differential Channels (Part 2 of 2) Note (1)
Package
1,508-pin
FineLine
BGA
Transmitter
Total
/Receiver Channels
Transmitter
(4)
Receiver
80 (72)
(7)
80 (56)
(7)
Maximum
Center Fast PLLs
Corner Fast PLLs (2)
Speed
(Mbps) PLL1 PLL2 PLL3 PLL4 PLL7 PLL8 PLL9 PLL10
840
10
(10)
10
(10)
10
(10)
10
(10)
20
(8)
20
(8)
20 (8)
20 (8)
840 (5),(8)
20
(20)
20
(20)
20
(20)
20
(20)
20
(8)
20
(8)
20 (8)
20 (8)
840
20
20
20
20
10
(14)
10
(14)
10
(14)
10
(14)
840 (5),(8)
40
40
40
40
10
(14)
10
(14)
10
(14)
10
(14)
Notes to Tables 5–11 through 5–14.
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
The first row for each transmitter or receiver reports the number of channels driven directly by the PLL. The second
row below it shows the maximum channels a PLL can drive if cross bank channels are used from the adjacent center
PLL. For example, in the 780-pin FineLine BGA EP1S30 device, PLL 1 can drive a maximum of 18 transmitter
channels at 840 Mbps or a maximum of 35 transmitter channels at 840 Mbps. The Quartus II software may also
merge transmitter and receiver PLLs when a receiver is driving a transmitter. In this case, one fast PLL can drive
both the maximum numbers of receiver and transmitter channels.
Some of the channels accessible by the center fast PLL and the channels accessible by the corner fast PLL overlap.
Therefore, the total number of channels is not the addition of the number of channels accessible by PLLs 1, 2, 3, and
4 with the number of channels accessible by PLLs 7, 8, 9, and 10. For more information on which channels overlap,
see the Fast PLL to High-Speed I/O Connections section in the relevant device pin table available on the web
(www.altera.com).
The corner fast PLLs in this device support a data rate of 840 Mbps for channels labeled “high” speed in the device
pin tables.
The numbers of channels listed include the transmitter clock output (tx_outclock) channel. You can use an extra
data channel if you need a DDR clock.
These channels span across two I/O banks per side of the device. When a center PLL clocks channels in the opposite
bank on the same side of the device it is called cross-bank PLL support. Both center PLLs can clock cross-bank
channels simultaneously if, for example, PLL_1 is clocking all receiver channels and PLL_2 is clocking all
transmitter channels. You cannot have two adjacent PLLs simultaneously clocking cross-bank receiver channels or
two adjacent PLLs simultaneously clocking transmitter channels. Cross-bank allows for all receiver channels on one
side of the device to be clocked on one clock while all transmitter channels on the device are clocked on the other
center PLL. Crossbank PLLs are supported at full-speed, 840 Mbps. For wire-bond devices, the full-speed is
624 Mbps.
PLLs 7, 8, 9, and 10 are not available in this device.
The number in parentheses is the number of slow-speed channels, guaranteed to operate at up to 462 Mbps. These
channels are independent of the high-speed differential channels. For the location of these channels, see the Fast
PLL to High-Speed I/O Connections section in the relevant device pin table available on the web (www.altera.com).
See device pin-outs channels marked “high” speed are 840 Mbps and “low” speed channels are 462 MBps.
The Quartus II software may also merge transmitter and receiver PLLs
when a receiver block is driving a transmitter block if the Use Common
PLLs for Rx and Tx option is set for both modules. The Quartus II
software does not merge the PLLs in multiple transmitter-only or
multiple receiver-only modules fed by the same clock.
Altera Corporation
July 2005
5–59
Stratix Device Handbook, Volume 2
Software Support
When you span two I/O banks using cross-bank support, you can route
only two load enable signals total between the plls. When you enable
rx_data_align, you use both rxloadena and txloadena of a PLL.
That leaves no loadena for the second PLL.
The only way you can use the rx_data_align is if one of the following
is true:
■
■
The RX PLL is only clocking RX channels (no resources for TX)
If all channels can fit in one I/O bank
LVDS Receiver Block
You only need to enter the input clock frequency, deserialization factor,
and the input data rate to implement an LVDS receiver block. The
Quartus II software then automatically sets the clock boost (W) factor for
the receiver. In addition, you can also indicate the clock and data
alignment for the receiver or add the pll_enable, rx_data_align,
and rx_locked output ports. Table 5–15 explains the function of the
available ports in the LVDS receiver block.
Table 5–15. LVDS Receiver Ports
Port Name
Direction
rx_in[number_of_channels - 1..0] Input
Function
Input Port
Source/Output Port
Destination
Input data channel
Pin
rx_inclock
Input
Reference input clock
Pin or output from a PLL
rx_pll_enable
Input
Enables fast PLL
Pin (1), (2), (3)
rx_data_align
Input
Control for the data
realignment circuitry
Pin or logic array (1),
(3), (4)
rx_locked
Output
Fast PLL locked pin
Pin or logic array (1), (3)
rx_out[Deserialization_factor *
number_of_channels -1..0]
Output
De-serialized data
Logic array
rx_outclock
Output
Internal reference clock
Logic array
Notes to Table 5–15:
(1)
(2)
(3)
(4)
This is an optional port.
Only one rx_pll_enable pin is necessary to enable all the PLLs in the device.
This is a non-differential pin.
See “Realignment Implementation” on page 5–28 for more information. For guaranteed performance and data
alignment, you must synchronize rx_data_align with rx_outclock.
5–60
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Use the altlvds MegaWizard Plug-In Manager to create an LVDS
receiver block. The following sections explain the parameters available in
the Plug-In Manager when creating an LVDS receiver block.
Page 3 of the altlvds_rx MegaWizard Plug-In Manager
On page 3 of the altlvds MegaWizard Plug-In Manager, you can
choose to create either an LVDS transmitter or receiver. Depending on
what you select, the MegaWizard Plug-In Manager provides you with
different options. Figure 5–40 shows page 3 of the altlvds MegaWizard
Plug-In Manager with options for creating an LVDS receiver.
Figure 5–40. Page 3 of the altlvds_rx MegaWizard Plug-In Manager
Number of Channels
The What is the number of channels? parameter specifies the number of
receiver channels required and the width of rx_out port. To set a fast
PLL to drive over 20 channels, type the required number in the Quartus II
window instead of choosing a number from the drop-down menu, which
only has selections of up to 20 channels.
Altera Corporation
July 2005
5–61
Stratix Device Handbook, Volume 2
Software Support
Deserialization Factor
Use the What is the deserialization factor? parameter to specify the
number of bits per channel. The Stratix LVDS receiver supports 4, 7, 8,
and 10 for deserialization factor (J) values. Based on the factor specified,
the Quartus II software determines the multiplication and/or division
factor for the LVDS PLL to deserialize the data.
See Table 5–5 for the differential bit naming convention. The parallel data
for the nth channel spans from the MSB (rx_out bit [(J × n) – 1]) to the
LSB (rx_out bit [J × (n – 1)]), where J is the deserialization factor. The
total width of the receiver rx_out port is equal to the number of channels
multiplied by your deserialization factor.
Input Data Rate
The What is the inclock boost(W)? parameter sets the data rate coming
into the receiver and is usually the deserialization factor (J) multiplied by
the inclock frequency. This parameter’s value must be larger than the
input clock frequency and has a maximum input data rate of 840 Mbps
for Stratix devices. You do not have to provide a value for the inclock
boost (W) when designing with Stratix devices because the Quartus II
software can calculate it automatically from this parameter and the clock
frequency or clock period.
The rx_outclock frequency is (W/J) × input frequency. The parallel
data coming out of the receiver has the same frequency as the
rx_outclock port. The clock-to-data alignment of the parallel data
output from the receiver depends on the What is the alignment of data
with respect to rx_inclock? parameter.
Data Alignment with Clock
The What is the alignment of data with respect to rx_inclock? parameter
adjusts the clock-to-data skew. For most applications, the data is source
synchronous to the clock. However, there are applications where you
must center-align the data with respect to the clock. You can use the What
is the alignment of data with respect to rx_inclock? parameter to align
the input data with respect to the rx_inclock port. The MegaWizard
Plug-In automatically calculates the phase for the fast PLL outputs from
the What is the alignment of data with respect to rx_inclock? parameter.
This parameter’s default value is EDGE_ALIGNED, and other values
available from the pull-down menu are EDGE_ALIGNED,
CENTER_ALIGNED, 45_DEGREES, 135_DEGREES, 180_DEGREES,
225_DEGREES, 270_DEGREES, and 315_DEGREES. CENTER_ALIGNED
is the same as 90 degrees aligned and is useful for applications like
HyperTransport technology.
5–62
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Clock Frequency or Clock Period
The fields in the Specify the input clock rate by box specify the input
frequency or the period of the input clock going into the fast PLL. When
using the same input clock to feed a transmitter and receiver
simultaneously, the Quartus II software can use one fast PLL for both the
transmitter and receiver.
Page 4 of the altlvds_rx MegaWizard Plug-In Manager
This section describes the parameters found on page 4 of the
altlvds_rx MegaWizard Plug-In Manager (see Figure 5–41).
Figure 5–41. Page 4 of the altlvds_rx MegaWizard Plug-In Manager
Altera Corporation
July 2005
5–63
Stratix Device Handbook, Volume 2
Software Support
Data Realignment
Check the Use the “rx_data_align” input port box within the Input Ports
box to add the rx_data_align output port and enable the data
realignment circuitry in Stratix SERDES. See “Receiver Data
Realignment” on page 5–25 for more information. If necessary, you can
create a state machine to send a pulse to the rx_data_align port to
realign the data coming in the LVDS receiver. You need to assert the port
for at least two clock cycles to enable the data realignment circuitry. Go
to the Altera web site at www.altera.com for a sample design written in
Verilog HDL.
For guaranteed performance when using data realignment, check the
Add Extra registers for rx_data_align input box when using the
rx_data_align port. The Quartus II software places one
synchronization register in the LE closest to the rx_data_align port.
Register Outputs
Check the Register outputs box to register the receiver’s output data. The
register acts as the module’s register boundary. If the module fed by the
receiver does not have a register boundary for the data, turn this option
on. The number of registers used is proportional to the deserialization
factor (J). The Quartus II software places the synchronization registers in
the LEs closest to the SERDES circuitry.
Use Common PLL for Both Transmitter & Receiver
Check the Use Common PLLs for Rx and Tx box to place both the LVDS
transmitter and the LVDS receiver in the same Stratix device I/O bank.
The Quartus II software allows the transmitter and receiver to share the
same fast PLL when they use the same input clock. Although you must
separate the transmitter and receiver modules in your design, the
Quartus II software merges the fast PLLs when appropriate and give you
the following message:
Receiver fast PLL <lvds_rx PLL name> and transmitter fast PLL <lvds_tx
PLL name> are merged together
The Quartus II software provides the following message when it cannot
merge the fast PLLs for the LVDS transmitter and receiver pair in the
design:
Can't merge transmitter-only fast PLL <lvds_tx PLL name> and receiveronly fast PLL <lvds_rx PLL name>
rx_outclock Resource
You can use either the global or regional clock for the rx_outclock
signal. If you select Auto in the Quartus II software, the tool uses any
available lines.
5–64
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
LVDS Transmitter Module
The Quartus II software calculates the inclock boost (W) factor for the
LVDS transmitter based on input data rate, input clock frequency, and
the deserialization factor. In addition to setting the data and clock
alignment, you can also set the outclock divide factor (B) for the
transmitter output clock and add the pll_enable, tx_locked, and
tx_coreclock ports. Table 5–16 explains the function of the available
ports in the LVDS transmitter block.
Table 5–16. LVDS Transmitter Ports
Port Name
Direction
Function
Input port
Source/Output port
Destination
tx_in[Deserialization_factor *
number_of_channels - 1..0]
Input
Input data
Logic array
tx_inclock
Input
Reference input clock
Pin or output clock
from a PLL
tx_pll_enable
Input
tx_out[number_of_channels - 1..0] Output
Fast PLL enable
Pin (1), (2), (3)
Serialized LVDS data
signal
Pin
tx_outclock
Output
External reference clock
Pin
tx_coreclock
Output
Internal reference clock
Pin, logic array, or
input clock to a fast
PLL (1)
tx_locked
Output
Fast PLL locked pin
Pin or logic array (1),
(2), (3)
Notes to Table 5–16:
(1)
(2)
(3)
This is an optional port.
Only one tx_pll_enable pin is necessary to enable all the PLLs in the device.
This is a non-differential pin.
You can also use the altlvds MegaWizard Plug-In Manager to create an
LVDS transmitter block. The following sections explain the parameters
available in the Plug-In Manager when creating an LVDS transmitter
block.
Page 3 of the altlvds_tx MegaWizard Plug-In Manager
This section describes the parameters found on page 3 of the
altlvds_tx MegaWizard Plug-In Manager (see Figure 5–42).
Altera Corporation
July 2005
5–65
Stratix Device Handbook, Volume 2
Software Support
Figure 5–42. Page 3 of the Transmitter altlvds MegaWizard Plug-In Manager
Number of Channels
The What is the number of channels? parameter specifies the number of
transmitter channels required and the width of the tx_in port. You can
have more than 20 channels in a transmitter or receiver module by typing
in the required number instead of choosing a number from the drop
down menu, which only has selections of up to 20 channels.
Deserialization Factor
The What is the deserialization factor? parameter specifies the number
of bits per channel. The transmitter block supports deserialization factors
of 4, 7, 8, and 10. Based on the factor specified, the Quartus II software
determines the multiplication and/or division factor for the LVDS PLL in
order to serialize the data.
Table 5–5 on page 5–32 lists the differential bit naming convention. The
parallel data for the nth channel spans from the MSB (rx_out bit
[(J × n) – 1]) to the LSB (rx_out bit [J × (n – 1)]), where J is the
5–66
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
deserialization factor. The total width of the tx_in port of the transmitter
is equal to the number of channels multiplied by the deserialization
factor.
Outclock Divide Factor
The What is the Output data rate? parameter specifies the ratio of the
tx_outclock frequency compared to the data rate. The default value for
this parameter is the value of the deserialization factor parameter. The
tx_outclock frequency is equal to [W/B] x input clock frequency.
There is also an optional tx_coreclock port which has the same
frequency as the [W/J] × input frequency.
The outclock divide factor is useful for applications that do not require
the data rate to be the same as the clock frequency. For example,
HyperTransport technology uses a half-clock data rate scheme where the
clock frequency is half the data rate. Table 5–17 shows the supported
outclock divide factor for a given deserialization factor.
Table 5–17. Deserialization Factor (J) vs. Outclock Divide Factor (B)
Deserialization Factor (J)
Outclock Divide Factor (B)
4
1, 2, 4
7
1, 7(1)
8
1, 2, 4, 8
10
1, 2, 10
Note to Table 5–17:
(1)
The clock does not have a 50% duty cycle when b=7 in x7 mode.
Output Data Rate
The What is the Output data rate parameter specifies the data rate out of
the fast PLL and determines the input clock boost/multiplication factor
needed for the transmitter. This parameter must be larger than the input
clock frequency and has a maximum rate of 840 Mbps for Stratix devices.
The input clock boost factor (W) is the output data rate divided by the
input clock frequency. The Stratix SERDES circuitry supports input clock
boost factors of 4, 7, 8, or 10. The maximum output data rate is 840 Mbps,
while the clock has a maximum output of 500 MHz.
Data Alignment with Clock
Use the What is the alignment of data with respect to tx_inclock?
parameter and the What is the alignment of tx_outclock? to align the
input and output data, respectively, with the clock. For most applications,
the data is edge-aligned with the clock. However, there are applications
where the data must be center-aligned with respect to the clock. With
Altera Corporation
July 2005
5–67
Stratix Device Handbook, Volume 2
Software Support
Stratix devices, you can align the input data with respect to the
tx_inclock port and align the output data with respect to the
tx_outclock port. The MegaWizard Plug-In Manager uses the
alignment of input and output data to automatically calculate the phase
for the fast PLL outputs. Both of these parameters default to
EDGE_ALIGNED, and other values are CENTER_ALIGNED, 45_DEGREES,
135_DEGREES, 180_DEGREES, 225_DEGREES, 270_DEGREES, and
315_DEGREES. CENTER_ALIGNED is the same as 180 degrees aligned
and is required for the HyperTransport technology I/O standard.
Clock Frequency & Clock Period
The fields in the Specify the input clock rate by box specify either the
frequency or the period of the input clock going into the fast PLL.
However, you cannot specify both. If your design uses the same input
clock to feed a transmitter and a receiver module simultaneously, the
Quartus II software can merge the fast PLLs for both the transmitter and
receiver when the Use common PLLs for Tx & Rx option is turned on.
Page 4 of the altlvds_tx MegaWizard Plug-In Manager
This section describes the parameters found on page 4 of the
altlvds_tx MegaWizard Plug-In Manager (see Figure 5–43).
5–68
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Figure 5–43. Page 4 of the Transmitter altlvds MegaWizard Plug-In Manager
Registered Inputs
Check the Register inputs box if the input data to the transmitter is not
registered just before it feeds the transmitter module. You can choose
either tx_clkin or tx_coreclk to clock the transmitter data
(tx_in[]) signal. This serves as the register boundary. The number of
registers used is proportional to the deserialization factor (J). The
Quartus II software places the synchronization registers with the LEs in
the same row and closest to the SERDES circuitry.
Use Common PLL for Transmitter & Receiver
Check the Use Common PLLs for Rx and Tx box to place both the LVDS
transmitter and receiver in the same I/O bank in Stratix devices. The
Quartus II software also allows the transmitter and receiver to share the
PLL when the same input clock is used for both. Although you must
Altera Corporation
July 2005
5–69
Stratix Device Handbook, Volume 2
Software Support
separate the transmitter and receiver in your design, the Quartus II
software merges the fast PLLs when appropriate and gives you the
following message:
Receiver fast PLL <lvds_rx pll name> and transmitter fast PLL
<lvds_tx pll name> are merged together
The Quartus II software gives the following message when it cannot
merge the fast PLLs for the LVDS transmitter and receiver pair in the
design:
Can't merge transmitter-only fast PLL
<lvds_tx pll name> and receiver-only fast PLL
<lvds_rx pll name>
tx_outclock Resource
You can use either the global or regional clock for the tx_outclock
signal. If you select Auto in the Quartus II software, the tool uses any
available lines.
SERDES Bypass Mode
You can bypass the SERDES block if your data rate is less than 624 Mbps,
and you must bypass the SERDES block for the ×1 and ×2 LVDS modules.
Since you cannot route the fast PLL output to an output pin, you must
create additional DDR I/O circuitry for the transmitter clock output. To
create an ×J transmitter output clock, instantiate an alt_ddio
megafunction clocked by the ×J clock with datain_h connected to VCC
and datain_l connected to GND.
×1 Mode
For ×1 mode, you only need to specify the I/O standard of the pins to tell
the Quartus II software that you are using differential signaling.
However, Altera recommends using the DDRIO circuitry when the input
or output data rate is higher than 231 Mbps. The maximum output clock
frequency for ×1 mode is 420 MHz.
×2 Mode
You must use the DDRIO circuitry for ×2 mode. The Quartus II software
provides the altddio_in and altddio_out megafunctions to use for
×2 receiver and ×2 transmitter, respectively. The maximum data rate in
×2 mode is 624 Mbps. Figure 5–44 shows the schematic for using DDR
circuitry in ×2 mode.
5–70
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Figure 5–44. LVDS x2 Mode Schematic Using DDR I/O Circuitry
DDIO In
RXp
RXn
datain[0]
inclock
DDIO Out
dataout_h[0]
dataout_l[0]
datain_h[0]
dataout[0]
TXp
TXn
datain_l[0]
Custom Logic
outclock
DDIO Out
RX_PLL
rx_inclk
inclock
/1 clock1
VCC
datain_h[0]
/2 clock0
GND
datain_l[0]
dataout[0]
tx_outclk
outclock
The transmitter output clock requires extra DDR output circuitry that has
the input high and input low connected to VCC and GND respectively. The
output clock frequency is the same as the input frequency of the DDR
output circuitry.
Other Modes
For other modes, you can still to use the DDR circuitry for better
frequency performance. You can use either the LEs or the M512 RAM
block for the deserialization.
M512 RAM Block as Serializer/Deserializer Interface
In addition to using the DDR circuitry and the M512 RAM block, you
need two extra counters per memory block to provide the address for the
memory: a fast counter powering up at 0 and a slow counter powering up
at 2. The M512 RAM block is configured as a simple dual-port memory
block, where the read enable and the write enable signals are always tied
high. Figures 5–45 and 5–46 show the block diagram for the SERDES
bypass receiver and SERDES bypass transmitter, respectively.
Altera Corporation
July 2005
5–71
Stratix Device Handbook, Volume 2
Software Support
Figure 5–45. SERDES Bypass LVDS Receiver Using M512 RAM Block as the Deserializer
waddr[7..5]
Simple Dual Port
RX_SESB
512 Bits
DDIO In
RXp
datain[0]
RXn
inclock
dataout_h[0]
datain[1..0]
dataout_l[0]
waddr[7..0]
dataout[7..0]
Core data
wclock
clock
q[4..0]
raddr[5..0]
rclock
W-UpCounter
RX_PLL
rx_inclk
inclock
÷1 clock1
÷2 clock0
clock
raddr[5..3]
q[2..0]
R-UpCounter
Core clock
Figure 5–46. SERDES Bypass LVDS Transmitter Using M512 RAM Block as Deserializer
waddr[7..5]
Simple Dual Port ×2×8
TX_SESB
512 Bits
core_data
datain[7..0]
clock
q[2..0]
DDIO Out
dataout[7..0]
datain_h[0]
dataout_h[0]
waddr[5..0]
datain_l[0]
dataout_l[0]
wclock
outclock
TXp
TXn
raddr[7..0]
rclock
W-UpCounter
RX_PLL
core_clk
inclock
÷1 clock1
×2 clock0
clock
q[5..0]
R-UpCounter
raddr[5..3]
RX_PLL
VCC
datain_h[0]
/1 clock1
GND
datain_l[0]
/2 clock0
tx_outclk
outclock
5–72
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
For the transmitter, the read counter is the fast counter and the write
counter is the slow counter. For the receiver, the write counter is the fast
counter and the read counter is the slow counter. Tables 5–18 and 5–19
provide the address counter configurations for the transmitter and the
receiver, respectively.
Table 5–18. Address Counters for SERDES Bypass LVDS Receiver
M512 Mode
Deserialization
Factor
Write Up-Counter
(Fast Counter)
Read Up-Counter
(Slow Counter)
Width
Starts at
Width
Starts at
Write
Read
Invalid Initial Cycles
×2×4
4
4
0
3
2
12
6
×2×8
8
5
0
3
2
24
6
×4×16
8
5
0
3
2
24
6
×2×16
16
6
0
3
2
48
6
Table 5–19. Address Counters for SERDES Bypass LVDS Transmitter
M512 Mode
Deserialization
Factor
Write Up-Counter
(Fast Counter)
Read Up-Counter
(Slow Counter)
Width
Starts at
Width
Starts at
Write
Read
Invalid Initial Cycles
×2×4
4
4
0
3
2
2
4
×2×8
8
5
0
3
2
2
8
×4×16
8
5
0
3
2
2
8
×2×16
16
6
0
3
2
2
16
In different M512 memory configurations, the counter width is smaller
than the address width, so you must ground some of the most significant
address bits. Table 5–20 summarizes the address width, the counter
width, and the number of bits to be grounded.
Table 5–20. Address & Counter Width
M512 Mode
Number of Grounded Bits
Write Counter Read Counter Write Address Read Address
Width
Width
Width
Width
Write Address Read Address
×2×4
4
3
8
7
4
4
×2×8
5
3
8
6
3
3
×4×16
6
3
7
5
1
2
×2×16
5
3
8
5
3
2
Altera Corporation
July 2005
5–73
Stratix Device Handbook, Volume 2
Software Support
Logic Array as Serializer/Deserializer Interface
The design can use the lpm_shift_reg megafunction instead of a
simple dual port memory block to serialize/deserialize data. The receiver
requires an extra flip-flop clocked by the slow clock to latch on to the
deserialized data. The transmitter requires a counter to generate the
enable signal for the shift register to indicate the times to load and
serialize the data. Figures 5–47 and 5–48 show the schematic of the ×8
LVDS receiver and ×8 LVDS transmitter, respectively, with the logic
array performing the deserialization.
This scheme can also be used for APEX II and Mercury device flexible
LVDS solutions.
Figure 5–47. SERDES Bypass LVDS Receiver with Logic Array as Deserializer
clock
Shift
Register
Serial
data in
×4 clock0
Clock
PLL
data_h
data
data[1, 3, 5, 7]
data[7..0]
D
DDR
Input
DFF[7..0]
Q
Data to
logic array
÷2 clock1
data_l
data
data[0, 2, 4, 6]
Shift
Register
CLK
clock
rx_clk
5–74
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
High-Speed Differential I/O Interfaces in Stratix Devices
Figure 5–48. SERDES Bypass LVDS Transmitter with Logic Array as Deserializer
Counter
Shift
Register
clock
load
data
Data[7..0]
data_l
×4 clock
Clock
PLL
×1 clock
DDR
Output
Shift
Register
data
Serial
data out
data_h
load
clock
tx_clk
Summary
Altera Corporation
July 2005
The Stratix device family of flexible, high-performance, high-density
PLDs delivers the performance and bandwidth necessary for complex
system-on-a-programmable-chip (SOPC) solutions. Stratix devices
support multiple I/O protocols to interface with other devices within the
system. Stratix devices can easily implement processing-intensive datapath functions that are received and transmitted at high speeds. The
Stratix family of devices combines a high-performance enhanced PLD
architecture with dedicated I/O circuitry in order to provide I/O
standard performances of up to 840 Mbps.
5–75
Stratix Device Handbook, Volume 2
Summary
5–76
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Section IV. Digital Signal
Processing (DSP)
This section provides information for design and optimization of digital
signal processing (DSP) functions and arithmetic operations in the onchip DSP blocks.
It contains the following chapters:
Revision History
■
Chapter 6, DSP Blocks in Stratix & Stratix GX Devices
■
Chapter 7, Implementing High Performance DSP Functions
in Stratix & Stratix GX Devices
The table below shows the revision history for Chapters 6 and 7.
Chapter
Date/Version
6
July 2005, v2.2
●
Changed Stratix GX FPGA Family data sheet reference to
Stratix GX Device Handbook, Volume 1.
Changes Made
September 2004, v2.1
●
Updated “Software Support” on page 6–28.
Deleted “Quartus II DSP Megafunctions” section. It was replaced by
the updated “Software Support” on page 6–28
Replaced references to AN 193 and AN 194 with a new reference
on page 6–28.
●
●
7
July 2003, v2.0
●
Minor content change.
April 2003, v1.0
●
No new changes in Stratix Device Handbook v2.0.
September 2004, v1.1
●
Corrected spelling error.
April 2003, v1.0
●
No new changes in Stratix Device Handbook v2.0.
Altera Corporation
Section IV–1
Digital Signal Processing (DSP)
Section IV–2
Stratix Device Handbook, Volume 2
Altera Corporation
6. DSP Blocks in Stratix &
Stratix GX Devices
S52006-2.2
Introduction
Traditionally, designers had to make a trade-off between the flexibility of
off-the-shelf digital signal processors and the performance of custombuilt devices. Altera® Stratix® and Stratix GX devices eliminate the need
for this trade-off by providing exceptional performance combined with
the flexibility of programmable logic devices (PLDs). Stratix and
Stratix GX devices have dedicated digital signal processing (DSP) blocks,
which have high-speed parallel processing capabilities, that are
optimized for DSP applications. DSP blocks are ideal for implementing
DSP applications that need high data throughput.
The most commonly used DSP functions are finite impulse response (FIR)
filters, complex FIR filters, infinite impulse response (IIR) filters, fast
Fourier transform (FFT) functions, discrete cosine transform (DCT)
functions, and correlators. These functions are the building blocks for
more complex systems such as wideband code division multiple access
(W-CDMA) basestations, voice over Internet protocol (VoIP), and highdefinition television (HDTV).
Although these functions are complex, they all use similar building
blocks such as multiply-adders and multiply-accumulators. Stratix and
Stratix GX DSP blocks combine five arithmetic operations—
multiplication, addition, subtraction, accumulation, and summation—to
meet the requirements of complex functions and to provide improved
performance.
This chapter describes the Stratix and Stratix GX DSP blocks, and
explains how you can use them to implement high-performance DSP
functions. It addresses the following topics:
■
■
■
f
Altera Corporation
July 2005
Architecture
Operational Modes
Software Support
See the Stratix Device Family Data Sheet section of the Stratix Device
Handbook, Volume 1 and the Stratix GX Device Family Data Sheet section of
the Stratix GX Device Handbook, Volume 1 for more information on Stratix
and Stratix GX devices, respectively.
6–1
DSP Block Overview
DSP Block
Overview
Each Stratix and Stratix GX device has two columns of DSP blocks that
efficiently implement multiplication, multiply accumulate (MAC), and
filtering functions. Figure 6–1 shows one of the columns with
surrounding LAB rows. You can configure each DSP block to support:
■
■
■
Eight 9 × 9 bit multipliers
Four 18 × 18 bit multipliers
One 36 × 36 bit multiplier
Figure 6–1. DSP Blocks Arranged in Columns
DSP Block
Column
8 LAB
Rows
DSP Block
The multipliers can then feed an adder or an accumulator block,
depending on the DSP block operational mode. Additionally, you can use
the DSP block input registers as shift registers to implement applications
such as FIR filters efficiently. The number of DSP blocks per column
6–2
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
DSP Blocks in Stratix & Stratix GX Devices
increases with device density. Tables 6–1 and 6–2 describe the number of
DSP blocks in each Stratix and Stratix GX device, respectively, and the
multipliers that you can implement.
Table 6–1. Number of DSP Blocks in Stratix Devices Note (1)
Device
DSP Blocks
9 × 9 Multipliers
18 × 18 Multipliers 36 × 36 Multipliers
EP1S10
6
48
24
6
EP1S20
10
80
40
10
EP1S25
10
80
40
10
EP1S30
12
96
48
12
EP1S40
14
112
56
14
EP1S60
18
144
72
18
EP1S80
22
176
88
22
Table 6–2. Number of DSP Blocks in Stratix GX Devices Note (1)
Device
DSP Blocks
9 × 9 Multipliers
EP1SGX10C
6
48
18 × 18 Multipliers 36 × 36 Multipliers
24
6
EP1SGX10D
6
48
24
6
EP1SGX25C
10
80
40
10
EP1SGX25D
10
80
40
10
EP1SGX25F
10
80
40
10
EP1SGX40D
14
112
56
14
EP1SGX40G
14
112
56
14
Note to Tables 6–1 and 6–2:
(1)
Each device has either the number of 9 × 9-, 18 × 18-, or 36 × 36-bit multipliers shown.The total number of
multipliers for each device is not the sum of all the multipliers.
Altera Corporation
July 2005
6–3
Stratix Device Handbook, Volume 2
DSP Block Overview
Figure 6–2 shows the DSP block operating as an 18 × 18 multiplier.
Figure 6–2. DSP Block in 18 × 18 Mode
Optional Serial Shift Register
Inputs from Previous
DSP Block
From the Row
Interface Block
D
PRN
Q
ENA
CLRN
D
PRN
Q
Multiplier Block
D
Adder Output Block
Output
Register
PRN
Q
ENA
CLRN
ENA
CLRN
Adder/
Subtractor/
Accumulator
D
PRN
Q
ENA
CLRN
D
PRN
Q
D
PRN
Q
ENA
CLRN
Summation
Block
ENA
CLRN
Adder
D
PRN
Q
ENA
CLRN
D
PRN
Q
D
PRN
Q
ENA
CLRN
ENA
CLRN
Adder/
Subtractor/
Accumulator
D
PRN
Q
ENA
CLRN
D
PRN
Q
Pipeline
Register
D
PRN
Q
ENA
CLRN
ENA
CLRN
6–4
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
DSP Blocks in Stratix & Stratix GX Devices
Architecture
The DSP block consists of the following elements:
■
■
■
■
■
■
A multiplier block
An adder/subtractor/accumulator block
A summation block
An output interface
Output registers
Routing and control signals
Multiplier Block
Each multiplier block has input registers, a multiplier stage, and a
pipeline register. See Figure 6–3.
Figure 6–3. Multiplier Block Architecture
signa
signb
aclr[3..0]
clock[3..0]
ena[3..0]
shiftinb
shiftina
D
Data A
Q
ENA
CLRN
D
Q
ENA
Data Out
to Adder
Blocks
CLRN
D
Data B
Q
ENA
CLRN
shiftoutb
Altera Corporation
July 2005
shiftouta
6–5
Stratix Device Handbook, Volume 2
Architecture
Input Registers
Each operand feeds an input register or the multiplier directly. The DSP
block has the following signals (one of each controls every input and
output register):
■
■
■
clock[3..0]
ena[3..0]
aclr[3..0]
The input registers feed the multiplier and drive two dedicated shift
output lines, shiftouta and shiftoutb. The shift outputs from one
multiplier block directly feed the adjacent multiplier block in the same
DSP block (or the next DSP block), as shown in Figure 6–4 on page 6–7, to
form a shift register chain. This chain can terminate in any block, i.e., you
can create any length of shift register chain up to 224 registers. A shift
register is useful in DSP applications such as FIR filters. When
implementing 9 × 9 and 18 × 18 multipliers, you do not need external
logic to create the shift register chain because the input shift registers are
internal to the DSP block. This implementation greatly reduces the
required LE count and routing resources, and produces repeatable
timing.
6–6
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
DSP Blocks in Stratix & Stratix GX Devices
Figure 6–4. Shift Register Chain
DSP Block 0
Data A
D
Q
ENA
A[n] × B[n]
CLRN
Data B
D
Q
Q
D
ENA
CLRN
ENA
CLRN
shiftoutb
shiftouta
D
Q
ENA
A[n - 1] × B[n - 1]
CLRN
D
Q
D
Q
ENA
CLRN
ENA
CLRN
shiftoutb
shiftouta
DSP Block 1
D
Q
ENA
A[n - 2] × B[n - 2]
CLRN
D
Q
D
Q
ENA
CLRN
ENA
CLRN
shiftoutb
Altera Corporation
July 2005
shiftouta
6–7
Stratix Device Handbook, Volume 2
Architecture
Multiplier Stage
The multiplier stage supports 9 × 9, 18 × 18, or 36 × 36 multiplication.
(The multiplier stage also support smaller multipliers. See “Operational
Modes” on page 6–18 for details.) Based on the data width, a single DSP
block can perform many multiplications in parallel.
The multiplier operands can be signed or unsigned numbers. Two
signals, signa and signb, indicate the representation of the two
operands. For example, a logic 1 on the signa signal indicates that data
A is a signed number; a logic 0 indicates an unsigned number. The result
of the multiplication is signed if any one of the operands is a signed
number, as shown in Table 6–3.
Table 6–3. Multiplier Signed Representation
Data A
Data B
Result
Unsigned
Unsigned
Unsigned
Unsigned
Signed
Signed
Signed
Unsigned
Signed
Signed
Signed
Signed
The signa and signb signals affect the entire DSP block. Therefore, all
of the data A inputs feeding the same DSP block must have the same sign
representation. Similarly, all of the data B inputs feeding the same DSP
block must have the same sign representation. The multiplier offers full
precision regardless of the sign representation.
1
By default, the Altera Quartus® II software sets the multiplier to
perform unsigned multiplication when the signa and signb
signals are not used.
Pipeline Registers
The output from the multiplier can feed a pipeline register or be
bypassed. You can use pipeline registers for any multiplier size;
pipelining is useful for increasing the DSP block performance,
particularly when using subsequent adder stages.
1
6–8
Stratix Device Handbook, Volume 2
In the DSP block, pipelining improves the performance of
36 × 36 multipliers. For 18 × 18 multipliers and smaller,
pipelining adds latency but does not improve performance.
Altera Corporation
July 2005
DSP Blocks in Stratix & Stratix GX Devices
Adder/Output Block
The adder/output block has the following elements (See Figure 6–5 on
page 6–10):
■
■
■
■
An adder/subtractor/accumulator block
A summation block
An output select multiplexer
Output registers
You can configure the adder/output block as:
■
■
■
■
■
A pure output interface
An accumulator
A simple one-level adder
A two-level adder with dynamic addition/subtraction control on the
first-level adder
The final stage of a 36-bit multiplier
The output select multiplexer sets the output of the DSP block. You can
register the adder/output block’s output using the output registers.
1
Altera Corporation
July 2005
You cannot use the adder/output block independently of the
multiplier.
6–9
Stratix Device Handbook, Volume 2
Architecture
Figure 6–5. Adder/Output Block
Accumulator Feedback
Output Select
Multiplexer
accum_sload0
Result A
Output
Registers
overflow0
Adder/
Subtractor/
Accumulator 0
addnsub1
Result B
signa
Adder
signb
Result C
Adder/
Subtractor/
Accumulator 1
addnsub3
overflow1
Result D
Accumulator Feedback
accum_sload1
Adder/Subtractor/Accumulator Block
The adder/subtractor/accumulator is the first level of the adder/output
block. You can configure the block as an accumulator or as an
adder/subtractor.
Accumulator
When the adder/subtractor/accumulator is configured as an
accumulator, the output of the adder/output block feeds back to the
accumulator as shown in Figure 6–5. You can use the
6–10
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
DSP Blocks in Stratix & Stratix GX Devices
accum_sload[1..0] signals to clear the accumulator asynchronously.
This action is not the same as resetting the output registers. You can clear
the accumulation and begin a new one without losing any clock cycles.
The overflow signal goes high on the positive edge of the clock when
the accumulator overflows or underflows. In the next clock cycle,
however, the overflow signal resets to zero even though an overflow (or
underflow) occurred in the previous clock cycle. Use a latch to preserve
the overflow condition indefinitely (until the latch is cleared).
Adder/Subtractor
The addnsub[1..0] signals select addition or subtraction: high for
addition and low for subtraction. You can control the addnsub[1..0]
signals using external logic; therefore, the first-level block can switch
from an adder to a subtractor dynamically, simply by changing the
addnsub[1..0] signals. If the first stage is configured as a subtractor,
the output is A - B and C - D.
The adder/subtractor also uses two signals, signa and signb, like the
multiplier block. These signals indicate the sign representation of both
operands together. You can register the signals with a latency of 1 or 2
clock cycles.
Summation Block
The output from the adder/subtractor feeds to an optional summation
block, which is an adder block that sums the outputs of the
adder/subtractor. The summation block is important in applications
such as FIR filters.
Output Select Multiplexer
The outputs from the various elements of the adder/output block are
routed through an output select multiplexer. Based on the DSP block
operational mode, the outputs of the multiplier block,
adder/subtractor/accumulator, or summation block feed straight to the
output, bypassing the remaining blocks in the DSP block.
1
The output select multiplier configuration is configured
automatically by software.
Output Registers
You can use the output registers to register the DSP block output. Like the
input registers, the output registers are controlled by the four
clock[3..0], aclr[3..0], and ena[3..0] signals. You can use the
output registers in any DSP block operational mode.
Altera Corporation
July 2005
6–11
Stratix Device Handbook, Volume 2
Architecture
1
The output registers form part of the accumulator in the
multiply-accumulate mode.
Routing Structure & Control Signals
This section describes the interface between the DSP blocks and the row
interface blocks. It also describes how the DSP block generates control
signals and how the signals route from the row interface to the DSP block.
DSP Block Interface
The DSP blocks are organized in columns, which provides efficient
horizontal communication between the blocks and the column-based
memory blocks. The DSP block communicates with other parts of the
device through an input and output interface. Each DSP block, including
the input and output interface, is 8 logic array blocks (LABs) long.
The DSP block and row interface blocks consist of eight blocks that
connect to eight adjacent LAB rows on the left and right. Each of the eight
blocks has two regions: right and left, one per row. The DSP block
receives 144 data input signals and 18 control signals for a total of
162 input signals. This block drives out 144 data output signals; 2 of the
data signals can be used as overflow signals (overflow). Figure 6–6
provides an overview of the DSP block and its interface to adjacent LABs.
Figure 6–6. DSP Block Interface to Adjacent LABs
DSP Block & Row Interface
144
8 LAB 162
Rows
Row
Interfaces
0 through 7
Data
DSP
Block
144
8 LAB
Rows
18
Control
DSP Block
Input Interface
DSP Block
Output Interface
Input Interface
The DSP block input interface has 162 input signals from adjacent LABs;
18 data signals per row and 18 control signals per block.
Output Interface
The DSP block output interface drives 144 outputs to adjacent LABs, 18
signals per row from 8 rows.
6–12
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
DSP Blocks in Stratix & Stratix GX Devices
Because the DSP block outputs communicate horizontally, and because
each DSP block row has more outputs than an LAB (18 from the DSP
block compared to 10 from an LAB), the DSP block has double the
number of row channel drivers compared to an LAB. The DSP block has
the same number of row channels, but the row channels are staggered as
if there were two LABs within each block. The DSP blocks have the same
number of column channels as LABs because DSP blocks communicate
primarily through row channels.
Row Interface Block
Each row interface block connects to the DSP block row structure with
21 signals. Because each DSP block has eight row interface blocks, this
block receives 162 signals from the eight row interfaces. Of the
162 signals, 144 are data inputs and 18 are control signals. Figure 6–7 on
page 6–14 shows one row block within the DSP block.
Altera Corporation
July 2005
6–13
Stratix Device Handbook, Volume 2
Architecture
Figure 6–7. DSP Row Interface Block
C4 and C8
Interconnects
DirectLink Interconnect
from Adjacent LAB
R4 and R8 Interconnects
Nine DirectLink Outputs
to Adjacent LABs
DirectLink Interconnect
from Adjacent LAB
18
DSP Block
Row Structure
LAB
10
LAB
9
9
10
3
Control
18
18
[17..0]
[17..0]
Row Interface
Block
DSP Block to
LAB Row Interface
Block Interconnect Region
18 Inputs per Row
18 Outputs per Row
Control Signals in the Row Interface Block
The DSP block has a set of input registers, a pipeline register, and an
output register. Each register is grouped in banks that share the same
clock and clear resources:
■
■
■
1- to 9-bit banks for the input register
1- to 18-bit banks for the pipeline register
18 bits for the output register
6–14
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
DSP Blocks in Stratix & Stratix GX Devices
The row interface block generates the control signals and routes them to
the DSP block. Each DSP block has 18 control signals:
■
■
■
■
■
■
Four clock signals (clock[3..0]), which are available to each bank
of DSP blocks
Four clear signals (aclr[3..0]), which are available to each bank
of DSP blocks
Four clock enable signals (ena[3..0]), which the whole DSP block
can use
signa and signb, which are specific to each DSP block
addnsub[1..0] signals
accum_sload[1..0] signals
The signa, signb, and addnsub[1..0], accum_sload[1..0]
signals have independent clocks and clears and can be registered
individually. When each 18 × 18 multiplier in the DSP block splits in half
to two 9 × 9 multipliers, each 9 × 9 multiplier has independent control
signals. Figure 6–8 shows the DSP block row interface and shows how it
generates the data and control signals.
Figure 6–8. DSP Block Row Interface
30 Local
Interconnect
Signals
21 Signals for
Data to Input
Register
DSP Row
LAB
Row
Clocks
Bit 0
Bit 1
Bit 2
Bit 3
Bit 4
Bit 5
Bit 6
Bit 7
DSP Block
Row Interface
Bit 8
DSP Row 2
DSP Row 1
DSP Row 3
DSP Row
Unit Control
Block
DSP Row
Bit 9
Bit 10
3
Detail of
1 DSP Row
DSP Row 4
DSP Row 5
Input
Registers
DSP
Block
DSP Row 6
DSP Row 7
DSP Row 8
Bit 11
Bit 12
Bit 13
Bit 14
Bit 15
Bit 16
Bit 17
Altera Corporation
July 2005
6–15
Stratix Device Handbook, Volume 2
Architecture
The DSP block interface generates the clock signals from LAB row clocks
or the local interconnect. The clear signals are generated from the local
interconnects within each DSP block row interface or from LAB row
clocks. The four clock enable signals are generated from the 30 local
interconnects from the same LAB rows that generate the clock signals.
The clock enable is paired with the clock because the enable logic is
implemented at the interface. Figure 6–9 shows the signal distribution
within the row interface block.
Figure 6–9. DSP Block Row Interface Signal Distribution
aclr[3..0]
ena[3..0]
data[17..0]
18
4
4
clock[3..0]
4
Input
18 Registers
18-Bit Data Routed
from 30 Local
Interconnects
18
A1
Four Clock Enable
Signals Routed from
30 Local Interconnects
18 × 18
Multiplier
Row 1
B1
18
Four Clear Signals
Routed from 30 Local
Interconnects or LAB
Row Clock
Four Clock Signals
Routed from LAB
Row Clock or Local
Interconnect
18
Row 2
18
18
A4
18 × 18
Multiplier
Row 7
B4
18
18
Row 8
6–16
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
DSP Blocks in Stratix & Stratix GX Devices
Each row block provides 18 bits of data to the multiplier (i.e., one of the
operands to the multiplier), which are routed through the 30 local
interconnects within each DSP row interface block. Any signal in the
device can be the source of the 18-bit multiplier data, by connecting to the
local row interconnect through any row or column.
Each control signal routes through one of the eight rows of the DSP block.
Table 6–4 shows the 18 control signals and the row to which each one
routes.
Table 6–4. Control Signals in DSP Block
Signal Name
Row
signa
1
signb
6
addnsub1
3
addnsub3
7
accum_sload0 2
accum_sload1 7
Description
DSP block-wide signed and unsigned control signals for all multipliers.
The multiplier outputs are unsigned only if both signa and signb are
low.
Controls addition or subtraction of the two one-level adders. The
addnsub0 signal controls the top two one-level adders; the addnsub1
signal controls the bottom two one-level adders. A high indicates
addition; a low indicates subtraction.
Resets the feedback input to the accumulator. The signal
asynchronously clears the accumulator and allows new accumulation to
begin without losing any clock cycles. The accum_sload0 controls the
top two one-level adders, and the accum_sload1 controls the bottom
two one-level adders. A low is for normal accumulation operations and
a high is for zeroing the accumulator.
clock0
3
clock1
4
DSP block-wide clock signals.
clock2
5
clock3
6
aclr0
1
aclr1
4
aclr2
5
aclr3
7
ena[3..0]
Same rows as the DSP block-wide clock enable signals.
Clock Signals
DSP block-wide clear signals.
Input/Output Data Interface Routing
The 30 local interconnects generate the 18 inputs to the row interface
blocks. The 21 outputs of the row interface block are the inputs to the DSP
row block (see Figure 6–7 on page 6–14).
Altera Corporation
July 2005
6–17
Stratix Device Handbook, Volume 2
Operational Modes
The row interface block has DirectLink™ connections that connect the
DSP block input or output signals to the left and right adjacent LABs at
each row. (The DirectLink connections provide interconnects between
LABs and adjacent blocks.) The DirectLink connection reduces the use of
row and column interconnects, providing higher performance and
flexibility.
Each row interface block receives 10 DirectLink connections from the
right adjacent LABs and 10 from the left adjacent LABs. Additionally, the
row interface block receives signals from the DSP block, making a total of
30 local interconnects for each row interface block. All of the row and
column resources within the DSP block can access this interconnect
region (see Figure 6–7 on page 6–14).
A DSP block has nine outputs that drive the right adjacent LAB and nine
that drive the left adjacent LAB through DirectLink interconnects. All
18 outputs drive any row or column.
Operational
Modes
You can use the DSP block in one of four operational modes, depending
on your application needs (see Table 6–4). The Quartus II software has
built-in megafunctions that you can use to control the mode. After you
have made your parameter settings using the megafunction’s
MegaWizard® Plug-In, the Quartus II software automatically configures
the DSP block.
Table 6–5. DSP Block Operational Modes
9× 9
Mode
18 × 18
36 × 36
Simple multiplier
Eight multipliers with eight Four multipliers with four
product outputs
product outputs
One multiplier
Multiply accumulator
Two 34-bit multiplyaccumulate blocks
–
Two-multiplier adder
Four two-multiplier adders Two two-multiplier adders
–
Four-multiplier adder
Two four-multiplier adders One four-multiplier adder
–
Two 52-bit multiplyaccumulate blocks
Simple Multiplier Mode
In simple multiplier mode, the DSP block performs individual
multiplication operations for general-purpose multipliers and for
applications such as equalizer coefficient updates that require many
individual multiplication operations.
6–18
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
DSP Blocks in Stratix & Stratix GX Devices
9- & 18-Bit Multipliers
You can configure each DSP block multiplier for 9 or 18 bits. A single DSP
block can support up to 8 individual 9-bit or smaller multipliers, or up to
4 individual multipliers with operand widths between 10- and 18-bits.
Figure 6–10 shows the simple multiplier mode.
Figure 6–10. Simple Multiplier Mode
signa
Adder Output Block
A
Q
D
A
ENA
CLRN
B
Q
D
ENA
CLRN
Q
D
ENA
CLRN
shiftoutb
shiftouta
signb
The multiplier operands can accept signed integers, unsigned integers, or
a combination. The signa and signb signals are dynamic and can be
registered in the DSP block. Additionally, you can register the multiplier
inputs and results independently. Pipelining the result, using the
pipeline registers in the block, increases the performance of the DSP
block.
36-Bit Multiplier
The 36-bit multiplier is a subset of the simple multiplier mode. It uses the
entire DSP block to implement one 36 × 36-bit multiplier. The four 18-bit
multipliers are fed part of each input, as shown in Figure 6–11 on
page 6–21. The adder/output block adds the partial products using the
Altera Corporation
July 2005
6–19
Stratix Device Handbook, Volume 2
Operational Modes
summation block. You can use pipeline registers between the multiplier
stage and the summation block. The 36 × 36-bit multiplier supports
signed and unsigned operation.
The 36-bit multiplier is useful when your application needs more than
18-bit precision, for example, for mantissa multiplication of precision
floating-point arithmetic applications.
6–20
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
DSP Blocks in Stratix & Stratix GX Devices
Figure 6–11. 36-Bit Multiplier
signa
signb
A[17..0]
D
Q
ENA
CLRN
Q
D
ENA
CLRN
B[17..0]
D
Q
ENA
A
CLRN
A[35..18]
D
Q
B
ENA
CLRN
D
Q
ENA
Partial
Product
Summation
Block
CLRN
B[35..18]
D
D
Q
ENA
Q
Data Out
CLRN
ENA
CLRN
A[35..18]
D
Q
C
ENA
CLRN
D
Q
ENA
CLRN
B[17..0]
D
D
Q
ENA
CLRN
A[17..0]
D
Q
ENA
CLRN
D
Q
ENA
CLRN
B[35..18]
D
Q
ENA
CLRN
Altera Corporation
July 2005
6–21
Stratix Device Handbook, Volume 2
Operational Modes
Multiply Accumulator Mode
In multiply accumulator mode, the output of the multiplier stage feeds
the adder/output block, which is configured as an accumulator or
subtractor (see Figure 6–12). You can implement up to two independent
18-bit multiply accumulators in one DSP block. The Quartus II software
implements smaller multiplier-accumulators by tying the unused loworder bits of an 18-bit multiplier to ground.
Figure 6–12. Multiply Accumulator Mode
signa (1)
signb (1)
aclr
clock
ena
shiftina
shiftinb
D
Data A
Q
D
ENA
D
Q
Q
Data Out
ENA
ENA
CLRN
Accumulator
CLRN
CLRN
D
Data B
Q
D
ENA
Q
overflow
ENA
CLRN
CLRN
shiftoutb
addnsub1
signa
shiftouta
signb
accum_sload1
Note to Figure 6–12:
(1)
The signa and signb signals are the same in the multiplier stage and the adder/output block.
The multiply accumulator output can be up to 52 bits wide for a
maximum 36-bit result with 16-bits of accumulation. In this mode, the
DSP block uses output registers and the accum_sload and overflow
signals. The accum_sload[1..0] signal synchronously loads the
multiplier result to the accumulator output. This signal can be
unregistered or registered once or twice. The DSP block can then begin a
new accumulation without losing any clock cycles. The overflow signal
indicates an overflow or underflow in the accumulator. This signal is
6–22
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
DSP Blocks in Stratix & Stratix GX Devices
cleared for the next accumulation cycle, and you can use an external latch
to preserve the signal. You can use the addnsub[1..0] signals to
perform accumulation or subtraction dynamically.
1
If you want to use DSP blocks and your design only has an
accumulator, you can use a multiply by one followed by an
accumulator to force the software to implement the logic in the
DSP block.
Two-Multiplier Adder Mode
The two-multiplier adder mode uses the adder/output block to add or
subtract the outputs of the multiplier block, which is useful for
applications such as FFT functions and complex FIR filters. Additionally,
in this mode, the DSP block outputs two sums or differences for
multipliers up to 18 bits, or 4 sums or differences for 9-bit or smaller
multipliers. A single DSP block can implement one 18 × 18-bit complex
multiplier or two 9 × 9-bit complex multipliers.
A complex multiplication can be written as:
(a + jb) × (c + jd) = (a × c – b × d) + j × (a × d + b × c)
In this mode, a single DSP block calculates the real part (a × c – b × d) using
one adder/subtractor/accumulator and the imaginary part (a × d + b × c)
using another adder/subtractor/accumulator for data up to 18 bits.
Figure 6–13 shows an 18-bit complex multiplication. For data widths up
to 9 bits, the DSP block can perform two complex multiplications using
four one-level adders. Resources outside of the DSP block route each
input to the two multiplier inputs.
1
Altera Corporation
July 2005
You can only use the adder block if it follows multiplication
operations.
6–23
Stratix Device Handbook, Volume 2
Operational Modes
Figure 6–13. Complex Multiplier Implemented Using Two-Multiplier Adder
Mode
18
DSP Block
18
A
36
18
C
37
A×C-B×D
(Real Part)
Subtractor
18
B
36
18
D
18
A
36
18
D
37
Adder
18
B
A×D+B×C
(Imaginary Part)
36
18
C
Four-Multiplier Adder Mode
In the four-multiplier adder mode, which you can use for 1-dimensional
and 2-dimensional filtering applications, the DSP block adds the results
of two adder/subtractor/accumulators in a final stage (the summation
block).
1
You can only use the adder block if it follows multiplication
operations.
9- & 18-Bit Summation Blocks
A single DSP block can implement one 18 × 18 or two 9 × 9 summation
blocks (see Figure 6–14 on page 6–25). The multiplier product widths
must be the same size.
6–24
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
DSP Blocks in Stratix & Stratix GX Devices
Figure 6–14. Four-Multiplier Adder Mode
signa
signb
aclr
clock
ena
shiftina
shiftinb
D
Data A
Q
ENA
CLRN
D
Q
ENA
Adder/
Subtractor
CLRN
D
Data B
Q
ENA
CLRN
D
Data A
Q
Q
D
ENA
ENA
CLRN
D
Q
ENA
CLRN
D
Data B
Q
addnsub0
signa
signb
addnsub1
Data Out
Adder
CLRN
ENA
CLRN
D
Data A
Q
ENA
CLRN
D
Q
ENA
Adder/
Subtractor
CLRN
D
Data B
Q
ENA
CLRN
D
Data A
Q
ENA
CLRN
Q
D
ENA
CLRN
D
Data B
Q
ENA
CLRN
shiftoutb
Altera Corporation
July 2005
shiftouta
6–25
Stratix Device Handbook, Volume 2
Operational Modes
FIR Filters
The four-multiplier adder mode can be used for FIR filter and complex
FIR filter applications. The DSP block combines a four-multiplier adder
with the input registers configured as shift registers. One set of shift
inputs contains the filter data, while the other holds the coefficients,
which can be loaded serially or in parallel (see Figure 6–15).
The input shift register eliminates the need for shift registers external to
the DSP block (e.g., implemented in device logic elements). This
architecture simplifies filter design and improves performance because
the DSP block implements all of the filter circuitry.
1
Serial shift inputs in 36-bit simple multiplier mode require
external registers.
One DSP block can implement an entire 18-bit FIR filter with up to four
taps. For FIR filters larger than four taps, you can cascade DSP blocks
with additional adder stages implemented in logic elements.
6–26
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
DSP Blocks in Stratix & Stratix GX Devices
Figure 6–15. Input Shift Registers Configured for a FIR Filter
Data A
D
Q
ENA
A[n] × B[n] (to adder)
CLRN
Data B
D
Q
Q
D
ENA
CLRN
ENA
CLRN
Data B
Data A
D
Q
ENA
A[n - 1] × B[n - 1] (to adder)
CLRN
D
Q
D
Q
ENA
CLRN
ENA
CLRN
Data B
Data A
D
Q
ENA
A[n - 2] × B[n - 2] (to adder)
CLRN
D
Q
Q
D
ENA
CLRN
ENA
CLRN
Altera Corporation
July 2005
6–27
Stratix Device Handbook, Volume 2
Software Support
Software
Support
Altera provides two distinct methods for implementing various modes of
the DSP block in your design: instantiation and inference. Both methods
use the following three Quartus II megafunctions:
■
■
■
lpm_mult
altmult_add
altmult_accum
You can instantiate the megafunctions in the Quartus II software to use
the DSP block. Alternatively, with inference, you can create a HDL design
an synthesize it using a third-party synthesis tool like LeonardoSpectrum
or Synplify or Quartus II Native Synthesis that infers the appropriate
megafunction by recognizing multipliers, multiplier adders, and
multiplier accumulators. Using either method, the Quartus II software
maps the functionality to the DSP blocks during compilation.
Conclusion
f
See the Implementing High-Performance DSP Functions in Stratix & Stratix
GX Devices chapter in the Stratix Device Handbook, Volume 2 or the
Stratix GX Device Handbook, Volume 2 for more information on using DSP
blocks to implement high-performance DSP functions such as FIR filters,
IIR filters, and discreet cosine transforms (DCTs).
f
See Quartus II On-Line Help for instructions on using the megafunctions
and the MegaWizard Plug-In Manager.
f
For more information on DSP block inference support, see the
Recommended HDL Coding Styles chapter of the Quartus II Development
Software Handbook v4.1, Volume 1.
The Stratix and Stratix GX device DSP blocks are optimized to support
DSP applications that need high data throughput, such as FIR filters, FFT
functions, and encoders. These DSP blocks are flexible and can be
configured in one of four operational modes to suit any application need.
The DSP block’s adder/subtractor/accumulator and the summation
blocks minimize the amount of logic resources used and provide efficient
routing. This efficiency results in improved performance and data
throughput for DSP applications. The Quartus II software, together with
the LeonardoSpectrum and Synplify software, provides a complete and
easy-to-use flow for implementing functionality in the DSP block.
6–28
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
7. Implementing High
Performance DSP Functions
in Stratix & Stratix GX Devices
S52007-1.1
Introduction
Digital signal processing (DSP) is a rapidly advancing field. With
products increasing in complexity, designers face the challenge of
selecting a solution with both flexibility and high performance that can
meet fast time-to-market requirements. DSP processors offer flexibility,
but they lack real-time performance, while application-specific standard
products (ASSPs) and application-specific integrated circuits (ASICs)
offer performance, but they are inflexible. Only programmable logic
devices (PLDs) offer both flexibility and high performance to meet
advanced design challenges.
The mathematical theory underlying basic DSP building blocks—such as
the finite impulse response (FIR) filter, infinite impulse response (IIR)
filter, fast fourier transform (FFT), and direct cosine transform (DCT)—is
computationally intensive. Altera® Stratix® and Stratix GX devices
feature dedicated DSP blocks optimized for implementing arithmetic
operations, such as multiply, multiply-add, and multiply-accumulate.
In addition to DSP blocks, Stratix and Stratix GX devices have
TriMatrix™ embedded memory blocks that feature various sizes that can
be used for data buffering, which is important for most DSP applications.
These dedicated hardware features make Stratix and Stratix GX devices
an ideal DSP solution.
This chapter describes the implementation of high performance DSP
functions, including filters, transforms, and arithmetic functions, using
Stratix and Stratix GX DSP blocks. The following topics are discussed:
■
■
■
■
■
Stratix &
Stratix GX DSP
Block Overview
Altera Corporation
September 2004
FIR filters
IIR filters
Matrix manipulation
Discrete Cosine Transform
Arithmetic functions
Stratix and Stratix GX devices feature DSP blocks that can efficiently
implement DSP functions, including multiply, multiply-add, and
multiply-accumulate. The DSP blocks also have three built-in registers
sets: the input registers, the pipeline registers at the multiplier output,
and the output registers. Figure 7–1 shows the DSP block operating in the
18 × 18-bit mode.
7–1
Stratix & Stratix GX DSP Block Overview
Figure 7–1. DSP Block Diagram for 18 x 18-bit Mode
Optional Serial Shift Register
Inputs from Previous
DSP Block
Multiplier Stage
D
Optional Stage Configurable
as Accumulator or Dynamic
Adder/Subtractor
Q
ENA
CLRN
D
D
ENA
CLRN
Q
Output Selection
Multiplexer
Q
ENA
CLRN
Adder/
Subtractor/
Accumulator
1
D
Q
ENA
CLRN
D
D
ENA
CLRN
Q
Q
ENA
CLRN
Summation
D
Q
ENA
CLRN
D
D
ENA
CLRN
Q
Q
Summation Stage
for Adding Four
Multipliers Together
Optional Output
Register Stage
ENA
CLRN
Adder/
Subtractor/
Accumulator
2
D
Optional Serial
Shift Register
Outputs to
Next DSP Block
in the Column
Q
ENA
CLRN
D
D
ENA
CLRN
Q
ENA
CLRN
7–2
Stratix Device Handbook, Volume 2
Q
Optional Pipeline
Register Stage
Optional Input Register
Stage with Parallel Input or
Shift Register Configuration
to MultiTrack
Interconnect
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
The DSP blocks are organized into columns enabling efficient horizontal
communication with adjacent TriMatrix memory blocks. Tables 7–1 and
7–2 show the DSP block resources in Stratix and Stratix GX devices,
respectively.
Table 7–1. DSP Block Resources in Stratix Devices
DSP Blocks
Maximum 9 × 9
Multipliers
Maximum 18 × 18
Multipliers
EP1S10
6
48
24
6
EP1S20
10
80
40
10
EP1S25
10
80
40
10
EP1S30
12
96
48
12
EP1S40
14
112
56
14
EP1S60
18
144
72
18
EP1S80
22
176
88
22
Device
Maximum 36 × 36
Multipliers
Table 7–2. DSP Block Resources in Stratix GX Devices
DSP Blocks
Maximum 9 × 9
Multipliers
Maximum 18 × 18
Multipliers
Maximum 36 × 36
Multipliers
EP1SGX10C
6
48
24
6
EP1SGX10D
6
48
24
6
EP1SGX25C
10
80
40
10
EP1SGX25D
10
80
40
10
EP1SGX25F
10
80
40
10
EP1SGX40D
14
112
56
14
EP1SGX40G
14
112
56
14
Device
Altera Corporation
September 2004
7–3
Stratix Device Handbook, Volume 2
TriMatrix Memory Overview
Each DSP block supports either eight 9 × 9-bit multipliers, four 18-bit
multipliers, or one 36 × 36-bit multiplier. These multipliers can feed an
adder or an accumulator unit based on the operation mode. Table 7–3
shows the different operation modes for the DSP blocks.
Table 7–3. Operation Modes for DSP Blocks
Number & Size of Multipliers per DSP Block
DSP Block Mode
9 x 9-bit
18 x 18-bit
36 x 36-bit
Simple multiplier
Eight multipliers with
eight product outputs
Four multipliers with four
product outputs
Multiply-accumulate
Two multiply and
accumulate (34 bit)
Two multiply and
accumulate (52 bit)
Two-multipliers adder
4 two-multipliers adders
2 two-multipliers adders
Four-multipliers adder
2 four-multipliers adder
1 four-multipliers adder
One multiplier with one
product output
Implementing multipliers, multiply-adders, and multiply-accumulators
in the DSP blocks has a performance advantage over logic cell
implementation. Using DSP blocks also reduces logic cell and routing
resource consumption. To achieve higher performance, register each
stage of the DSP block to allow pipelining. For implementing
applications, such as FIR filters, efficiently use the input registers of the
DSP block as shift registers.
f
TriMatrix
Memory
Overview
For more information on DSP blocks, see the DSP Blocks in Stratix &
Stratix GX Devices chapter.
Stratix and Stratix GX devices feature the TriMatrix memory structure,
composed of three sizes of embedded RAM blocks. These include the
512-bit size M512 block, the 4-Kbit size M4K block, and the 512-Kbit size
M-RAM block. Each block is configurable to support a wide range of
features.
Tables 7–4 and 7–5 show the number of memory blocks in each Stratix
and Stratix GX device, respectively.
Table 7–4. TriMatrix Memory Resources in Stratix Devices (Part 1 of 2)
Device
M512
M4K
M-RAM
EP1S10
94
60
1
EP1S20
194
82
2
EP1S25
224
138
2
7–4
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
Table 7–4. TriMatrix Memory Resources in Stratix Devices (Part 2 of 2)
Device
M512
M4K
M-RAM
EP1S30
295
171
4
EP1S40
384
183
4
EP1S60
574
292
6
EP1S80
767
364
9
Table 7–5. TriMatrix Memory Resources in Stratix GX Devices
Device
M512
M4K
M-RAM
EP1SGX10C
94
60
1
EP1SGX10D
94
60
1
EP1SGX25C
224
138
2
EP1SGX25D
224
138
2
EP1SGX25F
224
138
2
EP1SGX40D
384
183
4
EP1SGX40G
384
183
4
Most DSP applications require local data storage for intermediate
buffering or for filter storage. The TriMatrix memory blocks enable
efficient use of available resources for each application.
The M512 and M4K memory blocks can implement shift registers for
applications, such as multi-channel filtering, auto-correlation, and crosscorrelation functions. Implementing shift registers in embedded memory
blocks reduces logic cell and routing resource consumption.
f
For more information on TriMatrix memory blocks, see the TriMatrix
Embedded Memory Blocks in Stratix & Stratix GX Devices chapter.
DSP Function
Overview
The following sections describe commonly used DSP functions. Each
section illustrates the implementation of a basic DSP building block,
including FIR and IIR filters, in Stratix and Stratix GX devices using DSP
blocks and TriMatrix memory blocks.
Finite Impulse
Response (FIR)
Filters
This section describes the basic theory and implementation of basic FIR
filters, time-domain multiplexed (TDM) FIR filters, and interpolation and
decimation polyphase FIR filters. An introduction to the complex FIR
filter is also presented in this section.
Altera Corporation
September 2004
7–5
Stratix Device Handbook, Volume 2
Finite Impulse Response (FIR) Filters
FIR Filter Background
Digital communications systems use FIR filters for a variety of functions,
including waveform shaping, anti-aliasing, band selection,
decimation/interpolation, and low pass filtering. The basic structure of a
FIR filter consists of a series of multiplications followed by an addition.
The following equation represents an FIR filter operation:
y( n) = x( n) ⊗ h(n )
L–1
y( n) =
∑ x ( n – i )h ( i )
i=0
where:
x(n) represents the sequence of input samples
h(n) represents the filter coefficients
L is the number of filter taps
A sample FIR filter with L=8 is shown in Figure 7–2.
Figure 7–2. Basic FIR Filter
x(n)
h(0)
h(1)
h(2)
h(3)
h(5)
h(4)
h(6)
h(7)
y(n)
This example filter in Figure 7–2 uses the input values at eight different
time instants to produce an output. Hence, it is an 8-tap filter. Each
register provides a unit sample delay. The delayed inputs are multiplied
with their respective filter coefficients and added together to produce the
output. The width of the output bus depends on the number of taps and
the bit width of the input and coefficients.
7–6
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
Basic FIR Filter
A basic FIR filter is the simplest FIR filter type. As shown in Figure 7–2, a
basic FIR filter has a single input channel and a single output channel.
Basic FIR Filter Implementation
Stratix and Stratix GX devices’ dedicated DSP blocks can implement basic
FIR filters. Because these DSP blocks have closely integrated multipliers
and adders, filters can be implemented with minimal routing resources
and delays. For implementing FIR filters, the DSP blocks are configured
in the four-multipliers adder mode.
f
See the DSP Blocks in Stratix & Stratix GX Devices chapter for more
information on the different modes of the DSP blocks.
This section describes the implementation of an 18-bit 8-tap FIR filter.
Because Stratix and Stratix GX devices support modularity, cascading
two 4-tap filters can implement an 8-tap filter. Larger FIR filters can be
designed by extending this concept. Users can also increase the number
of taps available per DSP block if 18 bits of resolution are not required. For
example, by using only 9 bits of resolution for input samples and
coefficient values, 8 multipliers are available per DSP block. Therefore, a
9-bit 8-tap filter can be implemented in a single DSP block provided an
external adder is implemented in logic cells.
The four-multipliers adder mode, shown in Figure 7–3, provides four
18 × 18-bit multipliers and three adders in a single DSP block. Hence, it
can implement a 4-tap filter. The data width of the input and the
coefficients is 18 bits, which results in a 38-bit output for a 4-tap filter.
Altera Corporation
September 2004
7–7
Stratix Device Handbook, Volume 2
Finite Impulse Response (FIR) Filters
Figure 7–3. Hardware View of a DSP Block in Four-Multipliers Adder Mode Notes (1). (2), (3)
shiftout
input from
previous
block
shiftout
input from
previous
block
CLK2
CLR2
Data from
row
18
interface
block
Coefficients
18
from row
interface
block
CLK1
CLR1
D
D
Q
Multiplier A
36
D
Q
D
Q
h(0)
Q
Data from row
interface block 18
x(n) 18
18
D
Q
x(n-1) 18
37
36
Coefficients
from row 18
interface
block
Multiplier B
h(1)
D
Q
18
38
38
D
Data from row
interface block 18
D
Q
Output
y(n)
x(n-2) 18
Multiplier C
36
Coefficients
from row 18
interface
block
D
D
Q
D
Q
37
h(2)
Q
Data from row
interface block 18
Q
18
D
Q
x(n-3) 18
36
Coefficients
from row 18
interface
block
D
Q
shiftin
input to
next block
Multiplier D
h(3)
18
shiftin
input to
next block
Notes to Figure 7–3:
(1)
(2)
(3)
The input registers feed the multiplier blocks. These registers can increase the DSP block performance, but are
optional. These registers can also function as shift registers if the dedicated shiftin/shiftout signals are used.
The pipeline registers are fed by the multiplier blocks. These registers can increase the DSP block performance, but
are optional.
The output registers register the DSP block output. These registers can increase the DSP block performance, but are
optional.
7–8
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
Figure 7–4. Quartus II Software View of MegaWizard Implementation of a DSP Block in Four-Multipliers
Adder Mode
Each input register of the DSP block provides a shiftout output that
connects to the shiftin input of the adjacent input register of the same DSP
block. The registers on the boundaries of a DSP block also connect to the
registers of adjacent DSP blocks through the use of shiftin/shiftout
connections. These connections create register chains spanning multiple
DSP blocks, which makes it easy to increase the length of FIR filters.
Figure 7–5 shows two DSP blocks connected to create an 8-tap FIR filter.
Filters with more taps can be implemented by connecting DSP blocks in
a similar manner, provided sufficient DSP blocks are available in the
device.
1
Adding the outputs of the two DSP blocks requires an external
adder which can be implemented using logic cells.
The input data can be fed directly or by using the shiftout/shiftin chains,
which allow a single input to shift down the register chain inside the DSP
block. The input to each of the registers has a multiplexer, hence, the data
can be fed either from outside the DSP block or the preceding register.
This can be selected from the MegaWizard® in the Quartus® II software,
as shown in Figure 7–4. The example in Figure 7–5 uses the
shiftout/shiftin flip-flop chains where the multiplexers are configured to
use these chains. In this example, the flip-flops inside the DSP blocks
serve as the taps of the FIR filter.
Altera Corporation
September 2004
7–9
Stratix Device Handbook, Volume 2
Finite Impulse Response (FIR) Filters
When the coefficients are loaded in parallel, they can be fed directly from
memory elements or any other muxing scheme. This facilitates the
implementation of an adaptive (variable) filter.
Further, if the user wants to implement the shift register chains external
to the DSP block, this can be done by using the altshift_taps
megafunction. In this case, the coefficient and data shifting is done
external to the DSP block. The DSP block is only used to implement the
multiplications and the additions.
Parallel vs. Serial Implementation
The fastest implementations are fully parallel, but consume more logic
resources than serial implementations. To trade-off performance for logic
resources, implement a serial scheme with a specified number of taps. To
facilitate this, Altera provides the FIR Compiler core through its
MegaCore program. The FIR Compiler is an easy-to-use, fully-integrated
graphical user interface (GUI) based FIR filter design software.
f
For more information on the FIR Compiler MegaCore, visit the Altera
web site at www.altera.com and search for “FIR compiler” in the
“Intellectual Property” page.
It is important to note that the four-multipliers adder mode allows a DSP
block to be configured for parallel or serial input. When it is configured
for parallel input, as shown in Figure 7–6, the data input and the
coefficients can be loaded directly without the need for shiftin/shiftout
chains between adjacent registers in the DSP block. When the DSP block
is configured for serial input, as shown in Figure 7–5, the shiftin/shiftout
chains create a register cascade both within the DSP block and also
between adjacent DSP blocks.
7–10
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
Figure 7–5. Serial Loading 18-Bit 8-Tap FIR Filter Using Two DSP Blocks
Notes (1), (2), (3)
Data input x(n)
D
Q
D
Q
D
Q
D
Q
D
Q
D
Q
D
Q
D
Q
DSP block 1
Filter coefficients h(0)
x
x(n-1)
h(1)
x
x(n-2)
h(2)
x
x(n-3)
h(3)
Filter output
y(n)
x
x(n-4)
D
Q
D
Q
D
Q
D
Q
D
Q
D
Q
D
Q
D
Q
DSP block 2
h(4)
x
x(n-5)
h(5)
x
x(n-6)
h(6)
x
x(n-7)
h(7)
Notes to Figure 7–5:
(1)
(2)
(3)
Altera Corporation
September 2004
Unused ports grayed out.
The indexing x(n-1), ..., x(n-7) refers to the case of parallel loading and should be
ignored here. This indexing is retained in this figure for consistency with other
figures in this chapter.
To increase the DSP block performance, include the pipeline and output registers.
See Figure 7–3 on page 7–8 for the details.
7–11
Stratix Device Handbook, Volume 2
Finite Impulse Response (FIR) Filters
Figure 7–6. Parallel Loading 18-Bit 8-Tap FIR Filter Using Two DSP Blocks
Notes (1), (2)
Data input x(n)
D
Q
D
Q
D
Q
D
Q
D
Q
D
Q
D
Q
D
Q
DSP block 1
Filter coefficients h(0)
x(n-1)
h(1)
x(n-2)
h(2)
x(n-3)
h(3)
Filter output
y(n)
x(n-4)
D
Q
D
Q
D
Q
D
Q
D
Q
D
Q
D
Q
D
Q
DSP block 2
h(4)
x(n-5)
h(5)
x(n-6)
h(6)
x(n-7)
h(7)
Notes to Figure 7–6:
(1)
(2)
The indexing x(n-1), ..., x(n-7) refers to the case of parallel loading.
To increase the DSP block performance, include the input, pipeline, and output
registers. See Figure 7–3 on page 7–8 for the details.
7–12
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
Basic FIR Filter Implementation Results
Table 7–6 shows the results of the serial implementation of an 18-bit 8 tap
FIR filter as shown in Figure 7–5 on page 7–11
Table 7–6. Basic FIR Filter Implementation Results
Part
EP1S10F780
Utilization
LCELL: 130/10570 (1%)
DSP Block 9-bit elements: 16/48 (33%)
Memory bits: 288/920448 (<1%)
Performance
247 MHz
Basic FIR Filter Design Example
Download the Basic FIR Filter (base_fir.zip) design example from the
Design Examples section of the Altera web site at www.altera.com.
Time-Domain Multiplexed FIR Filters
A TDM FIR filter is clocked n-times faster than the sample rate in order to
reuse the same hardware. Consider the 8-tap filter shown in Figure 7–2.
The TDM technique can be used with a TDM factor of 2, i.e., n = 2, to
implement this filter using only four multipliers, provided the filter is
clocked two times faster internally.
To understand this concept, consider Figure 7–7 that shows a TDM filter
with a TDM factor of 2. A 2× -multiplied clock is required to run the filter.
On cycle 0 of the 2× clock, the user loads four coefficients into the four
multiplier inputs. The resulting output is stored in a register. On cycle 1
of the 2× clock, the user loads the remaining four coefficients into the
multiplier inputs. The output of cycle 1 is added with the output of cycle
0 to create the overall output. See the “TDM Filter Implementation” on
page 7–14 section for details on the coefficient loading schedule.
The TDM implementation shown in Figure 7–7 requires only four
multipliers to achieve the functionality of an 8-tap filter. Thus, TDM is a
good way to save logic resources, provided the multipliers can run at ntimes the clock speed. The coefficients can be stored in ROM/RAM, or
any other muxing scheme.
Altera Corporation
September 2004
7–13
Stratix Device Handbook, Volume 2
Finite Impulse Response (FIR) Filters
Figure 7–7. Block Diagram of 8-Tap FIR Filter with TDM Factor of n=2
Output
FIR filter with
four multipliers
18-bit input
D
Q
2x clock
TDM Filter Implementation
TDM FIR filters are implemented in Stratix and Stratix GX devices by
configuring the DSP blocks in the multiplier-adder mode. Figure 7–9
shows the implementation of an 8-tap TDM FIR filter (n=2) with 18 bits
of data and coefficient inputs. Because the input data needs to be loaded
into the DSP block in parallel, a shift register chain is implemented using
a combination of logic cells and the altshift_taps function. This shift
register is clocked with the same data sample rate (clock 1× ). The filter
coefficients are stored in ROM and loaded into the DSP block in parallel
as well. Because the TDM factor is 2, both the ROM and DSP block are
clocked with clock 2× .
Table 7–7. Operation of TDM Filter (Shown in Figure 7–9 on page 7–16)
Cycle of
2× Clock
Cycle Output
0
y0 = x(n-1)h(1) + x(n-3)h(3) + x(n-5)h(5) + x(n-7)h(7)
Store result
N/A
1
y1 = x(n)h(0)
+ x(n-2)h(2) + x(n-4)h(4) + x(n-6)h(6)
Generate output
y(n) = y0 + y1
2
y2 = x(n)h(1)
+ x(n-2)h(3) + x(n-4)h(5) + x(n-6)h(7)
Store result
N/A
3
y3 = x(n+1)h(0) + x(n-1)h(2) + x(n-3)h(4) + x(n-5)h(6)
Generate output
y(n) = y2 + y3
4
y4 = x(n+1)h(1) + x(n-1)h(3) + x(n-3)h(5) + x(n-5)h(7)
Store result
N/A
5
y5 = x(n+2)h(0) + x(n)h(2)
Generate output
y(n) = y4 + y5
6
y6 = x(n+2)h(1) + x(n)h(3) + x(n-2)h(5) + x(n-4)h(7)
Store result
N/A
7
y7 = x(n+3)h(0) + x(n+1)h(2) + x(n-1)h(4) + x(n-3)h(6) Generate output
+ x(n-2)h(4) + x(n-4)h(6)
Operation
Overall Output,
y(n)
y(n) = y6 + y7
Figure 7–8 and Table 7–7 show the coefficient loading schedule. For
example, during cycle 0, only the flip-flops corresponding to h(1), h(3),
h(5), and h(7) are enabled. This produces the temporary output, y0, which
is stored in a flip-flop outside the DSP block. During cycle 1, only the flip-
7–14
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
flops corresponding to h(0), h(2), h(4) and h(6) are enabled. This produces
the temporary output, y1, which is added to y0 to produce the overall
output, y(n). The following shows what the overall output, y(n), equals:
y ( n ) = y0 + y1
y ( n ) = x ( 0 )h ( 0 ) + x ( n – 1 )h ( 1 ) + x ( n – 2 )h ( 2 ) + x ( n – 3 )h ( 3 )
+ x ( n – 4 )h ( 4 ) + x ( n – 5 )h ( 5 ) + x ( n – 6 )h ( 6 ) + x ( n – 7 )h ( 7 )
This is identical to the output of the 8-tap filter shown in Figure 7–2. After
cycle 1, this process is repeated at every cycle.
Figure 7–8. Coefficient Loading Schedule in a TDM Filter
2x clock
1x clock
Cycle 0
Cycle 1
Cycle 2
Cycle 3
load h(1), h(3), h(5), h(7) load h(0), h(2), h(4), h(6) load h(1), h(3), h(5), h(7) load h(0), h(2), h(4), h(6)
Altera Corporation
September 2004
Cycle 4
load h(1), h(3), h(5), h(7)
7–15
Stratix Device Handbook, Volume 2
Finite Impulse Response (FIR) Filters
Figure 7–9. TDM FIR Filter Implementation Note (1)
Data input
x
x(n)
Shift register
DSP block
D
Q
D
Q
D
Q
D
Q
D
Q
Filter coefficients
D
Q
D
Q
D
Q
RAM / ROM 0
RAM / ROM 1
Accumulator
D
D
Q
D
Q
D
Q
RAM / ROM 2
RAM / ROM 3
D
Q
D
Q
D
Q
D
Q
Q
Filter output
y(n)
1x clock
Clock input
(1x clock)
PLL
2x clock
Note to Figure 7–9:
(1)
To increase the DSP block performance, include the pipeline and output registers. See Figure 7–3 on page 7–8 for
details.
If the TDM factor is more than 2, then a multiply-accumulator needs to be
implemented. This multiply-accumulator can be implemented using the
soft logic outside the DSP block if all the multipliers of the DSP block are
needed. Alternatively, the multiply-accumulator may be implemented
inside the DSP block if all the multipliers of the DSP block are not needed.
The accumulator needs to be zeroed at the start of each new sample input.
The user also needs a way to store additional sample inputs in memory.
For example, consider a sample rate of r and TDM factor of 4. Then, the
7–16
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
user needs a way to accept this sample data and send it at a 4r rate to the
input of the DSP block. One way to do this is using a first-in-first-out
(FIFO) memory with input clocked at rate r and output clocked at rate 4r.
The FIFO may be implemented in the TriMatrix memory.
TDM Filter Implementation Results
Table 7–8 shows the results of the implementation of an 18-bit 8-tap TDM
FIR filter as shown in Figure 7–9 on page 7–16.
Table 7–8. TDM Filter Implementation Results
Part
EP1S10F780
Utilization
Lcell: 196/10570 (1%)
DSP Block 9-bit elements: 8/48 (17%)
Memory bits: 360/920448 (<1%)
Performance
240 MHz (1)
Note to Table 7–8:
(1)
This refers to the performance of the DSP blocks. The input and output rate is 120
million samples per second (MSPS), clocked in and out at 120 MHz.
TDM Filter Design Example
Download the TDM FIR Filter (tdm_fir.zip) design example from the
Design Examples section of the Altera web site at www.altera.com.
Polyphase FIR Interpolation Filters
An interpolation filter can be used to increase sample rate. An
interpolation filter is efficiently implemented with a polyphase FIR filter.
DSP systems frequently use polyphase filters because they simplify
overall system design and also reduce the number of computations per
cycle required of the hardware. This section first describes interpolation
filters and then how to implement them as polyphase filters in Stratix and
Stratix GX devices. See the “Polyphase FIR Decimation Filters” on
page 7–24 section for a discussion of decimation filters.
Interpolation Filter Basics
An interpolation filter increases the output sample rate by a factor of I
through the insertion if I-1 zeros between input samples, a process
known as zero padding. After the zero padding, the output samples in
time domain are separated by Ts/I = 1/(I× fs), where Ts and fs are the
sample period and sample frequency of the original signal, respectively.
Figure 7–10 shows the concept of signal interpolation.
Altera Corporation
September 2004
7–17
Stratix Device Handbook, Volume 2
Finite Impulse Response (FIR) Filters
Inserting zeros between the samples creates reflections of the original
spectrum, thus, a low pass filter is needed to filter out the reflections.
Figure 7–10. Block Diagram Representation of Interpolation
Input
sample rate fs
I
LPF
Output
sample rate I*fs
To see how interpolation filters work, consider the Nyquist Sampling
Theorem. This theorem states that the maximum frequency of the input
to be sampled must be smaller than fs/2, where fs is the sampling
frequency, to avoid aliasing. This frequency, fs/2, is also known as the
Nyquist frequency (Fn). Typically, before a signal is sampled using an
analog to digital converter (ADC), it needs to be low pass filtered using
an analog anti-aliasing filter to prevent aliasing. If the input frequency
spectrum extends close to the Nyquist frequency, then the first alias is also
close to the Nyquist frequency. Therefore, the low pass filter needs to be
very sharp to reject this alias. A very sharp analog filter is hard to design
and manufacture and could increase passband ripple, thereby
compromising system performance.
The solution is to increase the sampling rate of the ADC, so that the new
Nyquist frequency is higher and the spacing between the desired signal
and the alias is also higher. Zero padding as described above increase the
sample rate. This process also known as upsampling (oversampling)
relaxes the roll off requirements of the anti-aliasing filter. Consequently, a
simpler filter achieves alias suppression. A simpler analog filter is easier
to implement, does not compromise system performance, and is also
easier to manufacture.
Similarly, the digital to analog converter (DAC) typically interpolates the
data before the digital to analog conversion. This relaxes the requirement
on the analog low pass filter at the output of the DAC.
The interpolation filter does not need to run at the oversampled
(upsampled) rate of fs× I. This is because the extra sample points added
are zeros, so they do not contribute to the output.
Figure 7–11 shows the time and frequency domain representation of
interpolation for a specific case where the original signal spectrum is
limited to 2 MHz and the interpolation factor (I) is 4. The Nyquist
frequency of the upsampled signal must be greater than 8 MHz, and is
chosen to be 9 MHz for this example.
7–18
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
Figure 7–11. Time & Frequency Domain Representations of Interpolation for I = 4
As an example, CD players use interpolation, where the nominal sample
rate of audio input is 44.1 kilosamples per second. A typical
implementation might have an interpolation (oversampling) factor of 4
generating 176.4 kilosamples per second of oversampled data stream.
Polyphase Interpolation Filters
A direct implementation of an interpolation filter, as shown in
Figure 7–10, imposes a high computational burden. For example, if the
filter is 16 taps long and a multiplication takes one cycle, then the number
of computations required per cycle is 16× I. Depending on the
interpolation factor (I), this number can be quite big and may not be
achievable in hardware. A polyphase implementation of the low pass
filter can reduce the number of computations required per cycle, often by
a large factor, as will be evident later in this section.
The polyphase implementation “splits” the original filter into I
polyphase filters whose impulse responses are defined by the following
equation:
h k ( n ) = h ( k + nI )
Altera Corporation
September 2004
7–19
Stratix Device Handbook, Volume 2
Finite Impulse Response (FIR) Filters
where:
k = 0,1, …, I-1
n = 0,1, …, P-1
P = L/I = length of polyphase filters
L = length of the filter (selected to be a multiple of I)
I = interpolation factor
h(n) = original filter impulse response
This equation implies that the first polyphase filter, h0(n), has coefficients
h(0), h(I), h(2I),..., h((P-1)I). The second polyphase filter, h1(n), has
coefficients h(1), h(1+I), h(1+2I), ..., h(1+(P-1)I). Continuing in this way,
the last polyphase filter, hI-1(n), has coefficients h(I-1), h((I - 1) + N), h((I
- 1) + 2I), ..., h((I - 1) + (P-1)I).
An example helps in understanding the polyphase implementation of
interpolation. Consider the polyphase representation of a 16-tap low pass
filter with an interpolation factor of 4. Thus, the output is given below:
15
y( n) =
∑ h ( n – iI )x ( i )
i=0
Referring back to Figure 7–11 on page 7–19, the only nonzero samples of
the input are x(0), x(4), x(8,) and x(12). The first output, y(0), only depends
on h(0), h(4), h(8) and h(12) because x(i) is zero for i ≠ 0, 4, 8, 12. Table 7–9
shows the coefficients required to generate output samples.
Table 7–9. Decomposition of a 16-Tap Interpolating Filter into Four Polyphase Filters
Output Sample
Coefficients Required
Polyphase Filter Impulse Response
y(0), y(4)...
h(0), h(4), h(8), h(12)
h0(n)
y(1), y(5)...
h(1), h(5), h(9), h(13)
h1(n)
y(2), y(6)...
h(2), h(6), h(10), h(14)
h2(n)
y(3), y(7)...
h(3), h(7), h(11), h(15)
h3(n)
Table 7–9 shows that this filter operation can be represented by four
parallel polyphase filters. This is shown in Figure 7–12. The outputs from
the filters are multiplexed to generate the overall output. The multiplexer
is controlled by a counter, which counts up modulo-I starting at 0.
It is illuminating to compare the computational requirements of the direct
implementation versus polyphase implementation of the low pass filter.
In the direct implementation, the number of computations per cycle
7–20
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
required is 16 × I = 16 × 4 = 64. In the polyphase implementation, the
number of computations per cycle required is 4 × 4 = 16. This is because
there are four polyphase filters, each with four taps.
Figure 7–12. Polyphase Representation of I=4 Interpolation Filter
Interpolation Using a Single Low-Pass Filter
Input
x(n)
I=4
Output
y(n)
LPF with
coefficients
h(0), h(1), ... h(15)
Interpolation Using a Polyphase Filter
Polyphase filter
with coefficients
h(0), h(4), h(8), h(12)
Polyphase filter
with coefficients
h(1), h(5), h(9), h(13)
Input
x(n)
y1(n)
0
1
2
3
Output
y(n)
y2(n)
Polyphase filter
with coefficients
h(2), h(6), h(10), h(14) y3(n)
Polyphase filter
with coefficients
h(3), h(7), h(11), h(15) y4(n)
4x Clock
Modulo 4 up
counter
initialized at 0
Polyphase Interpolation Filter Implementation
Figure 7–13 shows the Stratix or Stratix GX implementation of the
polyphase interpolation filter in Figure 7–12. The four polyphase filters
share the same hardware, which is a 4-tap filter. One Stratix or Stratix GX
DSP block can implement one 4-tap filter with 18-bit wide data and
coefficients. A multiplexer can be used to load new coefficient values on
every cycle of the 4× clock. Stratix and Stratix GX phase lock loops (PLLs)
can generate the 4× clock. In the first cycle of the 4× clock, the user needs
to load coefficients for polyphase filter h0(n); in the second cycle of the 4×
Altera Corporation
September 2004
7–21
Stratix Device Handbook, Volume 2
Finite Impulse Response (FIR) Filters
clock, the users needs to load coefficients of the polyphase filter h1(n) and
so on. Table 7–10 summarizes the coefficient loading schedule. The
output, y(n), is clocked using the 4× clock.
Table 7–10. Polyphase Interpolation (I=4) Filter Coefficient Loading
Schedule
Cycle of 4× Clock
Coefficients to Load
1, 5,...
h(0), h(4), h(8), h(12)
0, 1, 2, 3
2, 6,...
h(1), h(5), h(9), h(13)
0, 1, 2, 3
3, 7,...
h(2), h(6), h(10), h(14)
0, 1, 2, 3
4, 8,...
h(3), h(7), h(11), h(15)
0, 1, 2, 3
7–22
Stratix Device Handbook, Volume 2
Corresponding RAM/ROM
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
Figure 7–13. Implementation of the Polyphase Interpolation Filter (I=4) Notes (1), (2), (3)
Data input
x
x(n)
DSP block
D
Q
D
Q
D
Q
D
Q
Filter coefficients
h(0)
RAM / ROM 0
h(1)
RAM / ROM 1
Filter output
y(n)
D
Q
D
Q
D
Q
D
Q
h(2)
RAM / ROM 2
h(3)
RAM / ROM 3
Note (1)
Clock input
(1x clock)
Note (2)
1x clock
PLL
4x clock
Notes to Figure 7–13:
(1)
(2)
(3)
The 1× clock feeds the input data shiftin register chain.
The 4× clock feeds the input registers for the filter coefficients and other optional registers in the DSP block. See
Note (3).
To increase the DSP block performance, include the pipeline, and output registers. See Figure 7–3 for the details.
Altera Corporation
September 2004
7–23
Stratix Device Handbook, Volume 2
Finite Impulse Response (FIR) Filters
Polyphase Interpolation Filter Implementation Results
Table 7–11 shows the results of the polyphase interpolation filter
implementation in a Stratix device shown in Figure 7–13.
Table 7–11. Polyphase Interpolation Filter Implementation Results
Part
EP1S10F780
Utilization
Lcell: 3/10570 (<1%)
DSP Block 9-bit elements: 8/48 (17%)
Memory bits: 288/920448 (<1%)
Performance
240 MHz (1)
Note to Table 7–11:
(1)
This refers to the performance of the DSP blocks, as well as the output clock rate.
The input rate is 60 MSPS, clocked in at 60MHz.
Polyphase Interpolation Filter Design Example
Download the Interpolation FIR Filter (interpolation_fir.zip) design
example from the Design Examples section of the Altera web site at
www.altera.com.
Polyphase FIR Decimation Filters
A decimation filter can be used to decrease the sample rate. A decimation
filter is efficiently implemented with a polyphase FIR filter. DSP systems
frequently use polyphase filters because they simplify overall system
design and also reduce the number of computations per cycle required of
the hardware. This section first describes decimation filters and then how
to implement them as polyphase filters in Stratix devices. See the
“Polyphase FIR Interpolation Filters” section for a discussion of
interpolation filters.
Decimation Filter Basics
A decimation filter decreases the output sample rate by a factor of D
through keeping only every D-th input sample. Consequently, the
samples at the output of the decimation filter are separated by D× Ts=
D/fs, where Ts and fs are the sample period and sample frequency of the
original signal, respectively. Figure 7–14 shows the concept of signal
decimation.
The signal needs to be low pass filtered before downsampling can begin
in order to avoid the reflections of the original spectrum from being
aliased back into the output signal.
7–24
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
Figure 7–14. Block Diagram Representation of Decimation
Input
sample rate fs
LPF
D
Output
sample rate fs/D
Decimation filters reverse the effect of the interpolation filters. Before the
decimation process, a low pass filter is applied to the signal to attenuate
noise and aliases present beyond the Nyquist frequency. The filtered
signal is then applied to the decimation filter, which processes every D-th
input. Therefore the values between samples D, D-1, D-2 etc. are ignored.
This allows the filter to run M times slower than the input data rate.
In a typical system, after the analog to digital conversion is complete, the
data needs to be filtered to remove aliases inherent in the sampled data.
Further, at this point there is no need to continue to process this data at
the higher sample (oversampled) rate. Therefore, a decimation FIR filter
at the output of the ADC lowers the data rate to a value that can be
processed digitally.
Figure 7–15 shows a specific example where a signal spread over 8 MHz
is decimated by a factor of 4 to 2 MHz. The Nyquist frequency of the
downsampled signal must be greater than 2 MHz, and is chosen to be
2.25 MHz in this example.
Altera Corporation
September 2004
7–25
Stratix Device Handbook, Volume 2
Finite Impulse Response (FIR) Filters
Figure 7–15. Time & Frequency Domain Representations of Decimation for D=4
Polyphase Decimation Filters
Figure 7–14 shows a direct implementation of a decimation filter, which
imposes a high computational burden. For example, if the filter is 16 taps
long and a multiplication takes one cycle, the number of computations
required per cycle is 16× D. Depending on the decimation factor (D), this
number can be quite big and may not be achievable in hardware. A
polyphase implementation of the low pass filter can reduce the number
of computations required, often by a large ratio, as will be evident later in
this section.
The polyphase implementation “splits” the original filter into D
polyphase filters with impulse responses defined by the following
equation.
h k ( n ) = h ( k + nD )
7–26
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
where:
k = 0,1, …, D-1
n = 0,1, …, P-1
P = L/D = length of polyphase filters
L is the length of the filter (selected to be a multiple of D)
D is the decimation factor
h(n) is the original filter impulse response
This equation implies that the first polyphase filter, h0(n), has coefficients
h(0), h(D), h(2D)…h((P-1)D). The second polyphase filter, h1(n), has
coefficients h(1), h(1+D), h(1+2D), ..., h(1+(P-1)D). Continuing in this way,
the last polyphase filter, hD-1(n) has coefficients h(D-1), h((D - 1) + D),
h((D - 1) + 2D), ..., h((D - 1) + (P-1)D).
An example helps in the understanding of the polyphase implementation
of decimation. Consider the polyphase representation of a 16-tap low
pass filter with a decimation factor of 4. The output is given by:
15
y( n) =
∑ h ( i )x ( nD – i )
i=0
Referring to Figure 7–15 on page 7–26, it is clear that the output, y(n) is
discarded for n ≠ 0, 4, 8, 12, hence the only values of y(n) that need to be
computed are y(0), y(4), y(8), y(12). Table 7–12 shows which coefficients
are required to generate the output samples.
Table 7–12. Decomposition of a 16-Tap Decimation Filter into Four Polyphase Filters
Output Sample (1)
Coefficients Required
Polyphase Filter Impulse Response
y(0)0, y(4)0, . . .
h(0), h(4), h(8), h(12)
h0(n)
y(0)1, y(4)1, . . .
h(1), h(5), h(9), h(13)
h1(n)
y(0)2, y(4)2, . . .
h(2), h(6), h(10), h(14)
h2(n)
y(0)3, y(4)3, . . .
h(3), h(7), h(11), h(15)
h3(n)
Note to Table 7–12:
(1)
The output sample is the sum of the results from four polyphase filters: y(n) = y(n)0 + y(n)1 + y(n)2 + y(n)3.
Table 7–12 shows that the overall decimation filter operation can be
represented by 4 parallel polyphase filters. Figure 7–16 shows the
polyphase representation of the decimation filter. A demultiplexer at the
input ensures that the input is applied to only one polyphase filter at a
Altera Corporation
September 2004
7–27
Stratix Device Handbook, Volume 2
Finite Impulse Response (FIR) Filters
time. The demultiplexer is controlled by a counter, which counts down
modulo-D starting at 0. The overall output is taken by adding the outputs
of all the filters.
The polyphase representation of the decimation filter also reduces the
computational requirement. For the example in Figure 7–16, the direct
implementation requires 16 × D = 16 × 4 = 64 computations per cycle,
whereas the polyphase implementation requires only 4 × 4 × 1 = 16
computations per cycle. This saving in computational complexity is quite
significant and is often a very convincing reason to use polyphase filters.
Figure 7–16. Polyphase Filter Representation of a D=4 Decimation Filter
Decimation Using a Single Low-Pass Filter
Input
x(n)
LPF with
coefficients
h(0), h(1), ... h(15)
D=4
Output
y(n)
Decimation Using a Polyphase Filter
Input
x(n)
0
1
2
3
Polyphase filter
with coefficients
h(0), h(4), h(8), h(12)
Output
y(n)
Polyphase filter
with coefficients
h(1), h(5), h(9), h(13)
Polyphase Filter
with coefficients
h(2), h(6), h(10), h(14)
Polyphase Filter
with coefficients
h(3), h(7), h(11), h(15)
4x clock
7–28
Stratix Device Handbook, Volume 2
Modulo 4 down
counter
initialized at 0
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
Polyphase Decimation Filter Implementation
Figure 7–17 shows the decimation polyphase filter example of
Figure 7–16 as it would fit into Stratix or Stratix GX DSP blocks. The
coefficients of the polyphase filters need to be cycled using the schedule
shown in Table 7–13. The output y(n), is clocked using the 1× clock.
Table 7–13. Coefficient Loading Schedule for Polyphase Decimation Filter
(D=4)
Cycle of 4× Clock
Altera Corporation
September 2004
Coefficients to Load
Corresponding RAM/ROM
1, 5,...
h(0), h(4), h(8), h(12)
0, 1, 2, 3
2, 6,...
h(3), h(7), h(11), h(15)
0, 1, 2, 3
3, 7,...
h(2), h(6), h(10), h(14)
0, 1, 2, 3
4, 8,...
h(1), h(5), h(9), h(13)
0, 1, 2, 3
7–29
Stratix Device Handbook, Volume 2
Finite Impulse Response (FIR) Filters
Figure 7–17. Implementation of the Polyphase Decimation Filter (D=4) Notes (1), (2), (3)
Data input
x
x(n)
D
DSP block
Q
D
D
D
D
Q
Q
ROM
D
Q
Filter output
y(n)
Q
D
D
Q
D
Q
D
Q
D
Q
D
Q
D
Q
Q
Q
ROM
Q
Note (2)
Clock input
(1x clock)
Q
Q
D
D
Q
Q
ROM
D
D
Q
D
D
Q
Filter
coefficients
ROM
Q
D
D
Q
D
Note (1)
4x clock
PLL
1x clock
Notes to Figure 7–17:
(1)
(2)
(3)
The 1× clock feeds the register after the accumulator block.
The 4× clock feeds the shift register for the data, the input registers for both the data and filter coefficients, the other
optional registers in the DSP block (see Note (3)), and the accumulator block.
To increase the DSP block performance, include the pipeline, and output registers. See Figure 7–3 on page 7–8 for
the details.
7–30
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
Polyphase Decimation Filter Implementation Results
Table 7–14 shows the results of the polyphase decimation filter
implementation in a Stratix device shown in Figure 7–17.
Table 7–14. Polyphase Decimation Filter Implementation Results
Part
EP1S10F780
Utilization
Lcell: 168/10570 (1%)
DSP Block 9-bit elements: 8/48 (17%)
Memory bits: 300/920448(<1%)
Performance
240 MHz (1)
Note to Table 7–14:
(1)
This refers to the performance of the DSP blocks, as well as the input clock rate.
The output rate is 60 MSPS (clocked out at 60MHz).
Polyphase Decimation Filter Design Example
Download the Decimation FIR Filter (decimation_fir.zip) design
example from the Design Examples section of the Altera web site at
www.altera.com.
Complex FIR Filter
A complex FIR filter takes real and imaginary input signals and performs
the filtering operation with real and imaginary filter coefficients. The
output also consists of real and imaginary signals. Therefore, a complex
FIR filter is similar to a regular FIR filter except for the fact that the input,
output, and coefficients are all complex numbers.
One example application of complex FIR filters is equalization. Consider
a Phase Shift Keying (PSK) system; a single complex channel can
represent the I and Q data channels. A FIR filter with complex coefficients
could then process both data channels simultaneously. The filter
coefficients are chosen in order to reverse the effects of intersymbol
interference (ISI) inherent in practical communication channels. This
operation is called equalization. Often, the filter is adaptive, i.e. the filter
coefficients can be varied as desired, to optimize performance with
varying channel characteristics.
A complex variable FIR filter is a cascade of complex multiplications
followed by a complex addition. Figure 7–18 shows a block diagram
representation of a complex FIR filter. To compute the overall output of
the FIR filter, it is first necessary to determine the output of each complex
multiplier. This can be expressed as:
Altera Corporation
September 2004
7–31
Stratix Device Handbook, Volume 2
Finite Impulse Response (FIR) Filters
y real = x real × h real – x imag × h imag
y imag = x real × h imag + h real × x imag
where:
xreal is the real input signal
ximag is the imaginary input signal
hreal is the real filter coefficients
himag is the imaginary filter coefficients
yreal is the real output signal
yimag is the imaginary output signal
In complex representation, this equals:
y real + jy imag = ( x real + jx imag ) × ( h real + jh imag )
The overall real channel output is obtained by adding the real channel
outputs of all the multipliers. Similarly, the overall imaginary channel
output is obtained by adding the imaginary channel outputs of all the
multipliers.
Figure 7–18. Complex FIR Filter Block Diagram
xreal
ximag
Complex
FIR filter
hreal
yreal
yimag
himag
Complex FIR Filter Implementation
Complex filters can be easily implemented in Stratix devices with the DSP
blocks configured in the two-multipliers adder mode. One DSP block can
implement a 2-tap complex FIR filter with 9-bit inputs, or a single tap
complex FIR filter with 18-bit inputs. DSP blocks can be cascaded to
implement complex filters with more taps.
1
7–32
Stratix Device Handbook, Volume 2
The two-multipliers adder mode has two adders, each adding
the outputs of two multipliers. One of the adders is configured
as a subtractor.
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
f
For more information on the different modes of the DSP blocks, see the
DSP Blocks in Stratix & Stratix GX Devices chapter.
Figure 7–19 shows an example of a 2-tap complex FIR filter design with
18-bit inputs. The real and the complex outputs of the DSP blocks are
added externally to generate the overall real and imaginary output. As in
the case of basic, TDM, or polyphase FIR filters, the coefficients may be
loaded in series or parallel.
Figure 7–19. 2-Tap 18-Bit Complex FIR Filter Implementation
DSP block
Configured as a subtractor
xreal1
outreal1 = xreal1 * hreal1 - ximag1 * himag1
ximag1
Configured as a adder
Overall real output
hreal1
outimag1 = xreal1 * himag1 + ximag1 * hreal1
himag1
DSP block
Configured as a subtractor
xreal2
outreal2 = xreal2 * hreal2 - ximag2 * himag2
ximag2
Overall imaginary output
Configured as a adder
hreal2
outimag2 = xreal2 * himag2 + ximag2 * hreal2
himag2
Altera Corporation
September 2004
7–33
Stratix Device Handbook, Volume 2
Infinite Impulse Response (IIR) Filters
Infinite Impulse
Response (IIR)
Filters
Another class of digital filters are IIR filters. These are recursive filters
where the current output is dependent on previous outputs. In order to
maintain stability in an IIR filter, careful design consideration must be
given, especially to the effects of word-length to avoid unbounded
conditions. The following section discusses the general theory and
applications behind IIR Filters.
IIR Filter Background
The impulse response of an IIR filter extends for an infinite amount of
time because their output is based on feedback from previous outputs.
The general expression for IIR filters is:
n
y( n) =
n
∑ a ( i )x ( n – i ) – ∑ b ( i )y ( n – i )
i=0
i=1
where ai and bi represent the coefficients in the feed-forward path and
feedback path, respectively, and n represents the filter order. These
coefficients determine where the poles and zeros of the IIR filter lie.
Consequently, they also determine how the filter functions (i.e., cut-off
frequencies, band pass, low pass, etc.).
The feedback feature makes IIR filters useful in high data throughput
applications that require low hardware usage. However, feedback also
introduces complications and caution must be taken to make sure these
filters are not exposed to situations in which they may become unstable.
The complications include phase distortion and finite word length effects,
but these can be overcome by ensuring that the filter always operates
within its intended range.
Figure 7–20 shows a direct form II structure of an IIR filter.
7–34
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
Figure 7–20. Direct Form II Structure of an IIR Filter
w(n)
Σ
X(n)
Z
Σ
-b1
Σ
-b2
Σ
-bn
Z
Z
a0
Σ
a1
Σ
a2
Σ
an
Σ
Y(n)
-1
-1
-1
The transfer function for an IIR filter is:
n
∑ ai z
–i
i=0
H ( z ) = -----------------------------n
1+
∑ bi z
–i
i=1
The numerator contains the zeros of the filter and the denominator
contains the poles. The IIR filter structure requires a multiplication
followed by an accumulation. Constructing the filter directly from the
transfer function shown above may result in finite word length
limitations and make the filter potentiality unstable. This becomes more
critical as the filter order increases, because it only has a finite number of
bits to represent the output. To prevent overflow or instability, the
transfer function can be split into two or more terms representing several
second order filters called biquads. These biquads can be individually
scaled and cascaded, splitting the poles into multiples of two. For
example, an IIR filter having ten poles should be split into five biquad
sections. Doing this minimizes quantization and recursive accumulation
errors.
Altera Corporation
September 2004
7–35
Stratix Device Handbook, Volume 2
Infinite Impulse Response (IIR) Filters
This cascaded structure rearranges the transfer function. This is shown in
the equation below, where each product term is a second order IIR filter.
If n is odd, the last product term is a first order IIR filter:
(n + 1) ⁄ 2
H(z) = C ×
∏
k=1
–1
–2
a 0k + a 1k z + a 2k z
--------------------------------------------------- = C×
–1
–2
1 + b 1k z + b 2k z
(n + 1) ⁄ 2
∏
Hk ( z )
k=1
Figure 7–21 shows the cascaded structure.
Figure 7–21. Cascaded IIR Filter
x(n)
C
H 1 (z)
H k (z)
H n (z)
y(n)
Basic IIR Filters
In this section, the basic IIR filter is implemented using cascaded second
order blocks or biquads in the direct form II structure.
Basic IIR FIlter Implementation
Multiplier blocks, adders and delay elements can implement a basic IIR
filter. The Stratix architecture lends itself to IIR filters because of its
embedded DSP blocks, which can easily be configured to perform these
operations. The altmult_add megafunction can be used to implement
the multiplier-adder mode in the DSP blocks. Figure 7–22 shows the
implementation of an individual biquad using Stratix and Stratix GX DSP
blocks.
7–36
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
Figure 7–22. IIR Filter Biquad Note (1)
Note to Figure 7–22:
(1)
Altera Corporation
September 2004
Unused ports are grayed out.
7–37
Stratix Device Handbook, Volume 2
Infinite Impulse Response (IIR) Filters
The first DSP block in Figure 7–22 is configured in the two-multipliers
adder mode, and the second DSP block is in the four-multipliers adder
mode. For an 18-bit input to the IIR filter, each biquad requires five
multipliers and five adders (two DSP blocks). One of the adders is
implemented using logic elements. Cascading several biquads together
can implement more complex, higher order IIR filters. It is possible to
insert registers in between the biquad stages to improve the performance.
Figure 7–23 shows a 4thorder IIR filter realized using two cascaded
biquads in three DSP blocks.
Figure 7–23. Two Cascaded Biquads
First
biquad
a10
a11
Four-multipliers
adder mode
a12
DSP
block 1
b11
b12
x[n]
Two-multipliers
adder mode
b21
b22
a20
DSP
block 2
Four-multipliers
adder mode
Second
biquad
y[n]
a21
a22
7–38
Stratix Device Handbook, Volume 2
DSP
block 3
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
Basic IIR Filter Implementation Results
Table 7–15 shows the results of implementing a 4th order IIR filter in a
Stratix device.
Table 7–15. 4th Order IIR Filter Implementation Results
Part
EP1S10F780C5
Utilization
Lcell: 102/10570(<1%)
DSP Block 9-bit elements: 24/48 (50%)
Memory bits: 0/920448(0%)
Performance
73 MHz
Latency
4 clock cycles
Basic IIR Filter Design Example
Download the 4th Order IIR Filter (iir.zip) design example from the
Design Examples section of the Altera web site at www.altera.com.
Butterworth IIR Filters
Butterworth filters are the most popular version of IIR analog filters.
These filters are also known as “maximally flat” because they have no
passband ripple. Additionally, they have a monotonic response in both
the stopband and the passband. Butterworth filters trade-off roll off
steepness for their no ripple characteristic. The distinguishing
Butterworth filter feature is its poles are arranged in a uniquely
symmetrical manner along a circle in the s-plane. The expression for the
Butterworth filter’s magnitude-squared function is as follows:
H c ( jω)
2
1
= --------------------------jω 2N
⎛
1 + -------⎞
⎝ jωc⎠
where:
ωc is the cut-off frequency
N is the filter order
The filter’s cutoff characteristics become sharper as N increases. If a
substitution is made such that jω = s, then the following equation is
derived:
Altera Corporation
September 2004
7–39
Stratix Device Handbook, Volume 2
Infinite Impulse Response (IIR) Filters
1
H c ( s )H c ( – s ) = ---------------------------s 2N
1 + ⎛ -------⎞
⎝ jωc⎠
with poles at:
sk = ( –1 )
= ωc e
1
------2N
( jωc )
jπ ⎞
⎛ ------( 2k + N – 1 )
⎝ 2N⎠
for k=0,1,…,2N-1
There are 2N poles on the circle with a radius of ωc in the s-plane. These
poles are evenly spaced at π/N intervals along the circle. The poles chosen
for the implementation of the filter lie in the left half of the s-plane,
because these generate a stable, causal filter.
Each of the impulse invariance, the bilinear, and matched z transforms
can transform the Laplace transform of the Butterworth filter into the ztransform.
■
Impulse invariance transforms take the inverse of the Laplace
transform to obtain the impulse response, then perform a
z-transform on the sampled impulse response. The impulse
invariance method can cause some aliasing.
■
The bilinear transform maps the entire jω-axis in the s-plane to one
revolution of the unit circle in the z-plane. This is the most popular
method because it inherently eliminates aliasing.
■
The matched z-transform maps the poles and the zeros of the filter
directly from the s-plane to the z-plane. Usually, these transforms are
transparent to the user because several tools, such as MATLAB, exist
for determining the coefficients of the filter. The z-transform
generates the coefficients much like in the basic IIR filter discussed
earlier.
Butterworth Filter Implementation
For digital designs, consideration must be made to optimize the IIR
biquad structure so that it maps optimally into logic. Because speed is
often a critical requirement, the goal is to reduce the number of
operations per biquad. It is possible to reduce the number of multipliers
needed in each biquad to just two.
7–40
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
Through the use of integer feedforward multiplies, which can be
implemented by combining addition, shifting, and complimenting
operations, a Butterworth filter’s transfer function biquad can be
optimized for logic synthesis. The most efficient transformation is that of
an all pole filter. This is because there is a unique relationship between the
feedforward integer coefficients of the filter represented as:
–1
–2
1 + 2z + z H ( z ) = -----------------------------------------–1
–2
1 + b1 z + b2 z
As can be seen by this equation, the z-1 coefficient in the numerator
(representing the feedforward path) is twice the other two operands (z-2
and 1). This is always the case in the transformation from the frequency
to the digital domain. This represents the normalized response, which is
faster and smaller to implement in hardware than real multipliers. It
introduces a scaling factor as well, but this can be corrected at the end of
the cascade chain through a single multiply.
Figure 7–24 shows how a Butterworth filter biquad is implemented in a
Stratix or Stratix GX device.
Altera Corporation
September 2004
7–41
Stratix Device Handbook, Volume 2
Infinite Impulse Response (IIR) Filters
Figure 7–24. Butterworth Filter Biquad Notes (1), (2)
w(n-1)
DSP block
b1
D
w(n)
Q
w(n-2)
b2
D
Q
x(n)
w(n)
w(n-1)
y(n)
Notes to Figure 7–24:
(1)
Unused ports are grayed out.
(2)
The z-1 coefficient is a multiple of the other coefficients (z-2 and 1) in the
feedforward path. This is implemented using a shift operation.
7–42
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
The DSP block in Figure 7–24 is configured in multiply and add mode.
The three external adders are implemented in logic elements and
therefore are not part of the DSP block. Therefore, for an 18-bit input, each
biquad requires half a DSP block and three logic element adders. The gain
factor can be compensated for at the end of the filtering stage and is not
shown in this simple example. More complex, higher order Butterworth
filters can be realized by cascading several biquads together, as in the IIR
example. Figure 7–25 below shows a 4th order Butterworth filter using
two cascaded biquads in a single DSP block.
Altera Corporation
September 2004
7–43
Stratix Device Handbook, Volume 2
Infinite Impulse Response (IIR) Filters
Figure 7–25. Cascaded Butterworth Biquads Note (1)
D
Q
D
Q
D
Q
w1(n-2)
First
biquad
w1(n-1)
w1(n-1)
DSP
block
b11
D
Q
w1(n)
w1(n-2)
D
Q
b12
D
Q
x(n)
w2(n-1)
D
b21
D
Q
Second
biquad
Q
w2(n)
w2(n-2)
b22
D
D
Q
D
Q
D
Q
Q
w2(n-1)
w2(n-2)
y(n)
Note to Figure 7–25:
(1)
The gain factor is compensated for at the end of the filtering stage and is not shown in this figure.
7–44
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
Butterworth Filter Implementation Results
Table 7–16 shows the results of implementing a 4th order Butterworth
filter as shown in Figure 7–25.
Table 7–16. 4th Order Butterworth Filter Implementation Results
Part
EP1S10F780C6
Utilization
Lcell: 251/10570(2%)
DSP Block 9-bit elements: 16/48 (33%)
Memory bits: 0/920448 (0%)
Performance
80 MHz
Latency
4 clock cycles
Butterworth Filter Design Example
Download the 4th Order Butterworth Filter (butterworth.zip) design
example from the Design Examples section of the Altera web site at
www.altera.com.
Matrix
Manipulation
DSP relies heavily on matrix manipulation. The key idea is to transform
the digital signals into a format that can then be manipulated
mathematically.
This section describes an example of matrix manipulation used in 2-D
convolution filter, and its implementation in a Stratix device.
Background on Matrix Manipulation
A matrix can represent all digital signals. Apart from the convenience of
compact notation, matrix representation also exploits the benefits of
linear algebra. As with one-dimensional, discrete sequences, this
advantage becomes more apparent when processing multi-dimensional
signals.
In image processing, matrix manipulation is important because it
requires analysis in the spatial domain. Smoothing, trend reduction, and
sharpening are examples of common image processing operations, which
are performed by convolution. This can also be viewed as a digital filter
operation with the matrix of filter coefficients forming a convolutional
kernel, or mask.
Altera Corporation
September 2004
7–45
Stratix Device Handbook, Volume 2
Matrix Manipulation
Two-Dimensional Filtering & Video Imaging
FIR filtering for video applications and image processing in general is
used in many applications, including noise removal, image sharpening to
feature extraction.
For noise removal, the goal is to reduce the effects of undesirable,
contaminative signals that have been linearly added to the image.
Applying a low pass filter or smoothing function flattens the image by
reducing the rapid pixel-to-pixel variation in gray levels and, ultimately,
removing noise. It also has a blurring effect usually used as a precursor
for removing unwanted details before extracting certain features from the
image.
Image sharpening focuses on the fine details of the image and enhances
sharp transitions between the pixels. This acts as a high-pass filter that
reduces broad features like the uniform background in an image and
enhances compact features or details that have been blurred.
Feature extraction is a form of image analysis slightly different from
image processing. The goal of image analysis in general is to extract
information based on certain characteristics from the image. This is a
multiple step process that includes edge detection. The easiest form of
edge detection is the derivative filter, using gradient operators.
All of the operations above involve transformation of the input image.
This can be presented as the convolution of the two-dimensional input
image, x(m,n) with the impulse response of the transform, f(k,l), resulting
in y(m,n) which is the output image.
y ( m, n ) = f ( k, l ) ⊗ x ( m, n )
N
y ( m, n ) =
N
∑ ∑ f ( k,
l )x ( m – k, n – l )
k = –N l = –N
The f(k,l) function refers to the matrix of filter coefficients. Because the
matrix operation is analogous to a filter operation, the matrix itself is
considered the impulse response of the filter. Depending on the type of
operation, the choice of the convolutional kernel or mask, f(k,l) is
different. Figure 7–26 shows an example of convolving a 3 × 3 mask with
a larger image.
7–46
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
Figure 7–26. Convolution Using a 3 × 3 Kernel
The output pixel value, y(m,n) depends on the surrounding pixel values
in the input image, as well as the filter weights:
y ( m, n ) = w 1 x ( m – 1, n – 1 ) + w 2 x ( m – 1, n ) + w 3 x ( m – 1, n + 1 )
+ w 4 x ( m, n – 1 ) + w 5 x ( m, n ) + w 6 x ( m, n + 1 )
+ w 7 x ( m + 1, n – 1 ) + w 8 x ( m + 1, n ) + w 9 x ( m + 1, n + 1 )
To complete the transformation, the kernel slides across the entire image.
For pixels on the edge of the image, the convolution operation does not
have a complete set of input data. To work around this problem, the pixels
on the edge can be left unchanged. In some cases, it is acceptable to have
an output image of reduced size. Alternatively, the matrix effect can be
applied to edge pixels as if they are surrounded on the “empty” side by
Altera Corporation
September 2004
7–47
Stratix Device Handbook, Volume 2
Matrix Manipulation
black pixels, that is pixels with value zero. This is similar to padding the
edges of the input image matrix with zeros and is referred to as the free
boundary condition. This is shown in Figure 7–27.
Figure 7–27. Using Free Boundary Condition for Edge Pixels
3 x 3 kernel slides across image
0
0
0
0
Image boundary
x m , n + 1)
x(
x
x(
Image
0
x m + 1, n )
x(
x m + 1, n + 1)
x(
Image boundary
Convolution Implementation
This design example shows a 3 × 3 2-D FIR filter that takes in an 8 × 8
input image with gray pixel values ranging from 0-255 (8-bit). Data is fed
in serially starting from the top left pixel, moving horizontally on a rowby-row basis. Next the data is stored in three separate RAM blocks in the
buffering stage. Each M512 memory block represents a line of the image,
and this is cycled through. For a 32 × 32 input image, the design needs
M4K memory blocks. For larger images (640 × 480), this can be extended
to M-RAM blocks or other buffering schemes. The control logic block
provides the RAM control signals to interleave the data across all three
7–48
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
RAM blocks. The 9-bit signed filter coefficients feed directly into the filter
block. As the data is shifted out from the RAM blocks, the multiplexer
block checks for edge pixels and uses the free boundary condition.
Figure 7–28 shows a top-level diagram of the design.
Figure 7–28. Block Diagram on Implementation of 3 × 3 Convolutional Filter for an 8 × 8 Pixel Input Image
The 3 × 3 filter block implements the nine multiply-add operations in
parallel using two DSP blocks. One DSP block can implement eight of
these multipliers. The second DSP block implements the ninth multiplier.
The first DSP block is in the four-multipliers adder mode, and the second
is in simple multiplier mode. In addition to the two DSP blocks, an
external adder is required to sum the output of all nine multipliers.
Figure 7–29 shows this implementation.
Altera Corporation
September 2004
7–49
Stratix Device Handbook, Volume 2
Matrix Manipulation
Figure 7–29. Implementation of 3 × 3 Convolutional Filter Block
DSP Block in Four-Multipliers Adder Mode (9-bit)
w1
x(m - 1, n - 1)
w2
x(m - 1, n )
w3
x(m - 1, n + 1)
w4
x(m , n - 1)
w5
LE implemented
adder
x(m , n )
w6
x(m , n + 1)
y(m , n )
w7
x(m + 1, n - 1)
w8
x(m + 1, n )
DSP Block in Simple Multiplier Mode (8-bit)
w9
x(m + 1, n + 1)
Note: Unused multipliers and
adders grayed out. These
multipliers can be used by other
functions.
7–50
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
In cases where a symmetric 2-D filter is used, pixels sharing the same
filter coefficients from three separate line-stores A, B, and C can be added
together prior to the multiplication operation. This reduces the number of
multipliers used. Referring to Figure 7–30, w1, w2, and w3 are the filter
coefficients. Figure 7–31 shows the implementation of this circular
symmetric filter.
Figure 7–30. Symmetric 3 × 3 Kernel
Figure 7–31. Details on Implementation of Symmetric 3 × 3 Convolution Filter Block
Logic Elements
DSP Block - Four Multipliers
Adder Mode (9-bit)
A
w1
y(m , n )
w2
B
w3
C
GND
GND
Note: Unused multipliers and adders grayed out. These
multipliers can be used by other functions.
Altera Corporation
September 2004
7–51
Stratix Device Handbook, Volume 2
Discrete Cosine Transform (DCT)
Convolution Implementation Results
Table 7–17 shows the results of the 3 × 3 2-D FIR filter implementation in
Figure 7–28.
Table 7–17. 3 × 3 2-D Convolution Filter Implementation Results
Part
EP1S10F780
Utilization
Lcell: 372/10570 (3%)
DSP block 9-bit elements: 9/48 (18%)
Memory bits: 768/920448 (<1%)
Performance
226 MHz
Latency
15 clock cycles
The design requires the input to be an 8 × 8 image, with 8-bit input data
and 9-bit filter coefficient width. The output is an image of the same size.
Convolution Design Example
Download the 3 × 3 2-D Convolutional Filter (two_d_fir.zip) design
example from the Design Examples section of the Altera web site at
www.altera.com.
Discrete Cosine
Transform (DCT)
The discrete cosine transform (DCT) is widely used in video and audio
compression, for example in JPEG, MPEG video, and MPEG audio. It is a
form of transform coding, which is the preferred method for compression
techniques. Images tend to compact their energy in the frequency domain
making compression in the frequency domain much more effective. This
is an important element in compressing data, where the goal is to have a
high data compression rate without significant degradation in the image
quality.
DCT Background
Similar to the discrete fourier transform (DFT), the DCT is a function that
maps the input signal or image from the spatial to the frequency domain.
It transforms the input into a linear combination of weighted basis
functions. These basis functions are the frequency components of the
input data.
7–52
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
For 1-D with input data x(n) of size N, the DCT coefficients Y(k) are:
α( k )
Y ( k ) = ----------2
N–1
( 2n + 1 )πk
-⎞
∑ x ( n ) cos ⎛⎝ -------------------------⎠
2N
for 0 ≤ k ≤ N–1
n=0
where:
α( k ) =
1
---N
for k = 0
α( k ) =
2--N
for 1 ≤ k ≤ N –1
For 2-D with input data x(m,n) of size N × N, the DCT coefficients for the
output image, Y(p,q) are:
α( p )α( q )
Y ( p, q ) = ----------------------2
N–1N–1
∑ ∑ x ( m,
( 2m + 1 )πp
( 2n + 1 )πq
n ) cos ⎛ -----------------------------⎞ cos ⎛ ---------------------------⎞
⎝
⎠
⎝
⎠
2N
2N
m = 0n = 0
where:
α( p ) =
1--N
for p = 0
α( q ) =
1
---N
for q = 0
α( p ) =
2--N
for 1 ≤ p ≤ N –1
α( q ) =
2--N
for 1 ≤ q ≤ N–1
2-D DCT Algorithm
The 2-D DCT can be thought of as an extended 1-D DCT applied twice;
once in the x direction and again in the y direction. Because the 2-D DCT
is a separable transform, it is possible to calculate it using efficient 1-D
algorithms. Figure 7–32 illustrates the concept of a separable transform.
Altera Corporation
September 2004
7–53
Stratix Device Handbook, Volume 2
Discrete Cosine Transform (DCT)
Figure 7–32. A 2-D DCT is a Separable Transform
This section uses a standard algorithm proposed in [1]. Figure 7–33
shows the flow graph for the algorithm. This is similar to the butterfly
computation of the fast fourier transform (FFT). Similar to the FFT
algorithms, the DCT algorithm reduces the complexity of the calculation
by decomposing the computation into successively smaller DCT
components. The even coefficients (y0, y2, y4, y6) are calculated in the
upper half and the odd coefficients (y1, y3, y5, y7) in the lower half. As a
result of the decomposition, the output is reordered as well.
Figure 7–33. Implementing an N=8 Fast DCT
Stage 1
Stage 4
Stage 2 Stage 3
y0
x0
x1
C4
y4
x2
C6 -C2
y2
x3
C2 C6
y6
x4
C7 -C C
5 3 -C1
C5 -C C
1 7 C3
C3 -C -C
7 1 -C5
C1 C C
3 5 C7
y1
x5
x6
x7
a
y3
y5
y7
Sum a and b
b
Multiplied by -1
S31
S32
.
.
.
S3n
7–54
Stratix Device Handbook, Volume 2
Cm1
C.m2
.
.
.
Cmn
yk
Multiply-addition block
Stage 3 output (S3)
Matrix coefficent (Cmn)
yk = cm1s31 + cm2s32 + ... + cmns3n
where cx = cos
(16x π)
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
The following defines in matrix format, the 8-point 1-D DCT of
Figure 7–33:
Y 1D = x × Add 1 × Add 2 × Add 3 × C
where:
[x] is the 1 × 8 input matrix
Altera Corporation
September 2004
Add 1 =
1
0
0
0
0
0
0
1
0
1
0
0
0
0
1
0
0
0
1
0
0
1
0
0
0
0
0
1
1
0
0
0
0
0
0
1
–1
0
0
0
0
0
1
0
0
–1
0
0
0
1
0
0
0
0
–1
0
1
0
0
0
0
0
0
–1
Add 2 =
1
0
0
1
0
0
0
0
0
1
1
0
0
0
0
0
0
1
–1
0
0
0
0
0
1
0
0
–1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
1
Add 3 =
1
1
0
0
0
0
0
0
1
–1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
1
7–55
Stratix Device Handbook, Volume 2
Discrete Cosine Transform (DCT)
C =
1
0
0
C4
0
0
0
0
0
0
0
0
0
0
0
0
0
0
C6 –C2 0
0
0
0
0
0
C2 C6
0
0
0
0
0
0
0
C7 –C5 C3 –C1
0
0
0
0
C5 –C1 C7 C3
0
0
0
0
C3 –C7 –C1 –C5
0
0
0
0
C1 C3 C5 C7
0
C x = cos πx
-----16
All of the additions in stages 1, 2 and 3 of Figure 7–32 appear in
symmetric add and subtract pairs. The entire first stage is simply four
such pairs in a very typical cross-over pattern. This pattern is repeated in
stages 2 and 3. Multiplication operations are confined to stage 4 in the
algorithm. This implementation is shown in more detail in the next
section.
DCT Implementation
In taking advantage of the separable transform property of the DCT, the
implementation can be divided into separate stages; row processing and
column processing. However, some data restructuring is necessary
before applying the column processing stage to the results from the row
processing stage. The data buffering stage must transpose the data first.
Figure 7–34 shows the different stages.
Figure 7–34. Three Separate Stages in Implementing the 2-D DCT
Row
processing
Transpose
matrix
Column
processing
Because the row processing and column processing blocks share the same
1-D 8-point DCT algorithm, the hardware implementation shows this
block as being shared. The DCT algorithm requires a serial-to-parallel
conversion block at the input because it works on blocks of eight data
7–56
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
points in parallel. There is also a parallel-to-serial conversion block at the
output because the column processing stage generates the output image
column-by-column. In order to have the output in the same order as the
input (i.e., row-by-row), this conversion is necessary. Appropriate scaling
needs to be applied to the completed transform but this can be combined
with the quantization stage which often follows a DCT [1]. Figure 7–35
shows a top-level block diagram of this design.
Figure 7–35. Block Diagram on Serial Implementation of 2-D DCT
The implementation of the 1-D DCT block is based on the algorithm
shown in Figure 7–33. The simple addition and subtraction operations in
stages 1, 2 and 3 are implemented using logic cells. The multiply and
multiply-addition operations in stage 4 are implemented using DSP
blocks in the Stratix device in the simple multiplier mode, two-multiplier
adder mode, and the four-multiplier adder mode. An example of the
multiply-addition block is shown in Figure 7–36.
Altera Corporation
September 2004
7–57
Stratix Device Handbook, Volume 2
Discrete Cosine Transform (DCT)
Figure 7–36. Details on the Implementation of the Multiply-Addition Operation
in Stage 4 of the 1-D DCT Algorithm
DSP Block - Four-Multipliers Adder Mode (18-bit)
S31
Cm1
S32
Cm2
yk
S33
Cm3
S34
Cm4
Note to Figure 7–36:
(1)
Referring to Figure 7–33. S3n is an output from stage 3 of the DCT and Cmn is a
matrix coefficient. Cx = cos (xπ/16).
DCT Implementation Results
Table 7–18 shows the results of implementing a 2-D DCT with 18-bit
precision, as shown in Figure 7–35.
Table 7–18. 2-D DCT Implementation Results
Part
EP1S20F780
Utilization
Lcell: 1717/18460 (9%)
DSP Block 9-bit element: 18/80 (22%)
Memory bits: 2816/1669248 (<1%)
Performance
165 MHz
Latency
80 clock cycles
DCT Design Example
Download the 2-D convolutional filter (d_dct.zip) design example from
the Design Examples section of the Altera web site at www.altera.com.
7–58
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
Arithmetic
Functions
Arithmetic functions, such as trigonometric functions, including sine,
cosine, magnitude and phase calculation, are important DSP elements.
This section discusses the implementation of a simple vector magnitude
function in a Stratix device.
Background
Complex numbers can be expressed in two parts: real and imaginary.
z = a + jb
where:
a is the real part
b is the imaginary part
j2= –1
In a two-dimensional plane, a vector (a,b) with reference to the origin
(0,0) can also be represented as a complex number. In essence, the x-axis
represents the real part, and the y-axis represents the imaginary part (see
Figure 7–37).
Figure 7–37. Magnitude of Vector (a,b)
Complex numbers can be converted to phase and amplitude or
magnitude representation, using a Cartesian-to-polar coordinate
conversion. For a vector (a,b), the phase and magnitude representation is
the following:
Altera Corporation
September 2004
7–59
Stratix Device Handbook, Volume 2
Arithmetic Functions
Magnitude m =
2
a +b
2
Phase angle θ = tan-1(b/a)
This conversion is useful in different applications, such as position
control and position monitoring in robotics. It is also important to have
these transformations at very high speeds to accommodate real-time
processing.
Arithmetic Function Implementation
A common approach to implementing these arithmetic functions is using
the coordinate rotation digital computer (CORDIC) algorithm. The
CORDIC algorithm calculates the trigonometric functions of sine, cosine,
magnitude, and phase using an iterative process. It is made up of a series
of micro-rotations of the vector by a set of predetermined constants,
which are powers of 2.
Using binary arithmetic, this algorithm essentially replaces multipliers
with shift and add operations. In Stratix devices, it is possible to calculate
some of these arithmetic functions directly, without having to implement
the CORDIC algorithm.
This section describes a design example that calculates the magnitude of
a 9-bit signed vector (a,b) using a pipelined version of the square root
function available at the Altera IP Megastore. To calculate the sum of the
squares of the input (a2 + b2), configure the DSP block in the twomultipliers adder mode. The square root function is implemented using
an iterative algorithm similar to the long division operation. The binary
numbers are paired off, and subtracted by a trial number. Depending on
if the remainder is positive or negative, each bit of the square root is
determined and the process is repeated. This square root function does
not require memory and is implemented in logic cells only.
In this example, the input bit precision (IN_PREC) feeding into the square
root macro is set to twenty, and the output precision (OUT_PREC) is set to
ten. The number of precision bits is parameterizable. Also, there is a third
parameter, PIPELINE, which controls the architecture of the square root
macro. If this parameter is set to YES, it includes pipeline stages in the
square root macro. If set to NO, the square root macro becomes a singlecycled combinatorial function.
Figure 7–38 shows the implementation the magnitude design.
7–60
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
Figure 7–38. Implementing the Vector Magnitude Function
DSP Block - Two Multipliers
Adder Mode (9-bit)
LE implemented
square root function
a
b
Note: Unused multipliers
and adders grayed out.
Altera Corporation
September 2004
7–61
Stratix Device Handbook, Volume 2
Conclusion
Arithmetic Function Implementation Results
Table 7–19 shows the results of the implementation shown in Figure 7–38
with the PIPELINE parameter set to YES. Table 7–20 shows the results of
the implementation shown in Figure 7–38 with the PIPELINE parameter
set to NO.
Table 7–19. Vector Magnitude Function Implementation Results
(PIPELINE=YES)
Part
EP1S10F780
Utilization
Lcell: 497/10570 (4%)
DSP block 9-bit elements: 2/48 (4%)
Memory bits: 0/920448 (0%)
Performance
194 MHz
Latency
15 clock cycles
Table 7–20. Vector Magnitude Function Implementation Results
(PIPELINE=NO)
Part
EP1S10F780
Utilization
Lcell: 244/10570 (2%)
DSP block 9-bit elements: 2/48 (4%)
Memory bits: 0/920448 (0%)
Performance
30 MHz
Latency
3 clock cycles
Arithmetic Function Design Example
Download the Vector Magnitude Function (magnitude.zip) design
example from the Design Examples section of the Altera web site at
www.altera.com.
Conclusion
The DSP blocks in Stratix and Stratix GX devices are optimized to support
DSP functions requiring high data throughput, such as FIR filters, IIR
filters and the DCT. The DSP blocks are flexible and configurable in
different operation modes based on the application’s needs. The
TriMatrix memory provides the data storage capability often needed in
DSP applications.
The DSP blocks and TriMatrix memory in Stratix and Stratix GX devices
offer performance and flexibility that translates to higher performance
DSP functions.
7–62
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Implementing High Performance DSP Functions in Stratix & Stratix GX Devices
References
Altera Corporation
September 2004
See the following for more information:
■
Optimal DCT for Hardware Implementation
M. Langhammer. Proceedings of International Conference on Signal
Processing Applications & Technology (ICSPAT) '95, October 1995
■
Digital Signal Processing: Principles, Algorithms, and Applications
John G. Proakis, Dimitris G. Manolakis. Prentice Hall
■
Hardware Implementation of Multirate Digital Filters
Tony San. Communication Systems Design, April 2000
■
AN 73: Implementing FIR Filters in FLEX Devices
■
Efficient Logic Synthesis Techniques for IIR Filters
M.Langhammer. Proceedings of International Conference on Signal
Processing Applications & Technology (ICSPAT) '95, October 1995
7–63
Stratix Device Handbook, Volume 2
References
7–64
Stratix Device Handbook, Volume 2
Altera Corporation
September 2004
Section V. IP & Design
Considerations
This section provides documentation on some of the IP functions offered
by Altera® for Stratix® devices. (Also see the Intellectual Property section
of the Altera web site for a complete offering of IP cores for Stratix
devices.) The last chapter details design considerations for migrating
from the APEX™ architecture.
This section contains the following chapters:
Revision History
Chapter
8
■
Chapter 8, Implementing 10-Gigabit Ethernet Using Stratix &
Stratix GX Devices
■
Chapter 9, Implementing SFI-4 in Stratix & Stratix GX Devices
■
Chapter 10, Transitioning APEX Designs to Stratix & Stratix GX
Devices
The table below shows the revision history for Chapters 8 through 10.
Date/Version
July 2005, v2.0
Updated Stratix GX device information.
●
Table 8–2 on page 8–10: updated table, deleted Note 1, and updated
Note 2.
Updated Table 8–4 on page 8–12.
November 2003, v1.1
●
Removed support for series and parallel on-chip termination.
April 2003, v1.0
●
No new changes in Stratix Device Handbook v2.0.
July 2005, v2.0
●
Updated Stratix GX device information.
September 2004, v1.1
●
●
Table 9–2 on page 9–9: updated table, deleted Note 1, and updated
Note 2.
Updated Table 9–4 on page 9–10.
●
No new changes in Stratix Device Handbook v2.0.
September 2004, v1.2
9
Changes Made
April 2003, v1.0
Altera Corporation
●
Section V–1
IP & Design Considerations
Stratix Device Handbook, Volume 2
Chapter
Date/Version
10
July 2005, v3.0
●
Updated Stratix GX device information.
September 2004, v2.1
●
Updated Table 10–9 on page 10–26.
April 2004, v2.0
●
●
Synchronous occurrences renamed pipelined.
Asynchronous occurrences renamed flow-through.
November 2003, v1.2
●
Removed support for series and parallel on-chip termination.
October 2003, v1.1
●
Updated Table 10–6.
April 2003, v1.0
●
No new changes in Stratix Device Handbook v2.0.
Section V–2
Changes Made
Altera Corporation
8. Implementing 10-Gigabit
Ethernet Using Stratix &
Stratix GX Devices
S52010-2.0
Introduction
Ethernet has evolved to meet ever-increasing bandwidth demands and is
the most prevalent local-area network (LAN) communications protocol.
10-Gigabit Ethernet extends that protocol to higher bandwidth for future
high-speed applications. The accelerated growth of network traffic and
the resulting increase in bandwidth requirements is driving service
providers and enterprise network architects towards high-speed network
solutions. Potential applications for 10-Gigabit Ethernet include private
campus or LAN backbones, high-speed access links between service
providers and enterprises, and aggregation and transport in metropolitan
area networks (MANs).
The I/O features of Stratix® and Stratix GX devices enable support for 10Gigabit Ethernet, supporting 10-Gigabit 16-bit interface (XSBI) and 10Gigabit medium independent interface (XGMII). Stratix GX devices can
additionally support the 10-gigabit attachment unit interface (XAUI)
using the embedded 3.125-Gbps transceivers. You can find more
information on XAUI support in Section II, Stratix GX Transceiver User
Guide, of the Stratix GX Device Handbook, Volume 1.
This chapter discusses the following topics:
■
■
■
■
■
Fundamentals of 10-Gigabit Ethernet
Description and implementation of XSBI
Description and implementation of XGMII
Description of XAUI
I/O characteristics of XSBI, XGMII, and XAUI
Related Links
■
■
■
10-Gigabit
Ethernet
Altera Corporation
July 2005
10-Gigabit Ethernet Alliance at www.10gea.org
The Stratix Device Family Data Sheet section of the Stratix Device
Handbook, Volume 1 and the Stratix GX Device Family Data Sheet
section of the Stratix GX Device Handbook, Volume 1
The High-Speed Differential I/O Interfaces in Stratix Devices chapter
Ethernet speed has increased to keep pace with demand, initially to
10 megabits per second (Mbps), later to 100 Mbps, and recently to
1 gigabit per second (Gbps). Ethernet is the dominant network
technology in LANs, and with the advent of 10-Gigabit Ethernet, it is
entering the MAN and wide area network (WAN) markets.
8–1
10-Gigabit Ethernet
The purpose of the 10-Gigabit Ethernet proposed standard is to extend
the operating speed to 10 Gbps defined by protocol IEEE 802.3 and
include WAN applications. These additions provide a significant increase
in bandwidth while maintaining maximum compatibility with current
IEEE 802.3 interfaces.
Since its inception in March 1999, the 10-Gigabit Ethernet Task Force has
been working on the IEEE 802.3ae Standard. Some of the information in
the following sections is derived from Clauses 46, 47, 49, and 51 of the
IEEE Draft P802.3ae/D3.1 document. A fully ratified standard is
expected in the first half of 2002. Figure 8–1 shows the relationship of
10-Gigabit Ethernet to the Open Systems Interconnection (OSI) protocol
stack.
Figure 8–1. 10-Gigabit Ethernet Protocol in Relation to OSI Protocol Stack
Higher Layers
OSI Reference
Model Layers
LLC (1)
MAC (2)
Application
Reconciliation
Presentation
XGMII
Session
Transport
PCS (3)
XSBI
PHY (4)
PMA (5)
Network
PMD (6)
Data Link
MDI (7)
Physical
Medium
Notes to Figure 8–1:
(1)
(2)
(3)
(4)
(5)
(6)
(7)
LLC: logical link controller
MAC: media access controller
PCS: physical coding sublayer
PHY: physical layer
PMA: physical medium attachment
PMD: physical medium dependent
MDI: medium dependent interface
8–2
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Implementing 10-Gigabit Ethernet Using Stratix & Stratix GX Devices
The Ethernet PHY (layer 1 of the OSI model) connects the media (optical
or copper) to the MAC (layer 2). The Ethernet architecture further divides
the PHY (layer 1) into a PMD sublayer, a PMA sublayer, and a PCS. For
example, optical transceivers are PMD sublayers. The PMA converts the
data between the PMD sublayer and the PCS sublayer. The PCS is made
up of coding (e.g., 8b/10b, 64b/66b) and serializer or multiplexing
functions. Figure 8–2 shows the components of 10-Gigabit Ethernet and
how Altera implements certain blocks and interfaces.
10-Gigabit Ethernet has three different implementations for the PHY:
10GBASE-X, 10GBASE-R, and 10GBASE-W. The 10GBASE-X
implementation is a PHY that supports the XAUI interface. The XAUI
interface used in conjunction with the XGMII extender sublayer (XGXS)
allows more separation in distance between the MAC and PHY.
10GBASE-X PCS uses four lanes of 8b/10b coded data at a rate of
3.125 Gbps. 10GBASE-X is a wide wave division multiplexing (WWDM)
LAN PHY. 10GBASE-R and 10GBASE-W are serial LAN PHYs and serial
WAN PHYs, respectively. Unlike 10GBASE-X, 10GBASE-R and
10GBASE-W implementations have a XSBI interface and are described in
more detail in the following section.
Altera Corporation
July 2005
8–3
Stratix Device Handbook, Volume 2
10-Gigabit Ethernet
Figure 8–2. 10-Gigabit Ethernet Block Diagram
Interface directly covered in this
application note
MAC
Interface indirectly covered in this
application note
RS (1)
Can be implemented in Altera PLDs
XGMII (32 Bits at 156.25 Mbps DDR 1.5-V HSTL)
8b/10b
XGXS (2)
XAUI (4 Bits at 3.125 Gbps PCML)
XGXS
8b/10b
XGMII (32 Bits at 156.25 Mbps DDR 1.5-V HSTL)
PCS
8B/10B
PCS
64b/66b
PCS
64b/66b
WIS (3)
PHY
PMA
XSBI (16 Bits at
644.5 Mbps LVDS)
PMA
PMD
PMD
MDI
10GBASE-X
XSBI (16 Bits at
622.08 Mbps LVDS)
PMA
PMD
OC-192 Framing
MDI
10GBASE-R
10GBASE-W
Notes to Figure 8–2:
(1)
(2)
(3)
The reconciliation sublayer (RS) interfaces the serial MAC data stream and the parallel data of XGMII.
The XGMII extender sublayer (XGXS) extends the distance of XGMII when used with XUAI and provides the data
conversion between XGMII and XAUI.
The WAN interface sublayer (WIS) implements the OC-192 framing and scrambling functions.
8–4
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Implementing 10-Gigabit Ethernet Using Stratix & Stratix GX Devices
Interfaces
The following sections discuss XSBI, PCS, XGMII, and XAUI.
XSBI
One of the blocks of 10-Gigabit Ethernet is the XSBI interface. XSBI is the
interface between the PCS and the PMA sublayers of the PHY layer of the
OSI model. XSBI supports two types of PHY layers, LAN PHY and WAN
PHY. The LAN PHY is part of 10GBASE-R, and supports existing
Gigabit Ethernet applications at ten times the bandwidth. The WAN PHY
is part of 10GBASE-W, and supports connections to existing and future
installations of SONET/SDH circuit-switched access equipment.
10GBASE-R is a physical layer implementation that is comprised of the
PCS sublayer, the PMA, and the PMD. 10GBASE-R is based upon
64b/66b data coding. 10GBASE-W is a PHY layer implementation that is
comprised of the PCS sublayer, the WAN interface sublayer (WIS), the
PMA, and the PMD. 10GBASE-W is based on STS-192c/SDH VC-4-64c
encapsulation of 64b/66b encoded data. Figure 8–3 shows the
construction of these two PHY layers.
Figure 8–3. XSBI Interface for the Two PHY Layers
PCS
PCS
WIS
XSBI
PHY
PMA
PMA
PMD
PMD
MDI
Altera Corporation
July 2005
Medium
Medium
10GBASE-R
10GBASE-W
8–5
Stratix Device Handbook, Volume 2
Interfaces
Functional Description
XSBI uses 16-bit LVDS data to interface between the PCS and the PMA
sublayer. Figure 8–4 shows XSBI between these two sublayers.
Figure 8–4. XSBI Functional Block Diagram
REFCLK
Transmitter
TX_D[15..0]
Transmitter
PMA_TXCLK
PCS
PMA
PMA_TXCLK_SRC
Receiver
Receiver
RX_D[15..0]
PCS
PMA_RXCLK
PMA
Sync_Err (optional)
On the transmitter side, the transmit data (TX_D[15..0]) is output by
the PCS and input at the PMA using the transmitter clock (PMA_TXCLK),
which is derived from the PMA source clock (PMA_TXCLK_SRC). The
PMA source clock is generated from the PMA with its reference clock
(REFCLK). On the receiver side, the receiver data (RX_D[15..0]) is
output by the PMA and input at the PCS using the PMA-generated
receiver clock (PMA_RXCLK). The SYNC_ERR optional signal is sent to the
PCS by the PMA if the PMA fails to recover the clock from the serial data
stream.
The ratios for these two clocks and data are dependent on the type of PHY
used. Table 8–1 shows the rates for both PHY types.
Table 8–1. XSBI Clock & Data Rates for WAN & LAN PHY
Parameter
WAN PHY
LAN PHY
Unit
TX_D[15..0]
622.08
644.53125
Mbps
PMA_TXCLK
622.08
644.53125
MHz
PMA_TXCLK_SRC
622.08
644.53125
MHz
RX_D[15..0]
622.08
644.53125
Mbps
PMA_RXCLK
622.08
644.53125
MHz
8–6
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Implementing 10-Gigabit Ethernet Using Stratix & Stratix GX Devices
Implementation
The 16-bit full duplex LVDS implementation of XSBI in Stratix devices is
shown in Figure 8–5.
The source-synchronous I/O implemented in Stratix GX devices
optionally includes dynamic phase alignment (DPA). DPA automatically
and continuously tracks fluctuations caused by system variations and
self-adjusts to eliminate the phase skew between the multiplied clock and
the serial data, allowing for data rates of 1 Gbps. In non DPA mode the
I/O behaves similarly to that of the Stratix I/O. This document assumes
that DPA is disabled. However, it is simple to implement the same system
with DPA enabled to take advantage of its features. For more information
on DPA, see the Stratix GX Transceivers chapter in the Stratix GX Device
Handbook, Volume 1.
Figure 8–5. Stratix & Stratix GX Device XSBI Implementation
PMA
Data
TX_D[15..0]
Transmitter
SERDES
PMA_TXCLK
Stratix & Stratix GX PCS
PLL1
Stratix &
Stratix GX
Logic Array
×1
PMA_TXCLK_SRC
÷8
Transmitter
Transmitter
Receiver
PLL2
Phase Shift
÷8
180˚
Receiver
SERDES
PMA_RXCLK
RX_D[15..0]
Data
Receiver
Altera Corporation
July 2005
8–7
Stratix Device Handbook, Volume 2
Interfaces
The transmit serializer/deserializer (SERDES) clock comes from the
transmitter clock source (PMA_TXCLK_SRC). The receiver SERDES clock
comes from the PMA receiver recovered clock (PMA_RXCLK).
Figure 8–6 shows the transmitter output of the XSBI core. Data
transmitted from the PCS to the PMA starts at the core of the Stratix or
Stratix GX device and travels to the Stratix or Stratix GX transmitter
SERDES block. The transmitter SERDES block converts the parallel data
to serial data for 16 individual channels (TX_D[15..0]). The PMA
source clock (PMA_TXCLK_SRC) is used to clock out the signal data.
PMA_TXCLK is generated from the same phase-locked loop (PLL) as the
data, and it travels to the PMA at the same rate as the data. By using one
of the data channels in the middle of the bus as the clock (in this case, the
eighth channel CH8), the clock-to-data skew improves.
Figure 8–6. Stratix & Stratix GX Device XSBI Transmitter Implementation
Stratix & Stratix GX
PCS Transmitter
Stratix & Stratix GX SERDES
Parallel
Register
4 or 8
Parallel-to-Serial
Register
CH0
TX_D[0]
622 Mbps
Stratix & Stratix GX
Logic Array
4 or 8
÷J
Fast PLL
×W
622 MHz
CH7
TX_D[7]
CH8
PMA_TXCLK
CH9
TX_D[8]
CH16
TX_D[15]
PMA
Transmitter
W=1
J = 4 or 8
PMA_TXCLK_SRC
622 MHz
Figure 8–7 shows the receiver input of the XSBI core. From the receiver
side, data (RX_D[15..0]) comes from the PMA to the Stratix or
Stratix GX receiver SERDES block along with the PMA receiver clock
(PMA_RXCLK). The PMA receiver clock is used to convert the serial data
to parallel data. The phase shift or inversion on the PMA receiver clock is
needed to capture the receiver data.
8–8
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Implementing 10-Gigabit Ethernet Using Stratix & Stratix GX Devices
Stratix and Stratix GX devices contain up to eight fast PLLs. These PLLs
provide high-speed outputs for high-speed differential I/O support as
well as general- purpose clocking with multiplication and phase shifting.
The fast PLL incorporates this 180° phase shift. The Stratix and Stratix GX
device’s data realignment feature enables you to save more logic
elements (LEs). This feature provides a byte-alignment capability, which
is embedded inside the SERDES. The data realignment circuitry can
correct for bit misalignments by slipping data bits.
f
For more information about fast PLLs, see the Stratix Device Family Data
Sheet section of the Stratix Device Handbook, Volume 1 or the Stratix GX
Device Family Data Sheet section of the Stratix GX Device Handbook,
Volume 1.
Figure 8–7. Stratix & Stratix GX Device XSBI Receiver Implementation
Stratix & Stratix GX PCS Receiver
Stratix & Stratix GX SERDES
Parallel
Register
4 or 8
Parallel-to-Serial
Register
CH0
RX_D[0]
622 Mbps
Stratix & Stratix GX
Logic Array
PMA
Receiver
4 or 8
CH15
÷J
Fast PLL
×W
622 MHz
RX_D[15]
W=1
J = 4 or 8
PMA_RXCLK_SRC
622 MHz
Altera Corporation
July 2005
8–9
Stratix Device Handbook, Volume 2
Interfaces
With this XSBI transmitter and receiver block implementation, each XSBI
core requires two fast PLLs. The potential number of XSBI cores per
device corresponds to the number of fast PLLs each Stratix or Stratix GX
device contains. Tables 8–2 and 8–3 show the number of LVDS channels,
the number of fast PLLs, and the number of XSBI cores that can be
supported for each Stratix or Stratix GX device.
Table 8–2. Stratix Device XSBI Core Support
Stratix Device
Number of LVDS
Channels
Number of Fast
(Receive/Transmit)
PLLs
(1)
Number of XSBI
Interfaces
(Maximum)
EP1S10
44/44
4
2
EP1S20
66/66
4
2
EP1S25
78/78
4
2
EP1S30
82/82
8
4
EP1S40
90/90
8
4
EP1S60
116/116
8
4
EP1S80
152/156
8
4
Note to Table 8–2:
(1)
The LVDS channels can go up to 840 Mbps for flip-chip packages and up to
624 Mbps for wire-bond packages. This number includes both high speed and
low speed channels. The high speed LVDS channels can go up to 840 Mbps. The
low speed LVDS channels can go up to 462 Mbps. The High-Speed Differential I/O
Support chapter in the Stratix Device Handbook, Volume 1, and the device pin-outs
on the web (www.altera.com) specify which channels are high and low speed.
8–10
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Implementing 10-Gigabit Ethernet Using Stratix & Stratix GX Devices
Table 8–3. Stratix GX Device XSBI Core Support
Number of LVDS
Channels
Number of Fast
(Receive/Transmit)
PLLs
(1)
Stratix GX Device
Number of XSBI
Interfaces
(Maximum)
EP1SGX10
22/22
2
1
EP1SGX25
39/39
2
2
EP1SGX40
45/45
4
2
Note to Table 8–3:
(1)
The LVDS channels can go up to 840 Mbps for flip-chip packages and up to
624 Mbps for wire-bond packages. This number includes both high speed and
low speed channels. The high speed LVDS channels can go up to 840 Mbps. The
low speed LVDS channels can go up to 462 Mbps. The High-Speed Differential I/O
Support chapter in the Stratix Device Handbook, Volume 1, and the device pin-outs
on the web (www.altera.com) specify which channels are high and low speed.
AC Timing Specifications
Stratix and Stratix GX devices support a PCS interface. Figures 8–8 and
8–9 and Tables 8–4 and 8–5 illustrate timing characteristics of the PCS
transmitter and receiver interfaces.
Figure 8–8 shows the AC timing diagram for the Stratix and Stratix GX
PCS transmitter. You can determine PCS channel-to-channel skew by
adding the data invalid window before the rising edge (Tcq_pre) to the data
invalid window after the rising edge (Tcq_post).
Figure 8–8. PCS Transmitter Timing Diagram
Tperiod
PMA_TX_CLK
TX_DATA[15..0]
Tcq_pre
Altera Corporation
July 2005
Valid
Data
Tcq_post
Tsetup
Thold
8–11
Stratix Device Handbook, Volume 2
Interfaces
Table 8–4 lists the AC timing specifications for the PCS transmitter.
Table 8–4. PCS Transmitter Timing Specifications
Value
Parameter
Unit
Min
Typ
Max
PMA_TX_CLK Tperiod (WAN)
1,608
ps
PMA_TX_CLK Tperiod (LAN)
1,552
ps
Data invalid window before the rising edge
(Tcq_pre)
200
ps
Data invalid window after the rising edge (Tcq_post)
200
ps
40
PMA_TX_CLK duty cycle
PCS transmitter channel-to-channel skew
60
%
200
ps
Figure 8–9 shows the AC timing diagram for the Stratix and Stratix GX
PCS receiver interface. You can determine the PCS sampling window by
adding Tsetup to Thold. Receiver skew margin (RSKM) refers to the amount
of skew tolerated on the printed circuit board (PCB).
Figure 8–9. PCS Receiver Timing Diagram
Tperiod
Tperiod
PMA_RX_CLK
RX_DATA[15..0]
Tcq_pre
PMA_RX_CLK
Valid
Data
RX_DATA[15..0]
Tcq_post
Tsetup
Thold
RSKM
Transmitter Channel-to-Channel
Skew/2
RSKM
Sampling Window
Transmitter Channel-to-Channel
Skew/2
Table 8–5 lists the AC timing specifications for the PCS receiver interface.
Table 8–5. PCS Receiver Timing Specifications (Part 1 of 2)
Value
Parameter
Unit
Min
Typ
Max
PMA_RX_CLK Tperiod (WAN)
1,608
ps
PMA_RX_CLK Tperiod (LAN)
1,552
ps
Data invalid window before the rising edge (Tcq_pre)
200
ps
Data invalid window after the rising edge (Tcq_post)
200
ps
8–12
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Implementing 10-Gigabit Ethernet Using Stratix & Stratix GX Devices
Table 8–5. PCS Receiver Timing Specifications (Part 2 of 2)
Value
Parameter
Unit
Min
Typ
Max
PMA_RX_CLK duty cycle
45
Data set-up time (Tsetup)
300
55
ps
%
Data hold time (Thold)
300
ps
PCS sampling window
600
ps
RSKM (WAN)
304
ps
RSKM (LAN)
276
ps
XGMII
The purpose of XGMII is to provide a simple, inexpensive, and easy to
implement interconnection between the MAC sublayer and the PHY.
Though XGMII is an optional interface, it is used extensively in the
10-Gigabit Ethernet standard as the basis for the specification. The
conversion between the parallel data paths of XGMII and the serial MAC
data stream is carried out by the reconciliation sublayer. The
reconciliation sublayer maps the signal set provided at the XGMII to the
physical layer signaling (PLS) service primitives provided at the MAC.
XGMII supports a 10-Gbps MAC data rate.
Functional Description
The XGMII is composed of independent transmit and receive paths. Each
direction uses 32 data signals, TXD[31..0] and RXD[31..0], 4 control
signals, TXC[3..0] and RXC[3..0], and a clock TX_CLK and RX_CLK.
Figure 8–10 shows the XGMII functional block diagram.
Altera Corporation
July 2005
8–13
Stratix Device Handbook, Volume 2
Interfaces
Figure 8–10. XGMII Functional Block Diagram
TXC[3..0]
XGMII
TXD[31..0]
TX_CLK
RXC[3..0]
RXD[31..0]
RX_CLK
PCS
PCS
Transmit
tx_data[15..0]
PCS
Receive
XSBI
rx_data[15..0]
PMA
The 32 TXD and four TXC signals as well as the 32 RXD and four RXC
signals are organized into four data lanes. The four lanes in each direction
share a common clock (TX_CLK for transmit and RX_CLK for receive). The
four lanes are used in round-robin sequence to carry an octet stream
(8 bits of data per lane). The reconciliation sublayer generates continuous
data or control characters on the transmit path and expects continuous
data or control characters on the receive path.
Implementation
XGMII uses the 1.5-V HSTL I/O standard. Stratix and Stratix GX devices
support the 1.5-V HSTL Class I and Class II I/O standard (EIA/JESD8-6).
The standard requires a differential input with an external reference
voltage (VREF) of 0.75 V, as well as a termination voltage VTT of 0.75 V, to
which termination resistors are connected. The HSTL Class I standard
requires a 1.5-V VCCIO voltage, which is supported by Stratix and
Stratix GX devices.
Figure 8–11 shows the 32-bit full-duplex 1.5-V HSTL implementation of
XGMII.
8–14
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Implementing 10-Gigabit Ethernet Using Stratix & Stratix GX Devices
Figure 8–11. Stratix & Stratix GX XGMII Implementation
Stratix & Stratix GX PCS
DDR Output Circuitry
MAC (RS)
RX_D[31..0]
Data
Shift
Register
RX_C[3..0]
Clk
MAC_RXCLK
PLL1
Stratix &
Stratix GX
Logic Array
×4
Receiver
Transmitter
Receiver
PLL2
÷4
MAC_TXCLK
TX_D[31..0]
Clk
Shift
Register
Data
DDR Input Circuitry
TX_C[3..0]
Transmitter
For this implementation, the shift register clocks can either be generated
from a divided down MAC reconciliation sublayer transmitter clock
(MAC_TXCLK), or the asynchronous core clock, or both if using a FIFO
buffer.
Figure 8–12 shows one channel of the output half of XGMII. Data that is
transmitted from the PCS to the MAC reconciliation sublayer starts at the
core of the Stratix or Stratix GX device and travels to the shift register. The
shift register takes in the parallel data (even bits sent to the top register
and odd bits sent to the bottom register) and serializes the data. After the
data is serialized, it travels to the double data rate (DDR) output circuitry,
which is clocked with the × 4 clock from the PLL. Out of the DDR output
circuitry, the data drives off-chip along with the × 4 clock. This transaction
creates the DDR relationship between the clock and the data output. This
implementation only shows one channel, but can be duplicated to include
all 32 bits of the RX_D signal and all 4 bits of the RX_C signal.
Altera Corporation
July 2005
8–15
Stratix Device Handbook, Volume 2
Interfaces
Figure 8–12. Stratix & Stratix GX XGMII Output Implementation (One Channel)
Stratix & Stratix GX PCS Output
DDR Output Circuitry
D0,D2,D4,D6
4
Shift
Register
8
DFF
RX_D[0]
DATA
DATA
312.5 Mbps
Stratix &
Stratix GX
Logic Array
4
Shift
Register
D1,D3,D5,D7
DFF
MAC
Receiver
39.0625 MHz
CLK
PLL
×4
156.25 MHz
MAC_RXCLK
CLK
156.25 MHz
Figure 8–13 shows one channel of the input half of the XGMII interface.
From the receiver side, the DDR data is captured from the MAC to the
Stratix and Stratix GX PCS DDR input circuitry. The serial data is
separated into two individual data streams with the even bits routed to
the top register and odd bits routed to the bottom register. The DDR input
circuitry produces two output data streams that go into the shift registers.
From the shift registers, the data is deserialized using the clock from the
MAC, combining into an 8-bit word. This parallel data goes to a register
that is clocked by the divide-by-4 clock from the PLL. This data and clock
go to the Stratix and Stratix GX core. This implementation shows only one
channel, but can be duplicated to include all 32 bits of the TX_D signal and
all 4 bits of the TX_C signal.
8–16
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Implementing 10-Gigabit Ethernet Using Stratix & Stratix GX Devices
Figure 8–13. Stratix & Stratix GX XGMII Input Implementation (One Channel)
Stratix & Stratix GX PCS Input
DDR Input Circuitry
D0,D2,D4,D6
TX_D[0]
DATA
4
DFF
312.5 Mbps
Shift
Register
8
DFF
D1,D3,D5,D7
DFF
8
DATA
4
Latch
MAC
Transmitter
Shift
Register
Stratix &
Stratix GX
Logic Array
156.25 MHz
MAC_TXCLK
CLK
156.25 MHz
PLL
÷4
39.0625 MHz
CLK
Stratix and Stratix GX devices contain up to four enhanced PLLs. These
PLLs provide features such as clock switchover, spread-spectrum
clocking, programmable bandwidth, phase and delay control, and PLL
reconfiguration. Since the maximum clock rate is 156.25 MHz, you can
use a fast or enhanced PLL for both the XGMII output and input blocks.
f
For more information about fast PLLs, see the Stratix Device Family Data
Sheet section of the Stratix Device Handbook, Volume 1 or the Stratix GX
Device Family Data Sheet section of the Stratix GX Device Handbook,
Volume 1.
With this implementation for the XGMII output and input blocks, the
number of XGMII cores per device corresponds to the number of PLLs
each Stratix and Stratix GX device contains. Tables 8–6 and 8–7 show the
number of 1.5-V HSTL I/O pins, PLLs, and XGMII cores that are
supported in each Stratix and Stratix GX device. Each core requires 72 1.5-
Altera Corporation
July 2005
8–17
Stratix Device Handbook, Volume 2
Interfaces
V HSTL I/O pins for data and control and 2 clock pins for the transmitter
and receiver clocks. Each XGMII core also needs two PLLs (one for each
direction).
Table 8–6. Stratix XGMII Core Support
Number of 1.5-V
HSTL Class I I/O
Pins
Number of Fast
& Enhanced
PLLs
Number of XGMII
Interfaces
EP1S10
410
6
3
EP1S20
570
6
3
Stratix Device
EP1S25
690
6
3
EP1S30
718
10
5
EP1S40
814
12
6
EP1S60
1,014
12
6
EP1S80
1,195
12
6
Number of 1.5-V
HSTL Class I I/O
Pins
Number of Fast
& Enhanced
PLLs
Number of XGMII
Interfaces
226
4
2
Table 8–7. Stratix GX XGMII Core Support
Stratix Device
EP1SGX10 C, D
EP1SGX25 C
253
4
2
EP1SGX25 D, F
370
4
2
EP1SGX40 D, G
430
8
4
Reduced System Noise
The output buffer of each Stratix and Stratix GX device I/O pin has a
programmable drive strength control for certain I/O standards. The 1.5V HSTL Class I standard supports the minimum setting, which is the
lowest drive strength that guarantees IOH and IOL of the standard. Using
minimum settings provides signal slew rate control to reduce system
noise and signal overshoot.
f
For more information on IOH and IOL values, see Operating Conditions in
the DC & Switching Characteristics chapter of the Stratix Device Handbook,
Volume 1 or Operating Conditions in the DC & Switching Characteristics
chapter of the Stratix GX Device Handbook, Volume 1.
8–18
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Implementing 10-Gigabit Ethernet Using Stratix & Stratix GX Devices
Timing
XGMII signals must meet the timing requirements shown in Figure 8–14.
Make all XGMII timing measurements at the driver output (shown in
Figure 8–14) and a capacitive load from all sources of 20 pF that are
specified relative to the VIL_AC(max) and VIH_AC(min) thresholds.
Figure 8–14. XGMII Timing Diagram
TX_CLK
VIH_AC(min)
RX_CLK
VIL_AC(max)
VIH_AC(min)
TXC, TXD,
RXC, RXD
VIL_AC(max)
tsetup
tsetup
thold
thold
Table 8–8 shows the XGMII timing specifications.
Table 8–8. XGMII Timing Specifications Note (1)
Symbol
Driver
Receiver
Unit
Tsetup
960
480
ps
Thold
960
480
ps
Note to Table 8–8:
(1)
The actual set-up and hold times will be made available after device
characterization is complete.
Stratix and Stratix GX devices support DDR data with clock rates of up to
200 MHz, well above the XGMII clock rate of 156.25 MHz. For the HSTL
Class I I/O standard, Stratix and Stratix GX device I/O drivers provide a
1.0-V/ns slew rate at the input buffer of the receiving device.
XAUI
XAUI (pronounced Zowie) is located between the XGMII at the
reconciliation sublayer and the XGMII at the PHY layer. Figure 8–15
shows the location of XAUI. XAUI is designed to either extend or replace
XGMII in chip-to-chip applications of most Ethernet MAC to PHY
interconnects.
Altera Corporation
July 2005
8–19
Stratix Device Handbook, Volume 2
Interfaces
Figure 8–15. XAUI Location
MAC
Reconciliation
XGMII
XGMII Extender
Sublayer (XGXS)
XAUI
XGXS
XGMII
PHY
Functional Description
XAUI can replace the 32 bits of parallel data required by XGMII for
transmission with just 4 lanes of serial data. XAUI uses clock data
recovery (CDR) to eliminate the need for separate clock signals. 8b/10b
encoding is employed on the data stream to embed the clock in the data.
The 8b/10b protocol to encode an 8-bit word stream to 10-bit codes that
results in a DC-balanced serial stream and eases the receiver
synchronization. To support 10-Gigabit Ethernet, each lane must run at a
speed of at least 2.5 Gbps. Using 8b/10b encoding increases the rate for
each lane to 3.125 Gbps, which will be supported in Stratix GX Gbps
devices. This circuitry is supported by the embedded 3.125 Gbps
transceivers within the Stratix GX architecture. You can find more
8–20
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Implementing 10-Gigabit Ethernet Using Stratix & Stratix GX Devices
information on XAUI support in Section II, Stratix GX Transceiver User
Guide of the Stratix GX Device Handbook, Volume 2. Figure 8–16 shows how
XAUI is implemented.
Figure 8–16. Stratix GX XAUI Implementation
Stratix GX XAUI
PCS
8
CDR Tx
Stratix GX
Logic Array
8
CDR Tx
CH0
TX_D[0]
CH3
TX_D[3]
Transmitter
8
CDR Rx
8
CDR Rx
CH0
CH3
RX_D[0]
RX_D[3]
3.125 Gbps
Receiver
I/O
Characteristics
for XSBI, XGMII
& XAUI
The three interfaces of 10-Gigabit Ethernet (XSBI, XGMII, and XAUI) each
have different rates and I/O standards. Table 8–9 shows the
characteristics for each interface.
Table 8–9. 10-Gigabit Ethernet Interfaces Characteristics
Interface
Altera Corporation
July 2005
Width
Clock Rate
(MHz)
Data Rate
Per Channel
Clocking
Scheme
I/O Type
XGMII
32
156.25
312.5 Mbps
DDR source 1.5-V
synchronous HSTL
XSBI
16
644.5 or
622.08
644.5 or
622.08
Mbps
SDR source LVDS
synchronous
XAUI
4
None
3.125 Gbps
Clock data
recovery
(CDR)
1.5-V
PCML
8–21
Stratix Device Handbook, Volume 2
I/O Characteristics for XSBI, XGMII & XAUI
Software Implementation
You can use the Assignment Organizer in the Altera® Quartus® II
software to implement the I/O standards for a particular interface. For
example, set the I/O standard to LVDS for XSBI and to HSTL Class I for
XGMII. You can use the MegaWizard® Plug-In Manager to create the
PLLs and transmitter and receiver SERDES blocks for the XSBI
implementation and PLLs and DDR input and output circuitry for the
XGMII implementation. For more information on the Assignment
Organizer or MegaWizard Plug-In Manager, see the Quartus II Software
Help.
AC/DC Specifications
Table 8–10 lists the XSBI DC electrical characteristics, similar to Stratix
and Stratix GX devices, that are based on the ANSI/TIA-644 LVDS
specification.
Table 8–10. XSBI DC Specifications
Value
Parameter
Unit
Min
Output differential voltage (VOD)
Output offset voltage (VOS)
Output Impedance, single ended
Typ
Max
250
400 (1)
mV
1,125
1,375
mV
40
140
W
Change in VOD between ‘0’ and ‘1’
50
mV
Change in VOS between ‘0’ and ‘1’
50
mV
1,600
mV
Input voltage range (VI)
900
Differential impedance
100
W
Input differential voltage (VID)
100
600
mV
Receiver differential input impedance
70
130
W
50
mV
100
400
ps
Ground potential difference (between PCS and PMA)
Rise and fall times (20% to 80%)
Note to Table 8–10:
(1)
Larger VOD is possible for better signal intensity.
I/O characteristics for the 1.5-V HSTL standard for Stratix and Stratix GX
devices are shown in Figure 8–17 and comply with XGMII electrical
specifications available in 10-Gigabit Ethernet draft IEEE P802.3ae.
8–22
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Implementing 10-Gigabit Ethernet Using Stratix & Stratix GX Devices
Figure 8–17. Electrical Characteristics for Stratix & Stratix GX Devices
(1.5-V HSTL Class I)
tf(min) = 1 V/ns
tz(min) = 1 V/ns
80% VSWING
VREF
VSWING = 1.0 V
Input
20% VSWING
tPD
VOH = VCCN − 0.4 V = 1.1 V
Output
VTT = VCCN/2 = 0.75 V
VOL = 0.4 V
VIH(AC) = 0.95 V
tPH2
Tri-Stated
Output
VTT = 0.75 V
VIL(AC) = 0.55 V
tPL2
HSTL AC Load Circuit for Class I
VTT
Output Buffer
RT = 50 Ω
VOUT
CL = 20pF
VIN
Input Buffer
VREF
HSTL AC Waveform & I/O Interface
Altera Corporation
July 2005
8–23
Stratix Device Handbook, Volume 2
I/O Characteristics for XSBI, XGMII & XAUI
Table 8–11 lists the DC specifications for Stratix and Stratix GX devices
(1.5-V HSTL Class I).
Table 8–11. DC Specifications for Stratix & Stratix GX Devices (1.5-V HSTL Class I) Note (1)
Minimum
Typical
Maximum
Units
VCCIO
Symbol
I/O supply voltage
Parameter
Conditions
1.4
1.5
1.6
V
VREF
Input reference voltage
0.68
0.75
0.9
V
VTT
Termination voltage
0.7
0.75
0.8
V
VIH (DC)
DC high-level input voltage
VREF + 0.1
VIL (DC)
DC low-level input voltage
–0.3
VIH (AC)
AC high-level input voltage
VREF + 0.2
VIL (AC)
AC low-level input voltage
II
Input pin leakage current
0 < VIN < VCCIO
–10
VOH
High-level output voltage
IOH = –8 mA
VCCIO – 0.4
VOL
Low-level output voltage
IOL = 8 mA
IO
Output leakage current
(when output is high Z)
GND ≤VOUT ≤
VCCIO
–10
V
VREF – 0.1
V
V
VREF – 0.2
V
10
μA
V
0.4
V
10
μA
Note to Table 8–11:
(1)
Drive strength is programmable according to values shown in the Stratix Device Family Data Sheet section of the
Stratix Device Handbook, Volume 1 or the Stratix GX Device Family Data Sheet section of the Stratix GX Device
Handbook, Volume 1.
10-Gigabit Ethernet MAC Core
As an Altera Megafunction Partners Program (AMPPSM) member,
MorethanIP provides a 10-Gigabit Ethernet MAC core for Altera
customers. MorethanIP’s 10-Gigabit Ethernet MAC core implements the
RS, the MAC layer, and user-programmable FIFO buffers for clock and
data decoupling.
Core Features
MorethanIP’s 10-Gigabit Ethernet MAC core provides the following
features:
■
■
Includes automatic pause frame generation (per IEEE 802.3 × 31) with
user-programmable pause quanta and pause-frame termination
Includes a programmable 48-bit MAC address with a promiscuous
mode option, and a programmable Ethernet frame length that
supports IEEE 802.1Q VLAN-tagged frames or jumbo Ethernet
frames
8–24
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Implementing 10-Gigabit Ethernet Using Stratix & Stratix GX Devices
■
■
■
Supports broadcast traffic and multi-cast address resolution with a
64-entry hash table
Compliant with the IEEE802.3ae Draft 4.0
Implements XGMII, allowing it to interface to XAUI through a
10-Gigabit commercial SERDES
Conclusion
10-Gigabit Ethernet takes advantage of the existing Gigabit Ethernet
standard. With their rich I/O features, Stratix and Stratix GX devices
support the components of 10-Gigabit Ethernet as well as XSBI and
XGMII. Stratix GX devices also support XAUI. These interfaces are easily
implemented using the core architecture, differential I/O capabilities,
and superior PLLs of Stratix and Stratix GX devices.
Altera Corporation
July 2005
8–25
Stratix Device Handbook, Volume 2
I/O Characteristics for XSBI, XGMII & XAUI
8–26
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
9. Implementing SFI-4 in
Stratix & Stratix GX Devices
S52011-2.0
Introduction
The growth of the Internet has created huge bandwidth demands as
voice, video, and data push the limits of the existing wide area network
(WAN) backbones. To facilitate this bandwidth growth, speeds of OC-192
and higher are being deployed in WAN backbones (see Figure 9–1).
Today’s carrier backbone networks are supported by SONET/SDH
transmission technology. SONET/SDH is a transmission technology for
transporting optical signals at speeds ranging from 51 megabits per
second (Mbps) up to 40 gigabits per second (Gbps). SONET/SDH rings
make up the majority of the existing backbone infrastructure of the
Internet and the public switched telephone network (PSTN).
The Optical Internetworking Forum (OIF) standard SFI-4 is a 16-bit LVDS
interface used in an OC-192 SONET system to link the framer and the
serializer/deserializer (SERDES). Stratix® and Stratix GX devices support
the required data rates of up to 622.08 Mbps along with the one-to-one
relationship required between clock frequency and data rate. The fast
phase-locked loop (PLL) was designed to support the high clock
frequencies and the one-to-one relationship (between clock and data rate)
needed for interfaces such as XSBI and SFI-4. Support for SFI-4 extends
the reach of high-density programmable logic from the backplane to the
physical layer (PHY) devices.
This chapter focuses on the implementation of the interface between the
SERDES and the framer.
Figure 9–1. WAN Backbone
DWDM
40 G
SONET OC-48
SONET OC-192
~ ~
SDH STM-64
A SONET/SDH transmission network is composed of several pieces of
equipment, including terminal multiplexers, add-drop multiplexers, and
repeater and digital cross-connect systems. SONET is the standard used
in North America and SDH is the standard used outside North America.
Altera Corporation
July 2005
9–1
Introduction
The SONET/SDH specification outlines the frame format, multiplexing
method, synchronization method, and optical interface between the
equipment, as well as the specific optical interface.
SONET/SDH continues to play a key role in the next generation of
networks for many carriers. In the core network, the carriers offer services
such as telephone, dedicated leased lines, and Internet protocol (IP) data,
which are continuously transmitted. The individual data channels are not
transmitted on separate lines; instead, they are multiplexed into higher
speeds and transmitted on SONET/SDH networks at the corresponding
transmission speed.
Figure 9–2 shows a typical SONET/SDH line card. The system operates
as follows:
1.
The SONET/SDH line card first takes a high-speed serial optical
signal and converts it into a high-speed serial electrical signal. The
devices are called physical media dependent (PMD) devices.
2.
The system then recovers the clock from the electrical data using a
clock data recovery (CDR) unit.
3.
The SERDES parallelizes the data so that it can be manipulated
easily at lower clock rates.
4.
The interface between the SERDES and framer is called the SERDES
framer interface. The interface requirements are defined by the OIF.
5.
The framer identifies the beginning of the SONET/SDH frames and
monitors the performance of the system.
6.
The mapper following the framer maps asynchronous transfer
mode (ATM) cells, IP packets, or T/E carrier signals into the SONET
frame.
7.
The PHY-link layer interface provides a bus interface to packet/cell
processors or other link-layer devices.
9–2
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Implementing SFI-4 in Stratix & Stratix GX Devices
Figure 9–2. SONET/SDH Line Card
7
1
Optical
Signal
OE Module
2
CDR
3
5
6
SERDES
SONET/SDH
Framer
SONET/SDH
Mapper/Protocol
Processor
Optical-Electrical
Conversion
SERDES Framer
Interface
4
To Packet
Processor &
Switch Fabric
Link Layer
Interface
The OIF has defined the electrical interface (SFI) between the
SONET/SDH framer and high-speed SERDES devices. To keep up with
evolving transmission speeds and technology enhancements, different
versions of electrical interfaces are defined. SFI-4 is the version of SFI that
acts as an interface between an OC-192 SERDES and SONET framer, as
shown in Figure 9–2. An aggregate of 9953.28 Mbps is transferred in each
direction. With their differential I/O capabilities, Stratix and Stratix GX
devices are ideally suited to support the framer side of the SFI-4 interface.
Support for SFI-4 extends the reach of high-density programmable logic
from the backplane to the PHY devices.
System Topology
The SFI-4 interface uses 16 channels of source-synchronous LVDS to
interface between a SONET framer and an OC-192 SERDES. Figure 9–3
shows the SFI-4 interface.
Altera Corporation
July 2005
9–3
Stratix Device Handbook, Volume 2
Introduction
Figure 9–3. SFI-4 Interface Signals
REFCLK
SONET Framer
Transmitter
OC-192 SERDES
Transmitter
TXDATA[15..0]
TXCLK
TXCLK_SRC
SONET Framer
Receiver
Recovered Clock
Receiver
RXDATA[15..0]
RXCLK
The framer transmits outbound data via TXDATA[15..0] and is
received at the SERDES using TXCLK. TXCLK is derived from
TXCLK_SRC, which is provided by the OC-192 SERDES. The framer
receives incoming data on RXDATA[15..0] from the OC-192 SERDES.
The data received is latched on the rising edge of RXCLK. Table 9–1
provides the data rates and clock frequencies specified by SFI-4. The
modes of TXCLK are specified by the SFI-4 standard. In required mode
(622 MHz clock mode or × 1 mode), TXCLK should run at 622.08 MHz. In
optional mode (311 MHz clock mode or × 2 mode), TXCLK should run at
311.04 MHz.
Table 9–1. SFI-4 Interface Data Rates & Clock Frequencies
Signal
Performance
TXDATA[15..0]
622.08 Mbps
TXCLK
622.08 MHz or 311.04 MHz
TXCLK_SRC
622.08 MHz
RXDATA[15..0]
622.08 Mbps
RXCLK
622.08 MHz
REFCLK
622.08 MHz
9–4
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Implementing SFI-4 in Stratix & Stratix GX Devices
Interface Implementation in Stratix & Stratix GX Devices
The 16-bit full-duplex LVDS implementation of the framer part of the
SFI-4 interface is shown in Figure 9–4. Stratix devices support sourcesynchronous interfacing and LVDS differential signaling up to 840 Mbps.
Stratix devices have embedded SERDES circuitry for serial and parallel
data conversion.
The source-synchronous I/O implemented in Stratix GX devices
optionally includes dynamic phase alignment (DPA). DPA automatically
and continuously tracks fluctuations caused by system variations and
self-adjusts to eliminate the phase skew between the multiplied clock and
the serial data, allowing for data rates of 1 Gbps. In non DPA mode the
I/O behaves similarly to that of the Stratix I/O. This document assumes
that DPA is disabled. However, it is simple to implement the same system
with DPA enabled to take advantage of its features. For more information
on DPA, see the Stratix GX Transceivers chapter in the Stratix GX Device
Handbook, Volume 1.
The fast PLL enables 622.08 Mbps data transmission by transmitting and
receiving a differential clock at rates of up to 645 MHz. The clocks
required in the SERDES for parallel and serial data conversion can be
configured from the received RXCLK (divided down), the TXCLK_SRC
(divided down), or the asynchronous core clock. See Figure 9–4.
Altera Corporation
July 2005
9–5
Stratix Device Handbook, Volume 2
Introduction
Figure 9–4. Implementation of SFI-4 Interface Using Stratix & Stratix GX Devices
REFCLK
OC-192
SERDES
128
Data
TXDATA[15..0]
Transmitter
SERDES
Clk
TXCLK
Stratix Framer
PLL1
Stratix &
Stratix GX
Logic Array
×1
TXCLK_SRC
÷8
Transmitter
Transmitter
Receiver
PLL2
Phase Shift
÷8
Clk
180˚
Receiver
SERDES
RXCLK
RXDATA[15..0]
Data
128
Receiver
f
For details on differential I/O buffers, SERDES, and clock dividers using
PLLs, see the High-Speed Differential I/O Interfaces in Stratix Devices
chapter in the Stratix Device Handbook or the Stratix GX Device Handbook.
Figure 9–5 shows the transmitter block (from Figure 9–4) of the SFI-4
framer interface implemented in Stratix and Stratix GX devices. The data
starts in the logic array and goes into the Stratix and Stratix GX SERDES
block. The transmitter SERDES of the framer converts the parallel data to
serial data for the 16 TXDATA channels (TXDATA[15..0]). A fast PLL is
used to generate TXCLK from TXCLK_SRC. The fast PLL keeps the
TXDATA and TXCLK edge-aligned. A divided down (÷8) clock generated
from TXCLK_SRC is used to convert the parallel data to serial in the
transmitter SERDES. The divided down clock also clocks some of the
logic in the logic array.
9–6
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Implementing SFI-4 in Stratix & Stratix GX Devices
Figure 9–5. Framer Transmitter Interface in Stratix & Stratix GX Devices
Stratix & Stratix GX SFI-4 Transmitter
Stratix & Stratix GX SERDES
Parallel
Register
8
Parallel-to-Serial
Register
CH0
Stratix & Stratix GX
Logic Array
TXDATA[0]
622 Mbps
OC-192
SERDES
8
CH15
TXDATA[15]
TXCLK
622MHz
÷J
Fast PLL
×W
622 MHz
W=1
J=8
TXCLK_SRC
622 MHz
Figure 9–6 shows the receiver block (from Figure 9–4) of the SFI-4 framer
interface implemented in Stratix and Stratix GX devices.
RXDATA[15..0] is received from the OC-192 SERDES on the differential
I/O pins of the Stratix or Stratix GX device. The receiver SERDES
converts the high-speed serial data to parallel. You can generate the
clocks required in the SERDES for parallel and serial data conversion
from the received RXCLK. RXCLK is inverted (phase-shifted by 180° ) to
capture received data. While normal I/O operation guarantees that data
is captured, it does not guarantee the parallelization boundary, which is
randomly determined based on the power up of both communicating
devices. The SERDES has embedded data realignment capability, which
can be used to save logic elements (LEs).
Altera Corporation
July 2005
9–7
Stratix Device Handbook, Volume 2
Introduction
Figure 9–6. Framer Receiver Interface in Stratix & Stratix GX Devices
Stratix & Stratix GX SFI-4 Receiver
Stratix & Stratix GX SERDES
Parallel
Register
8
Serial-to-Parallel
Register
CH0
RXDATA[0]
Stratix & Stratix GX
Logic Array
OC-192
SERDES
8
CH15
RXDATA[15]
622 Mbps
÷J
Fast PLL
×W
622 MHz
W=1
J=8
RXCLK
622 MHz
Note to Figure 9–6:
(1)
The figure shows Stratix GX DPA disabled.
f
For more information on the byte-alignment feature in Stratix and
Stratix GX devices, see the High-Speed Differential I/O Interfaces in Stratix
Devices chapter in the Stratix Device Handbook or the Stratix GX Device
Handbook.
9–8
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Implementing SFI-4 in Stratix & Stratix GX Devices
Tables 9–2 and 9–3 list the number of SFI-4 cores that can be implemented
in Stratix and Stratix GX devices. See the High-Speed Differential I/O
Interfaces in Stratix Devices chapter in the Stratix Device Handbook or the
Stratix GX Device Handbook for the package type and the maximum
number of channels supported by each package.
Table 9–2. Stratix SFI-4 Core Support
Number of LVDS
Channels
Stratix Device
(Receiver/Transmitter)
(1)
Number of PLLs
Number of SFI-4
Interfaces
(Maximum)
EP1S10
44/44
4
2
EP1S20
66/66
4
2
EP1S25
78/78
4
2
EP1S30
82/82
8
4
EP1S40
90/90
8
4
EP1S60
116/116
8
4
EP1S80
152/156
8
4
Note to Table 9–2:
(1)
The LVDS channels can go up to 840 Mbps (or 1 Gbps using DPA in Stratix GX
devices). This number includes both high speed and low speed channels. The
high speed LVDS channels can go up to 840 Mbps. The low speed LVDS channels
can go up to 462 Mbps. The High-Speed Differential I/O Support chapters in the
Stratix Device Handbook, Volume 1 and the Stratix GX Device Handbook, Volume 1
and the device pin-outs on the web (www.altera.com) specify which channels are
high and low speed.
Table 9–3. Stratix GX SFI-4 Core Support
Number of LVDS
Channels
(Receiver/Transmitter)
(1)
Number of PLLs
Number of SFI-4
Interfaces
(Maximum)
EP1SGX10
22/22
2
1
EP1SGX25
39/39
2
2
EP1SGX40
45/45
4
2
Stratix GX
Device
Note to Table 9–3:
(1)
Altera Corporation
July 2005
The LVDS channels can go up to 840 Mbps, or 1 Gbps using DPA. This number
includes both high speed and low speed channels. The high speed LVDS channels
can go up to 840 Mbps. The low speed LVDS channels can go up to 462 Mbps. The
High-Speed Differential I/O Support chapter in the Stratix Device Handbook, Volume
1 and the Stratix GX Device Handbook, Volume 1 and the device pin-outs on the web
(www.altera.com) specify which channels are high and low speed.
9–9
Stratix Device Handbook, Volume 2
Introduction
AC Timing Specifications
Figures 9–7 through 9–9 and Tables 9–4 through 9–6 illustrate the timing
characteristics of SFI-4 at the framer. Stratix and Stratix GX devices
support all the timing requirements needed to support transmitter and
receiver functions of a SFI-4 framer; only framer-related timing
specifications are applicable.
f
For details on the timing specifications of LVDS I/O standards in Stratix
and Stratix GX devices, see the Stratix Device Family Data Sheet section of
the Stratix Device Handbook, Volume 1 and the High-Speed Differential I/O
Interfaces in Stratix Devices chapter or the Stratix GX Device Family Data
Sheet section of the Stratix GX Device Handbook, Volume 1 and the HighSpeed Differential I/O Interfaces in Stratix Devices chapter
Figure 9–7 shows the timing diagram for the Stratix and Stratix GX
framer transmitter × 1 (622 MHz clock) mode.
Figure 9–7. Framer Transmitter × 1 (622 MHz Clock) Mode Timing Diagram
Tperiod
Valid
Data
TX_DATA[15..0]
Tcq_pre
Tcq_post
Tsetup
Thold
Table 9–4 lists the timing specifications for the SFI-4 framer transmitter in
× 1 (622 MHz clock) mode.
Table 9–4. SFI-4 Framer Transmitter × 1 (622 MHz Clock) Mode Timing Specifications
Value
Parameter
Unit
Min
Typ
Max
1,608
TX_CLK (Tperiod)
ps
Data invalid window before the rising edge (Tcq_pre)
200
ps
Data invalid window after the rising edge (Tcq_post)
200
ps
TX_CLK duty cycle
Framer transmitter channel-to-channel skew
9–10
Stratix Device Handbook, Volume 2
40
60
%
200
ps
Altera Corporation
July 2005
Implementing SFI-4 in Stratix & Stratix GX Devices
Figure 9–8 shows the timing diagram for the SFI-4 framer transmitter in
× 2 (311 MHz clock) mode
Figure 9–8. Framer Transmitter × 2 (311 MHz Clock) Mode Timing Diagram
Tperiod/2
TX_CLK(P)
Valid
Data
TX_DATA[15..0]
Tcq_pre
Valid
Data
Tcq_post
Table 9–5 lists the timing specifications for the SFI-4 framer transmitter in
× 2 (311 MHz clock) mode.
Table 9–5. SFI-4 Framer Transmitter × 2 (311 MHz Clock) Mode Timing Specifications
Value
Parameter
Unit
Min
Typ
Max
3,215
TX_CLK (Tperiod)
ps
Data invalid window before the rising edge (Tcq_pre)
200
ps
Data invalid window after the rising edge (Tcq_post)
200
ps
52
%
200
ps
48
TX_CLK duty cycle
Framer transmitter channel-to-channel skew
Figure 9–9 shows the timing diagram for the SFI-4 framer receiver.
Figure 9–9. Framer Receiver Timing Diagram
Tperiod
Tperiod
RX_CLK(P)
RX_CLK(P)
RX_DATA[15..0]
Tcq_pre
Altera Corporation
July 2005
Valid
Data
Tcq_post
RX_DATA[15..0]
Tsetup
Thold
Transmitter Channel-to-Channel
Skew/2
RSKM
Sampling Window
RSKM
Transmitter Channel-to-Channel
Skew/2
9–11
Stratix Device Handbook, Volume 2
Introduction
Table 9–6 lists the timing specifications for the SFI-4 framer receiver.
Table 9–6. Framer Receiver Timing Specifications
Value
Parameter
Unit
Min
Typ
Max
1,608
RX_CLK (Tperiod)
Data invalid window before the rising edge (Tcq_pre)
Data invalid window after the rising edge (Tcq_post)
RX_CLK duty cycle
45
Data set-up time (Tsetup)
300
ps
200
ps
200
ps
55
%
ps
Data hold time (Thold)
300
ps
Framer sampling window
600
ps
Receiver skew margin (RSKM)
304
ps
Electrical Specifications
SFI-4 uses LVDS as a high-speed data transfer mechanism to implement
the SFI-4 interface. Table 9–7 lists the DC electrical characteristics for the
interface, which are based on the IEEE Std. 1596.3-1996 7 specification.
For more information on the voltage specification of LVDS I/O standards
in Stratix and Stratix GX devices, see the Stratix Device Family Data Sheet
section of the Stratix Device Handbook, Volume 1 and the High-Speed
Differential I/O Interfaces in Stratix Devices chapter or the Stratix GX Device
Family Data Sheet section of the Stratix GX Device Handbook, Volume 1 and
the High-Speed Differential I/O Interfaces in Stratix Devices chapter.
9–12
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Implementing SFI-4 in Stratix & Stratix GX Devices
Table 9–7. Framer LVDS DC Specifications
Value
Parameter
Unit
Min
Typ
Max
250
600 (1)
mV
1,125
1,375
mV
40
140
W
Change in VOD between ‘0’ and '1'
50
mV
Change in VOD between '1' and '0'
50
mV
2,400
mV
Output differential voltage (VOD)
Output offset voltage (VOS)
Output Impedance, single ended
Input voltage range (VI)
0
Differential impedance
100
W
Input differential voltage (VID)
100
600
mV
Receiver differential input impedance
70
130
W
50
mV
100
400
ps
Ground potential difference (between PCS and PMA)
Rise and fall times (20% to 80%)
Note to Table 9–7:
(1)
The IEEE standard requires 400 mV. A larger swing is encouraged, but not required.
Software Implementation
The SFI-4 interface uses a 16-bit LVDS I/O interface. The Altera®
Quartus® II software version 2.0 supports Stratix and Stratix GX devices,
allowing you to implement LVDS I/O buffers through the Quartus II
Assignment Organizer.
f
For information on the Quartus II Assignment Organizer, see the
Quartus II Software Help.
Conclusion
SFI-4 is the standard interface between SONET framers and optical
SERDES for OC-192 interfaces. With embedded SERDES and fast PLLs,
Stratix and Stratix GX devices can easily support the SFI-4 framer
interface, enabling
10-Gbps (OC-192) data transfer rates. Stratix and Stratix GX I/O supports
the required data rates of up to 622.08 Mbps. Stratix and Stratix GX fast
PLLs are designed to support the high clock frequencies and one-to-one
relationship needed for interfaces such as XSBI and SFI-4. Stratix and
Stratix GX devices can support multiple SFI-4 functions on one device.
Altera Corporation
July 2005
9–13
Stratix Device Handbook, Volume 2
Introduction
9–14
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
10. Transitioning APEX
Designs to Stratix &
Stratix GX Devices
S52012-3.0
Introduction
Stratix® and Stratix GX devices are Altera’s next-generation, system-ona-programmable-chip (SOPC) solution. Stratix and Stratix GX devices
simplify the block-based design methodology and bridge the gap
between system bandwidth requirements and programmable logic
performance.
This chapter highlights the new features in the Stratix and Stratix GX
devices and provides assistance when transitioning designs from
APEXTM II or APEX 20K devices to the Stratix or Stratix GX architecture.
You should be familiar with the APEX II or APEX 20K architecture and
available device features before using this chapter. Use this chapter in
conjunction with the Stratix Device Family Data Sheet section of the Stratix
Device Handbook, Volume 1 or the Stratix GX Device Family Data Sheet
section of the Stratix GX Device Handbook, Volume 1.
General
Architecture
Stratix and Stratix GX devices offer many new features and architectural
enhancements. Enhanced logic elements (LEs) and the MultiTrackTM
interconnect structure offer reduced resource utilization and
considerable design performance improvement. The MultiTrack
interconnect uses DirectDriveTM technology to ensure the availability of
deterministic routing resources for any design block, regardless of its
placement within the device.
All architectural changes between Stratix and Stratix GX and APEX II or
APEX 20K devices described in this section do not require any design
changes. However, you must resynthesize your design and recompile in
the Quartus® II software to target Stratix and Stratix GX devices.
Altera Corporation
July 2005
10–1
General Architecture
Logic Elements
Stratix and Stratix GX device LEs include several new, advanced features
that improve design performance and reduce logic resource consumption
(see Table 10–1). The Quartus II software automatically uses these new
LE features to improve device utilization.
Table 10–1. Stratix & Stratix GX LE Features
Feature
Function
Register chain interconnects Direct path between the register output
of an LE and the register input of an
adjacent LE within the same logic array
block (LAB)
Benefit
„ Conserves LE resources
„ Provides fast shift register
implementation
„ Saves local interconnect routing
resources within an LAB
Look-up table (LUT) chain
interconnects
Direct path between the combinatorial „ Allows LUTs within the same LAB to
output of an LE and the fast LUT input cascade together for high-speed wide
of an adjacent LE within the same LAB fan-in functions, such as wide XOR
operations
„ Bypasses local interconnect for
faster performance
Register-to-LUT feedback
path
Allows the register output to feed back
into the LUT of the same LE, such that
the register is packed with its own fanout LUT
„ Enhanced register packing mode
„ Uses resources more efficiently
Dynamic arithmetic mode
Uses one set of LEs for implementing
both an adder and subtractor
„ Improves performance for functions
that switch between addition and
subtraction frequently, such as
correlators
Carry-select chain
Calculates outputs for a possible carry- „ Gives immediate access to result for
in of 1 or 0 in parallel
both a carry-in of 1 or 0
„ Increases speed of carry functions
for high-speed operations, such as
counters, adders, and comparators
Asynchronous clear and
asynchronous preset
function
Supports direct asynchronous clear
and preset functions
„ Conserves LE resources
„ Does not require additional logic
resources to implement NOT-gate
push-back
In addition to the new LE features described in Table 10–1, there are
enhancements to the chains that connect LEs together. Carry chains are
implemented vertically in Stratix and Stratix GX devices, instead of
horizontally as in APEX II and APEX 20K devices, and continue across
rows, instead of across columns, as shown in Figure 10–1. Also note that
the Stratix and Stratix GX architectures do not support the cascade
primitive. Therefore, the Quartus II Compiler automatically converts
10–2
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Transitioning APEX Designs to Stratix & Stratix GX Devices
cascade primitives in APEX II and APEX 20K designs to a wire primitive
when compiled for Stratix and Stratix GX devices. These architectural
changes are transparent to the user and do not require design changes.
Figure 10–1. Carry Chain Implementation in APEX II & APEX 20K Devices vs.
Stratix & Stratix GX Devices
APEX II & APEX 20K Devices
Stratix Devices
Carry Chains
Carry-Select
Chains
LABs (with 10 LEs Each)
MultiTrack Interconnect
Stratix and Stratix GX devices use the MultiTrack interconnect structure
to provide a high-speed connection between logic resources using
performance-optimized routing channels of different lengths. This
feature maximizes overall design performance by placing critical paths
on routing lines with greater speed, resulting in minimal propagation
delay.
Altera Corporation
July 2005
10–3
Stratix Device Handbook, Volume 2
General Architecture
Stratix and Stratix GX device MultiTrack interconnect resources are
described in Table 10–2.
Table 10–2. Stratix & Stratix GX Device MultiTrack Interconnect Resources
Routing Type
Row
Interconnect
Direct link
Span
Adjacent LABs and/or blocks
Row
R4
Four LAB units horizontally
Row
R8
Eight LAB units horizontally
Row
R24
Horizontal routing across the width of the device
Column
C4
Four LAB units vertically
Column
C8
Eight LAB units vertically
Column
C16
Vertical routing across the length of the device
Direct link routing saves row routing resources while providing fast
communication paths between resource blocks. Direct link interconnects
allow an LAB, digital signal processing (DSP) block, or TriMatrixTM
memory block to drive data into the local interconnect of its left and right
neighbors. LABs, DSP blocks, and TriMatrix memory blocks can also use
direct link interconnects to drive data back into themselves for feedback.
The Quartus II software automatically uses these routing resources to
enhance design performance.
f
For more information about LE architecture and the MultiTrack
interconnect structure in Stratix and Stratix GX devices, see the Stratix
Device Family Data Sheet section of the Stratix Device Handbook, Volume 1
or the Stratix GX Device Family Data Sheet section of the Stratix GX Device
Handbook, Volume 1.
DirectDrive Technology
When using APEX II or APE 20K devices, you must place critical paths in
the same MegaLABTM column to improve performance. Additionally, you
should place critical paths in the same MegaLAB structure for optimal
performance. However, this restriction does not exist in Stratix and
Stratix GX devices because they do not contain MegaLAB structures.
With the new DirectDriveTM technology in Stratix and Stratix GX devices,
the actual distance between the source and destination of a path is the
most important criteria for meeting timing performance. DirectDrive
technology ensures that the same routing resources are available to each
design block, regardless of its location in the device.
10–4
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Transitioning APEX Designs to Stratix & Stratix GX Devices
Architectural Element Names
The architectural element naming system within Stratix and Stratix GX
devices differs from the row-column coordinate system (for example,
LC1_A2, LAB_B1) used in previous Altera device families. Stratix and
Stratix GX devices uses a new naming system based on the X-Y
coordinate system, (X, Y). A number (N) designates the location within the
block where the logic resides, such as LEs within an LAB. Because the
Stratix and Stratix GX architectures are column-based, this naming
simplifies location assignments. Stratix and Stratix GX architectural
blocks include:
■
■
■
■
■
■
LAB: logic array block
DSP: digital signal processing block
DSPOUT: adder/subtractor/accumulator or summation block of the
DSP block
M512: 512-bit memory block
M4K: 4-Kbit memory block
M-RAM: 512-Kbit memory block
Elements within architectural blocks include:
■
■
■
■
■
■
Altera Corporation
July 2005
LE: logic element
IOC: I/O element
PLL: phase-locked loop
DSPMULT: DSP block multiplier
SERDESTX: transmitter serializer/deserializer
SERDESRX: receiver serializer/deserializer
10–5
Stratix Device Handbook, Volume 2
General Architecture
Table 10–3 highlights the new location syntax used for Stratix and
Stratix GX devices.
Table 10–3. Stratix & Stratix GX Location Assignment Syntax
Architectural
Elements
Example of Location Syntax
Element Name
Location Syntax
Location
Description
Blocks
LAB, DSP,
<element_name>_X<number> LAB_X1_Y1
DSPOUT, M512, _Y<number>
M4K, M-RAM
Logic
LE, IOC, PLL,
DSPMULT,
SERDESTX,
SERDESRX
<element_name>_X<number> LC_X1_Y1_N0
_Y<number>_N<number>
Designates the first
LE, N0, in the LAB
located in row 1,
column 1
Pins (1)
I/O pins
pin_<pin_label>
Pin 5
pin_5
Designates the LAB in
row 1, column 1
Note to Table 10–3:
(1)
You can make assignments to I/O pads using IOC_X<number>_Y<number>_N<number>.
Use the following guidelines with the new naming system:
■
■
■
■
■
The anchor point, or origin, in Stratix and Stratix GX devices is in the
bottom-left corner, instead of the top-left corner as in APEX II and
APEX 20K devices.
The anchor point, or origin, of a large block element (e.g., a M-RAM
or DSP block) is also the bottom-left corner.
All numbers are zero-based, meaning the origin at the bottom-left of
the device is X0, Y0.
The I/O pins constitute the first and last rows and columns in the
X-Y coordinates. Therefore, the bottom row of pins resides in
X<number>, Y0, and the first left column of pins resides in X0,
Y<number>.
The sub-location of elements, N, numbering begins at the top.
Therefore, the LEs in an LAB are still numbered from top to bottom,
but start at zero.
Figure 10–2 show the Stratix and Stratix GX architectural element
numbering convention. Figure 10–3 displays the floorplan view in the
Quartus II software.
10–6
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Transitioning APEX Designs to Stratix & Stratix GX Devices
Figure 10–2. Stratix & Stratix GX Architectural Elements Note (1)
M4K RAM Blocks are
Two Units Wide and
One Unit High
LAB
(1,18)
LAB
(11,18)
M512
(12,18)
LAB
(13,18)
M4K
(14,18)
LAB
(16,18)
LAB
(1,17)
LAB
(11,17)
M512
(12,17)
LAB
(13,17)
M4K
(14,17)
LAB
(16,17)
LAB
(1,16)
LAB
(11,16)
M512
(12,16)
LAB
(13,16)
M4K
(14,16)
LAB
(16,16)
LAB
(1,15)
LAB
(11,15)
M512
(12,15)
LAB
(13,15)
M4K
(14,15)
LAB
(16,15)
M4K
(14,14)
M4K
(14,13)
Mega RAM Block is
13 Units Wide and
13 Units High
DSP Block (17,1)
is Two Units Wide
and Eight Units High
DSPMULT
(17,7,0)
and
(17,7,1)
(3)
DSPMULT
(17,5,0)
and
(17,5,1)
LAB
(16,14)
DSPOUT
(18,1,0)
and
(18,1,7)
LAB
(16,13)
DSPMULT
(17,3,0)
and
(17,3,1)
Mega RAM (1,2)
M4K
(14,2)
Pins
LAB
(16,2)
DSPMULT
(17,1,0)
and
(17,1,1)
(2)
PLL
(0,1,0)
LAB
(1,1)
LAB
(11,1)
M512
(12,1)
M4K
(14,1)
LAB
(13,1)
(2)
LAB
(16,1)
(2)
Origin (0, 0)
Notes to Figure 10–2:
(1)
(2)
(3)
Figure 10–2 shows part of a Stratix and Stratix GX device.
Large block elements use their lower-left corner for the coordinate location.
The Stratix GX architectural elements include transceiver blocks on the right side of the device.
Altera Corporation
July 2005
10–7
Stratix Device Handbook, Volume 2
TriMatrix Memory
Figure 10–3. LE Numbering as Shown in the Quartus II Software
TriMatrix
Memory
TriMatrix memory has three different sizes of memory blocks, each
optimized for a different purpose or application. M512 blocks consist of
512 bits plus parity (576 bits), M4K blocks consist of 4K bits plus parity
(4,608 bits), and M-RAM blocks consist of 512K bits plus parity
(589,824 bits). This new structure differs from APEX II and APEX 20K
devices, which feature uniformly sized embedded system blocks (ESBs)
either 4 Kbits (APEX II devices) or 2 Kbits (APEX 20K devices) large.
Stratix and Stratix GX TriMatrix memory blocks give you advanced
control of each memory block, with features such as byte enables, parity
bit storage, and shift-register mode, as well as mixed-port width support
and true dual-port mode operation.
10–8
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Transitioning APEX Designs to Stratix & Stratix GX Devices
Table 10–4 compares TriMatrix memory with ESBs.
Table 10–4. Stratix & Stratix GX TriMatrix Memory Blocks vs. APEX II & APEX 20K ESBs
Stratix & Stratix GX
Features
APEX II ESB
M512 RAM
Size (bits)
576
M4K RAM
4,608
APEX 20K ESB
M-RAM
589,824
4,096
2,048
Parity bits
Yes
Yes
Yes
No
No
Byte enable
No
Yes
Yes
No
No
True dual-port
mode
No
No
Yes
Yes
Yes
Includes support Includes support Includes support
for mixed width for mixed width for mixed width
Embedded shift
register
Yes
Yes
No
No
No
Dedicated
contentaddressable
memory (CAM)
support
No
No
No
Yes
Yes
Pre-loadable
initialization with a
.mif (1)
Yes
Yes
No
Yes
Yes
Packed mode (2)
No
Yes
No
Yes
Yes
Feed-through
behavior
Rising edge
Rising edge
Rising edge
Falling edge
Falling edge
Output power-up
condition
Powers up
cleared even if
using a .mif (1)
Powers up
cleared even if
using a .mif (1)
Powers up with
unknown state
Powers up
cleared or to
initialized value,
if using a .mif (1)
Powers up
cleared or to
initialized value,
if using a .mif (1)
Notes to Table 10–4:
(1)
(2)
.mif: Memory Initialization File.
Packed mode refers to combining two single-port RAM blocks into a single RAM block that is placed into true
dual-port mode.
Stratix and Stratix GX TriMatrix memory blocks only support pipelined
mode, while APEX II and APEX 20K ESBs support both pipelined and
flow-through modes. Since all TriMatrix memory blocks can be
pipelined, all input data and address lines are registered, while outputs
can be either registered or combinatorial. You can use Stratix and
Stratix GX memory block registers to implement input and output
registers without utilizing additional resources. You can compile designs
containing pipelined memory blocks (inputs registered) for Stratix and
Stratix GX devices without any modifications. However, if an APEX II or
Altera Corporation
July 2005
10–9
Stratix Device Handbook, Volume 2
TriMatrix Memory
APEX 20K design contains flow-through memory, you must modify the
memory modules to target the Stratix and Stratix GX architectures (see
“Memory Megafunctions” on page 10–12 for more information).
f
For more information about TriMatrix memory and converting flowthrough memory modules to pipelined, see the TriMatrix Embedded
Memory Blocks in Stratix & Stratix GX Devices chapter in the Stratix GX
Device Handbook and AN 210: Converting Memory from Asynchronous to
Synchronous for Stratix & Stratix GX Designs.
Same-Port Read-During-Write Mode
In same-port read-during-write mode, the RAM block can be in singleport, simple dual-port, or true dual-port mode. One port from the RAM
block both reads and writes to the same address location using the same
clock. When APEX II or APEX 20K devices perform a same-port readduring-write operation, the new data is available on the falling edge of
the clock cycle on which it was written, as shown in Figure 10–4. When
Stratix and Stratix GX devices perform a same-port read-during-write
operation, the new data is available on the rising edge of the same clock
cycle on which it was written, as shown in Figure 10–5. This holds true for
all TriMatrix memory blocks.
Figure 10–4. Falling Edge Feed-Through Behavior
(APEX II & APEX 20K Devices) Note (1)
inclock
data_in
A
B
wren
data_out
Old
A
Note to Figure 10–4:
(1)
Figures 10–4 and 10–5 assume that the address stays constant throughout and that
the outputs are not registered.
10–10
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Transitioning APEX Designs to Stratix & Stratix GX Devices
Figure 10–5. Rising Edge Feed-Through Behavior
(Stratix & Stratix GX Devices) Note (1)
inclock
data_in
A
B
wren
data_out Old
A
Note to Figure 10–5:
(1)
Figures 10–4 and 10–5 assume that the address stays constant throughout and that
the outputs are not registered.
Mixed-Port Read-During-Write Mode
Mixed-port read-during-write mode occurs when a RAM block in simple
or true dual-port mode has one port reading and the other port writing to
the same address location using the same clock. In APEX II and
APEX 20K designs, the ESB outputs the old data in the first half of the
clock cycle and the new data in the second half of the clock cycle, as
indicated by Figure 10–6.
Figure 10–6. Mixed-Port Feed-Through Behavior
(APEX II & APEX 20K Devices) Note (1)
inclock
Port A
data_in
A
B
Port A
wren
Port B
wren
Port B
data_out
Old
A
B
Note to Figure 10–6:
(1)
Figure 10–6 assumes that outputs are not registered.
Stratix and Stratix GX device RAM outputs the new data on the rising
edge of the clock cycle immediately after the data was written. When you
use Stratix and Stratix GX M512 and M4K blocks, you can choose whether
to output the old data at the targeted address or output a don’t care value
during the clock cycle when the new data is written. M-RAM blocks
Altera Corporation
July 2005
10–11
Stratix Device Handbook, Volume 2
TriMatrix Memory
always output a don’t care value. Figures 10–7 and 10–8 show the feedthrough behavior of the mixed-port mode. You can use the altsyncram
megafunction to set the output behavior during mixed-port read-duringwrite mode.
Figure 10–7. Mixed-Port Feed-Through Behavior (OLD_DATA) Note (1)
inclock
addressA and
addressB
Port A
data_in
Address Q
A
B
Port A
wren
Port B
wren
Port B
data_out
Old
A
B
Note to Figure 10–7:
(1)
Figures 10–7 and 10–8 assume that the address stays constant throughout and that
the outputs are not registered.
Figure 10–8. Mixed-Port Feed-Through Behavior (DONT_CARE) Note (1)
inclock
addressA and
addressB
Port A
data_in
Address Q
A
B
Port A
wren
Port B
wren
Port B
data_out
Unknown
B
Note to Figure 10–8:
(1)
Figures 10–7 and 10–8 assume that the address stays constant throughout and that
the outputs are not registered.
Memory Megafunctions
To convert RAM and ROM originally targeting the APEX II or APEX 20K
architecture to Stratix or Stratix GX memory, specify Stratix or Stratix GX
as the target family in the MegaWizard Plug-In Manager. The software
10–12
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Transitioning APEX Designs to Stratix & Stratix GX Devices
updates the memory module for the Stratix or Stratix GX architecture and
instantiates the new synchronous memory megafunction, altsyncram,
which supports both RAM and ROM blocks in the Stratix and Stratix GX
architectures.
FIFO Conditions
First-in first-out (FIFO) functionality is slightly different in Stratix and
Stratix GX devices compared to APEX II and APEX 20K devices. Stratix
and Stratix GX devices do not support simultaneous reads and writes
from an empty FIFO buffer. Also, Stratix and Stratix GX devices do not
support the lpm_showahead parameter when targeting a FIFO buffer
because the TriMatrix memory blocks are synchronous. The
lpm_showahead parameter for APEX II and APEX 20K devices puts the
FIFO buffer in “read-acknowledge” mode so the first data written into the
FIFO buffer immediately flows through to the output. Other than these
two differences, all APEX II and APEX 20K FIFO functions are fully
compatible with the Stratix and Stratix GX architectures.
Design Migration Mode in Quartus II Software
The Quartus II software features a migration mode for simplifying the
process of converting APEX II and APEX 20K memory functions to the
Stratix or Stratix GX architecture. If the design can use the Stratix or
Stratix GX altsyncram megafunction as a replacement for a previous
APEX II or APEX 20K memory function while maintaining functionally
similar behavior, the Quartus II software automatically converts the
memory. The software produces a warning message during compilation
reminding you to verify that the design migrated correctly.
For memory blocks with all inputs registered, the existing megafunction
is converted to the new altsyncram megafunction. The software
generates a warning when the altsyncram megafunction is
incompatible. For example, a RAM block with all inputs registered except
the read enable compiles with a warning message indicating that the
read-enable port is registered.
You can suppress warning messages for the entire project or for
individual memory blocks by setting the
SUPPRESS_MEMORY_CONVERSION_WARNINGS parameter to “on” as a
global parameter by selecting Assignment Organizer (Tools menu). In
the Assignment Organizer window, click Parameters in the Assignment
Categories box. Type SUPPRESS_MEMORY_CONVERSION_WARNINGS in
the Assignment Name box and type ON in the Assignment Setting box.
To suppress these warning messages on a per-memory-instance basis, set
the SUPPRESS_MEMORY_CONVERSION_WARNINGS parameter in the
Assignment Organizer to “on” for the memory instance.
Altera Corporation
July 2005
10–13
Stratix Device Handbook, Volume 2
TriMatrix Memory
If the functionality of the APEX II or APEX 20K memory megafunction
differs from the altsyncram functionality and at least one clock feeds
the memory megafunction, the Quartus II software converts the APEX II
or APEX 20K memory megafunction to the Stratix or Stratix GX
altsyncram megafunction. This conversion is useful for an initial
evaluation of how a design might perform in Stratix or Stratix GX devices
and should only be used for evaluation purposes. During this process, the
Quartus II software generates a warning that the conversion may be
functionally incorrect and timing results may not be accurate. Since the
functionality may be incorrect and the compilation is only intended for
comparative purposes, the Quartus II software does not generate a
programming file. A functionally correct conversion requires manually
instantiating the altsyncram megafunction and may require additional
design changes.
If the previous memory function does not have a clock (fully
asynchronous), the fitting-evaluation conversion results in an error
message during compilation and does not successfully convert the
design.
f
See AN 210: Converting Memory from Asynchronous to Synchronous for
Stratix & Stratix GX Designs for more information.
Table 10–5 summarizes the possible scenarios when using design
migration mode and the resulting behavior of the Quartus II software.
The most common cases where design-migration mode may have
difficulty converting the existing design are when:
■
A port is reading from an address that is being written to by another
port (mixed-port read-during-write mode). If both ports are using
the same clock, the read port in Stratix and Stratix GX devices do not
see the new data until the next clock cycle, after the new data was
written.
10–14
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Transitioning APEX Designs to Stratix & Stratix GX Devices
■
There are differences in power-up behavior between APEX II,
APEX 20K, and Stratix and Stratix GX devices. You should manually
account for these differences to maintain desired operation of the
system.
Table 10–5. Migration Mode Summary
Memory
Configuration
Single-port
Conditions
Possible
Instantiated
Megafunctions
All inputs are registered. altram
Quartus II Warning
Message(s)
Programming
File
Generated
Power-up differences. (1)
Yes
Power-up differences.
Mixed-port read- duringwrite. (1)
Yes
Yes
altrom
lpm_ram_dq
lpm_ram_io
lpm_rom
All inputs are registered. altdpram
Multi-port (two-,
three-, or four-port
lpm_ram_dp
functions)
altqpram
alt3pram
Dual-port
Read-enable ports are
unregistered.
Other inputs registered.
altdpram
lpm_ram_dp
altqpram
alt3pram
Power-up differences.
Mixed-port read- duringwrite.
Read enable will be
registered. (1)
Dual-port
Any other unregistered
port except read-enable
ports.
Clock available.
altdpram
lpm_ram_dp
altqpram
alt3pram
Compile for fitting- evaluation No
purposes.
Single-port
At least one registered
input.
Clock available.
altram
lpm_ram_dq
lpm_ram_io
Compile for fitting- evaluation No
purposes.
No clock
No clock.
altram
altrom
altdpram
altqpram
alt3pram
altdpram
lpm_ram_dq
lpm_ram_io
lpm_rom
lpm_ram_dp
lpm_ram_dp
Error – no conversion
possible.
No
Note to Table 10–5:
(1)
If the SUPPRESS_MEMORY_COUNVERSION_WARNINGS parameter is turned on, the Quartus II software does not
issue these warnings.
Altera Corporation
July 2005
10–15
Stratix Device Handbook, Volume 2
DSP Block
DSP Block
Stratix and Stratix GX device DSP blocks outperform LE-based
implementations for common DSP functions. Each DSP block contains
several multipliers that can be configured for widths of 9, 18, or 36 bits.
Depending on the mode of operation, these multipliers can optionally
feed an adder/subtractor/accumulator or summation unit.
You can configure the DSP block’s input registers to efficiently
implement shift registers for serial input sharing, eliminating the need for
external shift registers in LEs. You can add pipeline registers to the DSP
block for accelerated operation. Registers are available at the input and
output of the multiplier, and at the output of the
adder/subtractor/accumulator or summation block.
DSP blocks have four modes of operation:
■
■
■
■
Simple multiplier mode
Multiply-accumulator mode
Two-multipliers adder mode
Four-multipliers adder mode
Associated megafunctions are available in the Quartus II software to
implement each mode of the DSP block.
DSP Block Megafunctions
You can use the lpm_mult megafunction to configure the DSP block for
simple multiplier mode. You can set the lpm_mult Multiplier
Implementation option in the MegaWizard Plug-In Manager to either
use the default implementation, ESBs, or the DSP blocks. If you select the
Use Default option, the compiler first attempts to place the multiplier in
the DSP blocks. However, under certain conditions, the compiler may
implement the multiplier in LEs. The placement depends on factors such
as DSP block resource consumption, the width of the multiplier, whether
an operand is a constant, and other options chosen for the megafunction.
Stratix and Stratix GX devices do not support the Use ESBs option. If you
select this option, the Quartus II software tries to place the multiplier in
unused DSP blocks.
You can recompile APEX II or APEX 20K designs using the lpm_mult
megafunction for Stratix and Stratix GX devices in the Quartus II
software without changing the megafunction. This makes converting
lpm_mult megafunction designs to Stratix or Stratix GX devices
straightforward.
10–16
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Transitioning APEX Designs to Stratix & Stratix GX Devices
APEX II and APEX 20K designs use pipeline stages to improve registered
performance of LE-based multipliers at the expense of latency. However,
you may not need to use pipeline stages when targeting Stratix and
Stratix GX high-speed DSP blocks. The DSP blocks offer three sets of
dedicated pipeline registers. Therefore, Altera recommends that you
reduce the number of pipeline stages to three or fewer and implement
them in the DSP blocks. Additional pipeline stages are implemented in
LEs, which add latency without providing any performance benefit.
For example, you can configure a DSP block for 36 × 36-bit multiplication
using the lpm_mult megafunction. If you specify two pipeline stages,
the software uses the DSP block input and pipeline registers. If you
specify three pipeline stages, the software places the third pipeline stage
in the DSP block output registers. This design yields the same
performance with three pipeline stages because the critical path for a
36 × 36-bit operation is within the multiplier. With four or more pipeline
stages, the device inefficiently uses LE resources for the additional
pipeline stages. Therefore, if multiplier modules in APEX II or APEX 20K
designs are converted to Stratix or Stratix GX designs and do not require
the same number of pipeline stages, the surrounding circuitry must be
modified to preserve the original functionality of the design.
A design with multipliers feeding an accumulator can use the
altmult_accum (MAC) megafunction to set the DSP block in multiplyaccumulator mode. If the APEX II or APEX 20K design already uses LEbased multipliers feeding an accumulator, the Quartus II software does
not automatically instantiate the new altmult_accum (MAC)
megafunction. Therefore, you should use the MegaWizard Plug-In
Manager to instantiate the altmult_accum (MAC) megafunction. You
can also use LeonardoSpectrum™ or Synplify synthesis tools, which have
DSP block inference support, to instantiate the megafunction.
Designs that use multipliers feeding into adders can instantiate the new
altmult_add megafunction to configure the DSP blocks for twomultipliers adder or four-multipliers adder mode. You can also use the
altmult_add megafunction for stand-alone multipliers to take
advantage of the DSP blocks features such as dynamic sign control of the
inputs and the input shift register connections. These features are not
accessible through the lpm_mult megafunction. If your APEX II or
APEX 20K designs already use multipliers feeding an adder/subtractor,
the Quartus II software does not automatically infer the new
altmult_add megafunction. Therefore, you should step through the
MegaWizard Plug-In Manager to instantiate the new altmult_add
megafunction or use LeonardoSpectrum or Synplify synthesis tools,
which have DSP block inference support.
Altera Corporation
July 2005
10–17
Stratix Device Handbook, Volume 2
PLLs & Clock Networks
Furthermore, the altmult_add and altmult_accum (MAC)
megafunctions are only available for Stratix and Stratix GX devices
because these megafunctions target Stratix and Stratix GX DSP blocks,
which are not available in other device families. If you attempt to use
these megafunctions in designs that target other Altera device families,
the Quartus II software reports an error message. Use lpm_mult and an
lpm_add_sub or an altaccumulate megafunction for similar
functionality in other device families.
If you use a third-party synthesis tool, you may be able to avoid the
megafunction conversion process. LeonardoSpectrum and Synplify
provide inference support for lpm_mult, altmult_add, and
altmult_accum (MAC) to use the DSP blocks.
If your design does not require you to implement all the multipliers in
DSP blocks, you must manually set a global parameter or a parameter for
each instance to force the tool to implement the lpm_mult megafunction
in LEs. Depending on the synthesis tools, inference of DSP blocks is
handled differently.
f
PLLs & Clock
Networks
For more information about using DSP blocks in Stratix and Stratix GX
devices, see the DSP Blocks in Stratix & Stratix GX Devices chapter of the
Stratix Device Handbook.
Stratix and Stratix GX devices provide exceptional clock management
with a hierarchical clock network and up to four enhanced phase-locked
loops (PLLs) and eight fast PLLs versus the four general-purpose PLLs
and four True-LVDSTM PLLs in APEX II devices. By providing superior
clock interfacing, numerous advanced clocking features, and significant
enhancements over APEX II and APEX 20K PLLs, the Stratix and
Stratix GX device PLLs increase system performance and bandwidth.
Clock Networks
There are 16 global clock networks available throughout each Stratix or
Stratix GX device as well as two fast regional and four regional clock
networks per device quadrant, resulting in up to 40 unique clock
networks per device. The increased number of dedicated clock resources
available in Stratix and Stratix GX devices eliminate the need to use
general-purpose I/O pins as clock inputs.
Stratix EP1S25 and smaller devices have 16 dedicated clock pins and
EP1S30 and larger devices have four additional clock pins to feed various
clocking networks. In comparison, APEX II devices have eight dedicated
clock pins and APEX 20KE and APEX 20KC devices have four dedicated
clock pins.
10–18
Stratix Device Handbook, Volume 2
Altera Corporation
July 2005
Transitioning APEX Designs to Stratix & Stratix GX Devices
The dedicated clock pins in Stratix and Stratix GX devices can feed the
PLL clock inputs, the global clock networks, and the regional clock
networks. PLL outputs and internally-generated signals can also drive
the global clock network. These global clocks are available throughout
the entire device to clock all device resources.
Stratix and Stratix GX devices are divided into four quadrants, each
equipped with four regional clock networks. The regional clock network
can be fed by either the dedicated clock pins or the PLL outputs within its
device quadrant. The regional clock network can only feed device
resources within its particular device quadrant.
Each Stratix and Stratix GX device provides eight dedicated fast clock
I/O pins FCLK[7..0] versus four dedicated fast I/O pins in APEX II
and APEX 20K devices. The fast regional clock network can be fed by
these dedicated FCLK[7..0] pins or by the I/O interconnect. The I/O
interconnect allows internal logic or any I/O pin to drive the fast regional
clock network. The fast regional clock network is available for generalpurpose clocking as well as high fan-out control signals such as clear,
preset, enable, TRDY and IRDY for PCI applications, or bidirectional or
output pins.
EP1S25 and smaller devices have eight fast regional clock networks, two
per device quadrant. The quadrants in EP1S30 and larger devices are
divided in half, and each half-quadrant can be clocked by one of the eight
fast regional networks. Additionally, each fast regional clock network can
drive its neighboring half-quadrant (within the same device quadrant).
PLLs
Table 10–6 highlights Stratix and Stratix GX PLL enhancements to
existing APEX II, APEX 20KE and APEX 20KC PLL features.
Table 10–6. Stratix & Stratix GX PLL vs. APEX II, APEX 20KE & APEX 20KC PLL Features (Part 1 of 2)
Stratix & Stratix GX
Feature
APEX II PLLs
Enhanced PLLs
Fast PLLs
APEX 20KE &
APEX 20KC PLLs
Number of PLLs
Two (EP1S30 and
smaller devices);
four (EP1S40 and
larger devices) (9)
Four (EP1S25 and Four generalpurpose PLLs and
smaller devices);
four LVDS PLLs
eight (EP1S30
and larger
devices) (10)
Up to four generalpurpose PLLs. Up
to two LVDS PLLs.
(1)
Minimum input frequency
3 MHz
15 MHz
1.5 MHz
1.5 MHz
Maximum input frequency
250 to 582 MHz (2) 644.5 MHz (11)
420 MHz
420 MHz
Altera Corporation
July 2005
10–19
Stratix Device Handbook, Volume 2
PLLs & Clock Networks
Table 10–6. Stratix & Stratix GX PLL vs. APEX II, APEX 20KE & APEX