TI TAS3103DBT

Data Manual
February 2004
Digital Audio Solutions
SLES038C
IMPORTANT NOTICE
Texas Instruments Incorporated and its subsidiaries (TI) reserve the right to make corrections, modifications,
enhancements, improvements, and other changes to its products and services at any time and to discontinue
any product or service without notice. Customers should obtain the latest relevant information before placing
orders and should verify that such information is current and complete. All products are sold subject to TI’s terms
and conditions of sale supplied at the time of order acknowledgment.
TI warrants performance of its hardware products to the specifications applicable at the time of sale in
accordance with TI’s standard warranty. Testing and other quality control techniques are used to the extent TI
deems necessary to support this warranty. Except where mandated by government requirements, testing of all
parameters of each product is not necessarily performed.
TI assumes no liability for applications assistance or customer product design. Customers are responsible for
their products and applications using TI components. To minimize the risks associated with customer products
and applications, customers should provide adequate design and operating safeguards.
TI does not warrant or represent that any license, either express or implied, is granted under any TI patent right,
copyright, mask work right, or other TI intellectual property right relating to any combination, machine, or process
in which TI products or services are used. Information published by TI regarding third-party products or services
does not constitute a license from TI to use such products or services or a warranty or endorsement thereof.
Use of such information may require a license from a third party under the patents or other intellectual property
of the third party, or a license from TI under the patents or other intellectual property of TI.
Reproduction of information in TI data books or data sheets is permissible only if reproduction is without
alteration and is accompanied by all associated warranties, conditions, limitations, and notices. Reproduction
of this information with alteration is an unfair and deceptive business practice. TI is not responsible or liable for
such altered documentation.
Resale of TI products or services with statements different from or beyond the parameters stated by TI for that
product or service voids all express and any implied warranties for the associated TI product or service and
is an unfair and deceptive business practice. TI is not responsible or liable for any such statements.
Following are URLs where you can obtain information on other Texas Instruments products and application
solutions:
Products
Applications
Amplifiers
amplifier.ti.com
Audio
www.ti.com/audio
Data Converters
dataconverter.ti.com
Automotive
www.ti.com/automotive
DSP
dsp.ti.com
Broadband
www.ti.com/broadband
Interface
interface.ti.com
Digital Control
www.ti.com/digitalcontrol
Logic
logic.ti.com
Military
www.ti.com/military
Power Mgmt
power.ti.com
Optical Networking
www.ti.com/opticalnetwork
Microcontrollers
microcontroller.ti.com
Security
www.ti.com/security
Telephony
www.ti.com/telephony
Video & Imaging
www.ti.com/video
Wireless
www.ti.com/wireless
Mailing Address:
Texas Instruments
Post Office Box 655303 Dallas, Texas 75265
Copyright  2004, Texas Instruments Incorporated
Contents
Section
1
2
3
Title
Page
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1−1
1.1
Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1−1
1.2
Terminal Assignments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1−2
1.3
Hardware Block Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1−3
1.4
Functional Block Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1−4
1.5
Ordering Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1−5
1.6
Terminal Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1−5
1.7
Operational Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1−8
1.7.1
Terminal-Controlled Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1−9
1.7.2
I2C Bus-Controlled Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . 1−10
Hardware Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−1
2.1
Input and Output Serial Audio Ports (SAPs) . . . . . . . . . . . . . . . . . . . . . . 2−3
2.1.1
SAP Configuration Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−3
2.1.2
Processing Flow—SAP Input to SAP Output . . . . . . . . . . . 2−10
2.2
DPLL and Clock Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−14
2.3
Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−16
2.3.1
8051 Microprocessor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−16
2.3.2
I2C Bus Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−16
2.4
Digital Audio Processor (DAP) Arithmetic Unit . . . . . . . . . . . . . . . . . . . 2−21
2.5
Reset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−23
2.6
Power Down . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−23
2.7
Watchdog Timer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−24
2.8
General-Purpose I/O (GPIO) Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−24
2.8.1
GPIO Functionality—I2C Master Mode . . . . . . . . . . . . . . . . 2−25
2.8.2
GPIO Functionality—I2C Slave Mode . . . . . . . . . . . . . . . . . . 2−26
Firmware Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−1
3.1
I2C Coefficient Number Formats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−1
3.1.1
28-Bit 5.23 Number Format . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−1
3.1.2
48-Bit 25.23 Number Format . . . . . . . . . . . . . . . . . . . . . . . . . . 3−2
3.2
Input Crossbar Mixers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−4
3.3
3D Effects Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−8
3.3.1
CH1/CH2 Effects Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−8
3.3.2
CH3 Effects Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−8
3.4
Biquad Filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−10
3.5
Bass and Treble Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−11
3.5.1
Treble and Bass Processing and Concurrent I2C
Read Transactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−15
3.6
Soft Volume/Loudness Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−17
iii
4
iv
3.6.1
Soft Volume . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−17
3.6.2
Loudness Compensation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−24
3.6.3
Time Alignment and Reverb Delay Processing . . . . . . . . . . 3−26
3.7
Dynamic Range Control (DRC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−29
3.7.1
DRC Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−31
3.7.2
Compression/Expansion Coefficient Computation
Engine Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−33
3.7.3
DRC Compression/Expansion Implementation Examples 3−35
3.8
Spectrum Analyzer/VU Meter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−45
3.9
Dither . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−47
3.9.1
Dither Seeds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−48
3.9.2
Dither Mix Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−50
3.9.3
Dither Gain Mixers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−50
3.9.4
Dither Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−51
3.10 Output Crossbar Mixers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−54
Electrical Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4−1
4.1
Absolute Maximum Ratings Over Operating Temperature Ranges . . 4−1
4.2
Recommended Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . 4−1
4.3
Electrical Characteristics Over Recommended Operating
Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4−2
4.4
TAS3100 Timing Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4−3
4.4.1
Master Clock Signals Over Recommended Operating
Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4−3
4.4.2
Control Signals Over Recommended Operating Conditions
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4−4
4.4.3
Serial Audio Port Slave Mode Signals Over Recommended
Operating Conditions ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4−5
4.4.4
Serial Audio Port Master Mode Signals Over Recommended
Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4−6
4.4.5
I2C Slave Mode Interface Signals Over Recommended
Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4−7
A.1
I2C Subaddress Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A−1
A.2
TAS3103 Firmware Block Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A−19
List of Illustrations
Figure
Title
Page
2−1 TAS3103 Detailed Hardware Block Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−2
2−2 Discrete Serial Data Formats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−3
2−3 Four-Channel TDM Serial Data Formats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−3
2−4 SAP Configuration Subaddress Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−4
2−5 Recommended Procedure for Issuing SAP Configuration Updates . . . . . . . 2−5
2−6 Format Options: Input Serial Audio Port . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−7
2−7 TDM Format Options: Output Serial Audio Port . . . . . . . . . . . . . . . . . . . . . . . . 2−8
2−8 Discrete Format Options: Output Serial Audio Port . . . . . . . . . . . . . . . . . . . . . 2−9
2−9 Word Size Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−9
2−10 8 CH TDM Format Using SAP Modes 0101 and 1000 . . . . . . . . . . . . . . . . 2−10
2−11 6 CH Data, 8 CH Transfer TDM Format Using SAP Modes
0101 and 1000 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−10
2−12 SAP Input-to-Output Latency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−11
2−13 SAP Input-to-Output Latency for I2S Format Conversions . . . . . . . . . . . . . 2−12
2−14 DPLL and Clock Management Block Diagram . . . . . . . . . . . . . . . . . . . . . . . 2−15
2−15 I2C Slave Mode Communication Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−17
2−16 I2C Subaddress Access Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−18
2−17 Digital Audio Processor Arithmetic Unit Block Diagram . . . . . . . . . . . . . . . . 2−21
2−18 DAP Arithmetic Unit Data Word Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−22
2−19 DAP ALU Operation With Intermediate Overflow . . . . . . . . . . . . . . . . . . . . . 2−22
2−20 TAS3103 Reset Circuitry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−23
2−21 GPIO Port Circuitry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2−25
2−22 Volume Adjustment Timing—Master I2C Mode . . . . . . . . . . . . . . . . . . . . . . . 2−27
3−1 5.23 Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−1
3−2 Conversion Weighting Factors—5.23 Format to Floating Point . . . . . . . . . . . 3−1
3−3 Alignment of 5.23 Coefficient in 32-Bit I2C Word . . . . . . . . . . . . . . . . . . . . . . . 3−2
3−4 25.23 Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−2
3−5 Alignment of 5.23 Coefficient in 32-Bit I2C Word . . . . . . . . . . . . . . . . . . . . . . . 3−3
3−6 Alignment of 25.23 Coefficient in Two 32-Bit I2C Words . . . . . . . . . . . . . . . . . 3−3
3−7 Serial Input Port to Processing Node Topology . . . . . . . . . . . . . . . . . . . . . . . . . 3−5
3−8. Input Mixer and Effects Block Topology—Internal Processing
Nodes A, B, C, D, E, and F . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−6
3−9 Input Mixer Topology—Internal Processing Nodes G and H . . . . . . . . . . . . . . 3−7
v
3−10 TAS3103 3D Effects Processing Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−9
3−11 Biquad Filter Structure and Coefficient Subaddress Format . . . . . . . . . . . . 3−10
3−12 Bass and Treble Filter Selections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−12
3−13 Bass and Treble Application Example—Subaddress Parameters . . . . . . . 3−14
3−14 I2C Bass/Treble Activity Monitor Procedure . . . . . . . . . . . . . . . . . . . . . . . . . 3−16
3−15 Soft Volume and Loudness Compensation Block Diagram . . . . . . . . . . . . . 3−18
3−16 Detailed Block Diagram—Soft Volume and Loudness Compensation . . . 3−25
3−17 Delay Line Memory Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−27
3−18 Maximum Delay Line Lengths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−28
3−19 DRC Positioning in TAS3103 Processing Flow . . . . . . . . . . . . . . . . . . . . . . . 3−29
3−20 Dynamic Range Compression (DRC) Transfer Function Structure . . . . . . 3−30
3−21 DRC Block Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−32
3−22 DRC Input Word Structure for 0-dB Channel Processing Gain . . . . . . . . . 3−36
3−23 DRC Transfer Curve—Example 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−38
3−24 DRC Transfer Curve—Example 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−40
3−25 DRC Transfer Curve—Example 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−42
3−26 DRC Transfer Curve—Example 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−44
3−27 Spectrum Analyzer/VU Meter Block Diagram . . . . . . . . . . . . . . . . . . . . . . . . 3−46
3−28 Logarithmic Number Conversions—Spectrum Analyzer/VU Meter . . . . . . 3−47
3−29 Dither Data Block Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−49
3−30 Dither Data Magnitude (Gain = 1.0) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−51
3−31 Triangular Dither Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−52
3−32 Triangular Dither Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−53
3−33 Processing Node to Serial Output Port Topology . . . . . . . . . . . . . . . . . . . . . 3−55
3−34 Output Crossbar Mixer Topology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3−56
4−1 Master Clock Signals Timing Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4−3
4−2 Control Signals Timing Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4−4
4−3 Serial Audio Port Slave Mode Timing Waveforms . . . . . . . . . . . . . . . . . . . . . . 4−5
4−4 TAS3100 Serial Audio Port Master Mode Timing Waveforms . . . . . . . . . . . . . 4−6
4−5 I2C SCL and SDA Timing Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4−8
4−6 I2C Start and Stop Conditions Timing Waveforms . . . . . . . . . . . . . . . . . . . . . . 4−8
vi
List of Tables
Table
Title
2−1 TAS3103 Throughput Latencies vs MCLK and LRCLK . . . . . . . . . . . . . . . . .
2−2 TAS3103 Clock Default Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2−3 I2C EEPROM Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2−4 Four Byte Write Exceptions—Reserved and Factory-Test I2C
Subaddresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2−5 Four Byte Read Exceptions—Reserved and Factory-Test I2C
Subaddresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2−6 GPIO Port Functionality—I2C Master Mode . . . . . . . . . . . . . . . . . . . . . . . . . .
3−1 Biquad Filter Breakout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3−2 Bass Shelf Filter Indices for 1/2-dB Adjustments . . . . . . . . . . . . . . . . . . . . . .
3−3 Treble Shelf Filter Indices for 1/2-dB Adjustments . . . . . . . . . . . . . . . . . . . . .
3−4 Volume Adjustment Gain Coefficients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3−5 DRC Example 2 Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3−6 DRC Example 3 Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3−7 DRC Example 4 Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3−8 Mixer Gain Setting for LSB Dither Data Insertion . . . . . . . . . . . . . . . . . . . . . .
Page
2−13
2−16
2−19
2−20
2−21
2−25
3−10
3−14
3−15
3−20
3−39
3−41
3−45
3−50
vii
viii
1 Introduction
The TAS3103 is a fully configurable digital audio processor that preserves high-quality audio by using a 48-bit data
path, 28-bit filter coefficients, and a single cycle 28 x 48-bit multiplier and 76-bit accumulator. Because of the
coefficient-configurable fixed-program architecture of the TAS3103, a complete set of user-specific audio processing
functions can be realized, with short development times, in a small, low power, low-cost device. A personal computer
(PC) GUI-based software development package and a comprehensive evaluation board provide additional facilities
to further reduce development times. The TAS3103 uses 1.8-V core logic with 3.3-V I/O buffers, and requires only
3.3-V power. The TAS3103 is available in a 38-pin TSSOP package.
1.1 Features
•
Audio Input/Output
− Four Serial Audio Input Channels
− Three Serial Audio Output Channels
− 8-kHz to 96-kHz Sample Rates Supported
− 15 Stereo/TDM Data Formats Supported
− Input/Output Data Format Selections Independent
− 16-, 18-, 20-, 24-, and 32-Bit Word Sizes Supported
•
Serial Master/Slave I2C Control Channel
•
Three Independent Monaural Processing Channels
− Programmable Four Stereo Input Digital Mixer
− 3D Effect and Reverb Structure and Filters
− Programmable 12 Band Digital Parametric EQ
− Programmable Digital Bass and Treble Controls
− Programmable Digital Soft Volume Control (24 dB to −-∞ dB)
− Soft Mute/Unmute
− Programmable Dither
− Programmable Loudness Compensation
− VU Meter and Spectral Analysis I2C Output
− Programmable Channel Delay (Up to 42 ms at 48 kHz)
− 192-dB Dynamic Range (Supports Up to 32-Bit Audio Data)
− Dual Threshold Dynamic Range Compression/Expansion
•
Electrical and Physical
− Single 3.3-V Power Supply
− 38-Pin TSSOP Package
− Low Power Standby
1−1
1.2 Terminal Assignments
DBT PACKAGE
(TOP VIEW)
SCLKIN
PWRDN
REGULATOR_EN
XTALI (1.8-V logic)
XTALO (1.8-V logic)
AVDD_BYPASS_CAP
A_VDDS (3.3 V)
AVSS
MCLKI
TEST
MICROCLK_DIV
I2C_SDA
I2C_SCL
SDIN1
SDIN2
SDIN3
SDIN4
GPIO0
GPIO1
1−2
1
38
2
37
3
36
4
35
5
34
6
33
7
32
8
31
9
30
10
29
11
28
12
27
13
26
14
25
15
24
16
23
17
22
18
21
19
20
LRCLK
ORIN
SCLKOUT2
SCLKOUT1
MCLKO
SDOUT3
SDOUT2
VDDS (3.3 V)
SDOUT1
DVDD_BYPASS_CAP
DVSS
I2CM_S
RST
CS1
CS0
PLL1
PLL0
GPIO3
GPIO2
1.3 Hardware Block Diagram
SDIN1
SDIN2
SDIN3
SDIN4
SDOUT1
SDOUT2
SDOUT3
28
64
64
Serial
Audio
Port
SCLKOUT2
SCLKOUT1
MCLKO
LRCLK
SCLKIN
TEST
PWRDN
PLL1
PLL0
XTALO
XTALI
RST
Digital Audio Processor
64
64
64
64
64
48
76-Bit
ALU
Controller
Oscillator
and
PLL
External
Data
RAM
Code
ROM
COEF
RAM
Data
Path
Data
RAM
48
Memory
Interface
54
Delay
Memory
(4K x 16)
Internal
Data
RAM
8051
MCU
(8-Bit)
Code
ROM
Volume
Update
Control
Registers
I2C
Serial
Interface
8
4
I2C_SDA
I2C_SCL
CS1
CS0
I2CM_S
GPIO[3:0]
1−3
1−4
SDIN4
SDIN3
SDIN2
SDIN1
XTALO
MCLKI
XTALI
MultiMode
Serial
to
PCM
Input
Port
Input
Crossbar
Mixer
Oscillator
PLL0
MultiMode
3D
Effects
Block
PLL
and
Dividers
Ch3
Ch2
Ch1
I2C_
PLL1 MC/Div SDA
1.4 Functional Block Diagram
Ch3
12
Biquad
Filters
12
Biquad
Filters
12
Biquad
Filters
I2C_
SCL
CS0
Microprocessor
RST
LRCLK
Treble
and
Bass
Treble
and
Bass
Treble
and
Bass
SCLKIN
Loudness
Compensation
Loudness
Compensation
DRC
Ganged
DRC
SCLKOUT1
SCLKOUT2
MCLKO
Delay
Delay
Programmable
Dither
Programmable
Dither
Programmable
Dither
Ch3
Ch2
Center
Ch1
Output
Crossbar
Multiplexer
A_VDDS
AVDD_
BYPASS_
CAP
AVSS
Voltage Regulation
DVSS
DVDD_
BYPASS_
CAP
Delay
VU Meter
or
Spectrum Analyzer
I2C_
Test PWRDN M_S
Loudness
Compensation
CS1
I2S Clock Input/Generation
Soft
Volume
Soft
Volume
Soft
Volume
3 Mono Processing Channels
GPIO(3:0)
VDDS
Multimode
PCM
to
Serial
Output
Port
SDOUT3
SDOUT2
SDOUT1
ORIN
1.5 Ordering Information
TA
PLASTIC
38-PIN TSSOP
(DBT)
0°C to 70°C
TAS3103DBT
−40°C to 85°C
TAS3103IDBT
1.6 Terminal Functions
TERMINAL
NAME
NO.
I/O
TYPE(1)
DESCRIPTION
PULLUP/
DOWN(2)
A_VDDS (3.3 V)
7
PWR
The PWR pin is used to input 3.3-V power to the DPLL and clock oscillator.
This pin can be connected to the same power source used to drive the DVSS
power pin. To achieve low DPLL jitter, this pin should be bypassed to AVSS
with a 0.01-µF capacitor (low ESR preferable).
None
AVDD_BYPASS_CAP
6
PWR
AVDD_BYPASS_CAP is a pinout of the internally regulated 1.8-VDC power
used by the DPLL and crystal oscillator. This pin should be connected to pin 8
with a 0.01-µF capacitor (low ESR preferable). This pin must not be used to
power external devices.
None
AVSS
8
PWR
None
CS0
24
I
D
CS1
25
I
D
AVSS is the ground reference for the internal DPLL and oscillator circuitry.
This pin needs to reference the same ground as DVSS power pin. To achieve
low DPLL jitter, ground noise at this pin must be minimized. The availability of
the AVSS pin allows a designer to use optimizing techniques such as star
ground connections, separate ground planes, or other quiet ground
distribution techniques to achieve a quiet ground reference at this pin.
CS0 is the LSB of a 2-bit code used to generate part of an I2C device address
that makes it possible to address four TAS3103 ICs on the same bus without
additional chip select logic. The pulldowns on the inputs select 00 as a default
when neither pin is connected.
CS1 is the MSB of a 2-bit code used to generate part of an I2C device address
that makes it possible to address four TAS3103 ICs on the same bus without
additional chip select logic.
DVDD_BYPASS_CAP
29
DVSS
28
GPIO0
18
I/O
D
GPIO1
19
I/O
GPIO2
20
GPIO3
I2CM_S
Pulldown
Pulldown
PWR
DVDD_BYPASS_CAP is a pin-out of the internally regulated 1.8-V power
used by all internal digital logic. This pin must not be used to power external
devices. A low ESR capacitor of at least 470 nF should be placed as close to
the device as possible between this pin and pin 28.
None
PWR
DVSS is the digital ground pin.
None
GPIO0 is a general-purpose I/O, controlled by the internal microprocessor
through I2C commands. When in the I2C master mode, GPIO0 serves as a
volume up command for CH1/CH2.
Pullup
D
GPIO1 is a general-purpose I/O, controlled by the internal microprocessor
through I2C commands. When in the I2C master mode, GPIO1 serves as a
volume down command for CH1/CH2.
Pullup
I/O
D
GPIO2 is a general-purpose I/O, controlled by the internal microprocessor
through I2C commands. When in the I2C master mode, GPIO2 serves as a
volume up command for CH3.
Pullup
21
I/O
D
GPIO3 is a general-purpose I/O, controlled by the internal microprocessor
through I2C commands. When in the I2C master mode, GPIO3 serves as a
volume down command for CH3.
Pullup
27
I
D
I2CM_S is a non-latched input that determines whether the TAS3103 acts as
an I2C master or slave. Logic high, or no connection, sets the TAS3103 as an
I2C master device. A logic low sets the TAS3103 as an I2C slave device. As a
master I2C device, the TAS3103 I2C port must have access to an external
EEPROM for input.
Pullup
1−5
TERMINAL
PULLUP/
DOWN(2)
NO.
I/O
TYPE(1)
DESCRIPTION
I2C_SCL
13
I/O
D
I2C_SCL is the I2C clock pin. When the TAS3103 I2C port is a master,
I2C_SCL is (1/2N) x (1/(M+1)) x 1/10 times the microprocessor clock, where N
and M are set to 2 and 8 respectively. When the TAS3103 I2C port is a slave,
input clock rates up to 400 kHz can be supported. This pin must be provided an
external pullup (5 kΩ is recommended for most applications).
External
pullup
required
I2C_SDA
12
I/O
D
I2C_SDA is the I2C bidirectional data pin. The TAS3103 I2C port can support
data rates up to 400K bits/sec. This pin must be provided an external pullup
(5 kΩ is recommended for most applications).
External
pullup
required
LRCLK
38
I/O
D
LRCLK is either an input or an output, depending on whether the TAS3103 is in
a master or slave serial audio port mode, which is determined by bit 22 of
subaddress 0xF9.
Pulldown
MCLKI
9
I
D
MCLKI is a master clock input that provides an alternative to using a fixed
crystal frequency. In DPLL modes, the input frequency of this clock can range
from 2.8 MHz to 24.576 MHz. In PLL bypass mode, frequencies up to 136 MHz
can be used. Whenever MCLKI is not used and XTALI/XTALO provide the
master clock input, MCLKI must be grounded.
None
MCLKO
34
O
D
MCLKO is the master output clock pin. It is produced by dividing MCLKI/XTALI
by 1, 2, or 4 (depending on the setting of a subaddress control field). MCLKO is
provided to interconnect, without the need for additional glue logic, the
TAS3103 interfaces chips that require different multiples of the audio sample
rate (FS) as a master clock.
None
MICROCLK_DIV
11
I
D
MICROCLK_DIV sets the division ratio between the digital audio processing
clock and the internal microprocessor clock. The audio-processing clock is the
DPLL output clock if PLL_bypass is not enabled. The audio-processing clock
is MCLKI/XTALI master clock if PLL_bypass is enabled. Logic high on this pin
sets the microprocessor clock equal to the audio-processing clock. A logic low
sets the microprocessor clock to 1/4 the digital audio-processing clock.
MICROCLK_DIV must be set low if the audio processing clock is > 36 MHz.
MICROCLK_DIV must be set high if the audio processing clock is ≤ 36 MHz.
Pulldown
ORIN
37
I
D
ORIN allows the processing of a multichannel signal set through two
TAS3103s without any additional components. One use of ORIN would be to
fully emulate a 6-channel audio processor at speeds up to a 96-kHz sample
rate with only two TAS3103s and no glue logic.
Pulldown
NAME
The two-chip configuration is accomplished by wiring the SDOUT1 port of one
of the two TAS3103 chips to the ORIN port of the second TAS3103. Internal to
the chip, the ORIN input is OR’ed with internal SDOUT1 data to generate the
resulting output data on channel SDOUT1. For TDM output formats, the
SDOUT1 outputs of the two chips differ in phasing in both the left and right
channels to arrive at the proper composite output. For discrete outputs, one
chip contributes the left channel of the composite SDOUT1, and the other chip
contributes the right channel of the composite SDOUT1.
If not used, ORIN must be connected to ground.
PLL0
22
I
D
PLL0 is the LSB of a 2-bit code used to select four different modes of DPLL
multiplexer/input divider operation. PLL[1:0] values of 00, 01, and 10 select
the DPLL input clock to be MCLKI/XTALI divided by 1, 2, and 4 respectively. A
value of 11 results in MCLKI/XTALI being substituted for the DPLL output. The
pullup/pulldown combination provides a default of 01 when neither pin is
connected.
Pullup
PLL1
23
I
D
PLL1 is the MSB of a 2-bit code used to select four different modes of DPLL
multiplexer/input divider operation. PLL[1:0] values of 00, 01, and 10 select
the DPLL input clock to be MCLKI/XTALI divided by 1, 2, and 4 respectively. A
value of 11 results in MCLKI/XTALI being substituted for the DPLL output. The
pullup/pulldown combination provides a default of 01 when neither pin is
connected.
Pulldown
1−6
TERMINAL
DESCRIPTION
PULLUP/
DOWN(2)
NO.
I/O
TYPE(1)
PWRDN
2
I
D
PWRDN powers down all logic and stops all clocks whenever logic high is
applied. However, the coefficient memory remains stable through a power
down cycle, as long as a reset is not sent after a power down cycle.
REGULATOR_EN
3
I
D
REGULATOR_EN is only used in factory tests. This pin should always be tied
to ground.
RST
26
I
D
RST is the master reset input. Applying a logic low to this pin generates a
master reset. The master reset results in all coefficients being set to their
power-up default state, all data memories being cleared, and all logic signals
being returned to their default values.
Pullup
SCLKIN
1
I
D
SCLKIN is the serial audio port (SAP) input data clock. This clock is only used
when the SAP is a slave. In master mode, SCLKOUT1 internally provides the
serial input clock (SCLKOUT1 from a given TAS3103 must not be connected
to SCLKIN on the same TAS3103 chip).
Pulldown
SCLKOUT1
35
O
D
SCLKOUT1 is one of two serial output bit clocks. It is divided from
MCLKI/XTALI in master mode, and SCLKIN in slave mode. Subaddress
control fields determine the divide ratio in both cases. When the serial audio
port is in a master mode, SCLKOUT1 is used to receive incoming serial data
and should be wired to the data source(s) providing data to the SDIN inputs.
None
SCLKOUT2
36
O
D
SCLKOUT2 is one of two serial output bit clocks. It is divided from
MCLKI/XTALI in master mode, and SCLKIN in slave mode. Subaddress
control fields determine the divide ratio in both cases. SCLKOUT2 is always
used to clock out serial data from the three serial SDOUT output data
channels. SCLKOUT2 is provided separately from SCLKOUT1 to allow
discrete in to TDM out and TDM in to discrete out data format conversions
without the use of external glue logic.
Output
SDIN1
14
I
D
SDIN1, SDIN2, SDIN3, and SDIN4 are the four TAS3103 serial data input
ports. All four input ports support four discrete (stereo) data formats. SDIN1 is
the only data input port that also supports eleven time division multiplexed
data formats. All four ports are capable of receiving data with bit rates up to
24.576 MHz.
Pulldown
SDIN2
15
I
D
SDIN2 is one of the four TAS3103 serial data input ports. SDIN2 supports four
discrete (stereo) data formats, and is capable of receiving data with bit rates
up to 24.576 MHz.
Pulldown
SDIN3
16
I
D
SDIN3 is one of the four TAS3103 serial data input ports. SDIN4 supports four
discrete (stereo) data formats, and is capable of receiving data with bit rates
up to 24.576 MHz.
Pulldown
SDIN4
17
I
D
SDIN4 is one of the four TAS3103 serial data input ports. SDIN4 supports four
discrete (stereo) data formats, and is capable of receiving data with bit rates
up to 24.576 MHz.
Pulldown
SDOUT1
30
O
D
SDOUT1, SDOUT2, and SDOUT3 are the three TAS3103 serial data output
ports. All three output ports support four discrete (stereo) data formats.
SDOUT1 is the only data output port that also supports eleven time division
multiplexed data formats. All three ports are capable of outputting data at bit
rates up to 24.576 MHz.
None
SDOUT2
32
O
D
SDOUT2 is one of the three serial data output ports. SDOUT2 supports four
discrete (stereo) data formats, and is capable of outputting data at bit rates up
to 24.576 MHz.
None
SDOUT3
33
O
D
SDOUT3 is one of the three serial data output ports. SDOUT3 supports four
discrete (stereo) data formats, and is capable of outputting data at bit rates up
to 24.576 MHz.
None
TEST
10
I
D
TEST is only used in factory tests. This pin must be left unconnected or
grounded.
Pulldown
NAME
Pulldown
None
1−7
TERMINAL
PULLUP/
DOWN(2)
NO.
I/O
TYPE(1)
DESCRIPTION
VDDS (3.3 V)
31
-
PWR
VDDS is the 3.3-V pin that powers (1) the 1.8-V internal power regulator used
to supply logic power to the chip and (2) the I/O ring. It is recommended that
this pin be bypassed to DVSS (pin 28) with a low ESR capacitor in the range of
0.01 µF.
None
XTALI (1.8-V logic)
4
I
A
XTALO and XTALI provide a master clock for the TAS3103 via use of an
external fundamental mode crystal. XTALI is the 1.8-V input port for the
oscillator circuit. See Note 3 for recommended crystal type and accompanying
circuitry. This pin should be grounded when the MCLKI pin is used as the
source for the master clock.
None
XTALO (1.8-V logic)
5
O
A
XTALO and XTALI provide a master clock for the TAS3103 via use of an
external fundamental mode crystal. XTALO is the 1.8-V output drive to the
crystal. XTALO can support crystal frequencies between 2.8 MHz and
20 MHz. See Note 3 for recommended crystal type and accompanying
circuitry. This pin should be left unconnected in applications using an external
clock input to MCLKI.
None
NAME
NOTES: 1. TYPE: A = analog; D = 3.3-V digital; PWR = power/ground/decoupling
2. All pullups are 20-µA weak pullups and all pulldowns are 20-µA weak pulldowns. The pullups and pulldowns are included to assure
proper input logic levels if the pins are left unconnected (pullups => logic 1 input; pulldowns => logic 0 input). Devices that drive inputs
with pullups must be able to sink 20 µA while maintaining a logic 0 drive level. Devices that drive inputs with pulldowns must be able
to source 20 µA, while maintaining a logic 1 drive level.
3. Crystal type and recommended circuit:
TAS3103
C1
OSC
Circuit
rd
XO
C2
XI
AVSS
•
Crystal type = parallel-mode, fundamental-mode crystal
•
rd = drive level control resistor—vendor specified
•
CL = Crystal load capacitance (capacitance of circuitry between the two terminals of the crystal)
•
CL = (C1 × C2) / (C1 + C2) + CS (where CS = board stray capacitance ~2 pF)
−
Example: Vendor recommended CL = 18 pF, CS = 3 pF ⇒ C1 = C2 = 2 x (18 − 3) = 30 pF
1.7 Operational Modes
The TAS3103 operation is governed by I/O terminal voltage level settings and register / coefficient settings within the
TAS3103. The terminal settings are wholly sufficient to address all external environments - allowing the remaining
configuration settings to be determined by either I2C commands or by the content of an I2C serial EEPROM (when
the I2C master mode is selected).
1−8
1.7.1 Terminal-Controlled Modes
1.7.1.1 Clock Control
MCLKI
PLL1
PLL0
0
0
DAP CLOCK
11 x MCLK
0
1
(11 x MCLK)/2
1
0
(11 x MCLK)/4
1
1
MCLK (PLL bypass)
MICROCLK_DIV
MICROPROCESSOR CLOCK
0
DAP clock/4
1
DAP clock
XTALI
PLL0
PLL1
MICROCLK_DIV
MCLK
Reference
Divider
Microprocessor
Scaler
PLL
÷ 11
Digital
Audio Processor
(DAP) Clock
1400 x Fs 3 DAP Clock 3 136 MHz
Microprocessor
Clock < 36 MHz
1.7.1.2 I2C Bus Setup
SLAVE ADDRESS
CS1
CS0
0x68/69
0
0
0x6A/6B
0
1
0x6C/D
1
0
0x6E/6F
1
1
I2C BUS MODE
I2CM_S
0
Slave
1
Master
TAS3103 I2C Slave Address
SDA
a6 = 0
SCL
1
ÎÎÎÎÎÎÎ ÎÎ ÎÎ ÎÎ ÎÎ ÎÎ ÎÎ
ÎÎÎÎÎÎÎ ÎÎ ÎÎ ÎÎ ÎÎ ÎÎ ÎÎ
a5 = 1
a4 = 1
a3 = 0
a2 = 1
a1=CS1
a0=CS0
2
3
4
5
6
7
R/W
8
ACK
9
Start
1.7.1.3 Power-Down/Sleep Selection
PWRDN
POWER STATUS
0
Active
1
Power down/sleep
1−9
1−10
Slave Addr
S
Slave Addr
Slave Addr
S
Slave Addr
Ack
m
s xxxxxxx
b
Ack
xxxxxxxx
Ack
xxxxxxxx
Ack
xxxxxxxx
xxxxxxxx
xxxxxxxx
xxxxxxxx
xxxxxxxx
xxxxxxxx
Ack
Ack
Ack
Ack
Ack
Ack
Ack
Ack
Ack
Ack
xxxxxxxx
Ack
Ack
Ack
xxxxxxxx
xxxxxxxx
0x4E
Channel 3
xxxxxxxx
0x4D
Channel 2
Ack
0x4C
Channel 1
m
Ack Sub-Addr Ack 0000 s xxx
b
m
0000 s xxx
b
Subaddress
xxxxxxxx
xxxxxxxx
xxxxxxxx
xxxxxxxx
xxxxxxxx
Reverberation Block
Reverberation Block Gains
m
Ack Sub-Addr Ack 0000 s xxx
b
m
0000 s xxx
b
m
0000 s xxx
b
m
0000 s xxx
b
m
0000 s xxx
b
Ack xxxxxx
Ack xxxxxx
Ack xxxxxx
Ack xxxxxx
Ack xxxxxx
l
s
b
l
s
b
l
s
b
l
s
b
l
s
b
b0
b1
Ack
Ack
G0
G1
Ack
Ack
b2
a2
Ack
Ack
a1
Ack
Ack
Ack
l
xxxxxx s
b
l
Ack xxxxxx s
b
Ack
l
Ack xxxxxx s
b
l
Ack xxxxxx s
b
xxxxxxxx
Effects Block BiQuad Filter Coefficients
0x34−0x4B
Ack
Output Mixer 28-Bit Gain Coefficients
0x84 – 0xA1
Input Mixer 28-Bit Gain Coefficients
0x01 − 0x33
Sub-Addr
m
Ack Sub-Addr Ack 0000 s xxx
b
Ack
SUBADDRESS(es)
0x00 – Starting I2C Check Word
0xFC – Ending I2C Check Word
I2C Bus-Controlled Modes
S
S
1.7.2
PARAMETER(s)
Node
48
48
48
28
28
b2
b1
76
76
76
Σ
76
76
76
a2
48
48
28
28
a1
Magnitude
Truncation
48
48
28
Reverberation Block
Reverberation
Delay
48
48
Gain Coefficient G1
(Format = 5.23)
28
Gain Coefficient G0
(Format = 5.23)
NOTE: All gain coefficients 5.23 numbers.
z−1
z−1
28
b0
32 or 48
28
(Format = 5.23)
Gain Coefficient
z−1
z−1
Σ
48
Node
48
S Check words apply to I2C master mode only
S In master I2C mode, the two check words are compared after EEPROM
download. If comparison fails, a second attempt is made. If the second
comparison fails, the parameters default to the slave default values.
S In slave I2C mode, the default value for both check words is:
0x81_42_24_18
1−11
S
xxxxxxxx
xxxxxxxx
xxxxxxxx
xxxxxxxx
Ack
Ack
Ack
Ack
Ack
Ack
Ack
Ack
Ack
xxxxxxxx
xxxxxxxx
Ack
Ack
Channel 3 = 0x75
Channel 2 = 0x74
Channel 1 = 0x73
Ack
Ack
xxxxxxxx
l
s
b
l
s
b
l
s
b
l
s
b
l
s
b
b0
b1
Ack
Ack
b2
a2
Ack
Ack
a1
Ack
Ack
Inline Gain
Ack Bypass Gain
Ack xxxxxx
Ack xxxxxx
Ack xxxxxx
Ack xxxxxx
Ack xxxxxx
l
Ack xxxxxx s
b
l
Ack xxxxxx s
b
xxxxxxxx
xxxxxxxx
xxxxxxxx
xxxxxxxx
xxxxxxxx
xxxxxxxx
Bass and Treble Gain Coefficients
xxxxxxxx
0x67−0x72
Channel 3
Ack
0x5B−0x66
Channel 2
m
Ack Sub-Addr Ack 0000 s xxx
b
m
0000 s xxx
b
Slave Addr
Slave Addr
S
0x4F−0x5A
Channel 1
m
Ack Sub-Addr Ack 0000 s xxx
b
m
0000 s xxx
b
m
0000 s xxx
b
m
0000 s xxx
b
m
0000 s xxx
b
Subaddress
MAIN FILTER
BLOCK
Cascaded (Twelve/Channel) Main Filter BiQuads
SUBADDRESS(es)
z−1
z−1
48
48
28
28
b2
b1
28
76
76
76
Σ
76
76
76
a2
Bass
Shelf Filter
48
48
z−1
z−1
28
Bypass Gain Coeeficient
(Format = 5.23)
28
Inline Gain Coeeficient
(Format = 5.23)
Bass and Treble Block
Treble
Shelf Filter
48
28
28
a1
Magnitude
Truncation
NOTE: All gain coefficients 5.23 numbers
48
b0
PARAMETER(s)
Σ
1−12
CH3
CH2
0x7B = Mix n to m
0x78 = Mix w to m
0x7A = Mix l to k
0x77 = Mix v to k
0x79 = Mix j to i
Ack
xxxxxxxx
CH 3
Bass and Treble
Block
Mix_w_to_m
CH 2
Bass and Treble
Block
Σ
w
Σ
CH 2 Soft
Volume
CH 3 Soft
Volume
Σ
l
Mix_l_to_k
Σ
n
Mix_n_to_m
Σ
o
DRC_bypass_3
Σ
q
Mix_n_to_q_via_DRC_mult
DRC_bypass_2
Σ
p
Mix_l_to_p_via_DRC_mult
Mix_j_to_o_via_DRC_mult
DRC_bypass_1
CH3
CH2
CH1
m
S Slave Addr Ack Sub-Addr Ack 0000 s xxx
b
m
0000 s xxx
b
Mix_j_to_i
Dynamic
Range Control
Loudness
m
Σ
j
Ack
Dynamic
Range Control
Loudness
k
Σ
Mix_v_to_k
i
Σ
l
Ack xxxxxx s
b
Loudness
CH 1 Soft
Volume
xxxxxxxx
Σ
v
Σ
u
Ack
Mix_u_to_i
CH 1
Bass and Treble
Block
m
Ack Sub-Addr Ack 0000 s xxx
b
0x76 = Mix u to i
Slave Addr
CH1
S
Dynamic Range Control (DRC) Mixer Coefficients
xxxxxxxx
Ack
Ack
Ack
CH1− 0x7E
CH2− 0x7D
CH1− 0x7C
xxxxxxxx
Ack
l
Ack xxxxxx s
b
l
Ack xxxxxx s
b
Ack
Ack
Word2
Word1
Word 2 = Mix n to q − Bypass
Word 1 = Mix n to q − Inline
Word 2 = Mix l to p − Bypass
Word 1 = Mix l to p − Inline
Word 2 = Mix j to o − Bypass
Word 1 = Mix j to o − Inline
xxxxxxxx
xxxxxxxx
1−13
Slave Addr
Ack
xxxxxxxx
Ack
xxxxxxxx
l
Ack xxxxxx s
b
Ack
xxxxxxxx
Ack
xxxxxxxx
0x82 = Mix Channel 3 Output to o − 28-Bit Coefficient
0x83 = Mix Channel 3 Output to p − 28-Bit Coefficient
m
Ack Sub-Addr Ack 0000 s xxx
b
Ack
Ack
l
Ack xxxxxx s
b
Channel 3 to Channel 1 and Channel 2 Mix Gain Coefficients
0x7F = Mix Dither 1 to o − 28-Bit Coefficient
0x80 = Mix Dither 2 to p − 28-Bit Coefficient
0x81 = Mix Dither 3 to q − 28-Bit Coefficient
m
Ack Sub-Addr Ack 0000 s xxx
b
Slave Addr
Channel 1
Channel 2
S
Channel 1
Channel 2
Channel 3
S
Dither Mix Gain Coefficients
SUBADDRESS(es)
Σ
Σ
Σ
Node q
Node p
Node o
Channel 3
Loudness / Soft Volume
Processed Output
Dynamic
Range Control
Channel 2
Loudness / Soft Volume
Processed Output
Dynamic
Range Control
Channel 1
Loudness / Soft Volume
Processed Output
32-Bit Truncate
Mix−Delay3_to_p
32-Bit Truncate
Mix−Delay3_to_o
32-Bit Truncate
Σ
Σ
Σ
q
p
o
Dither 1
Dither−Processed Audio Out _ CH 1
Delay 3
Delay 2
Delay 1
Channel 3
Processed Audio
Channel 2
Processed Audio
Channel 1
Processed Audio
Dither−Processed Audio Out _ CH 3
Dither 3
Mix_Dither3_to_q
Dither−Processed Audio Out _ CH 2
Dither 2
Mix_Dither2_to_p
Mix_Dither1_to_o
PARAMETER(s)
1−14
Volume
Commands
I2C Bus
Slave Addr
Ack Sub-Addr Ack
0xF2
CH1
S
S
xxxxxxxx
Ack xxxxxxxx
CCC
Ack xxxxxxxx Ack xxxxx HHH Ack
321
Ack xxxxxxxx
l
Ack xxxxxxx s Ack
b
ttransition
Commanded
Volume
48
AUDIO IN
= x16 BoostMAX
Volume Command = 1/223 Cut
MAX (LSB)
(5.23 Precision) = Zero Output For 0x0000000 Volume Control
Slave Addr
m
Ack Sub-Addr Ack 0000 s xxx Ack xxxxxxxx
b
Mute Command = 1 => 0x0000000 Volume Control
CH 1 = 0xF2
CH 2 = 0xF3
Volume
CH 3 = 0xF4
Command
Ack Sub-Addr Ack
Original
Volume
Note: Negative Volume Commands Result In Audio Polarity Inversion
Slave Addr
v
Ack xxxxxxx c Ack xxxxxxxx Ack
s
SOFT VOLUME
Ack xxxxxxxx
0xF1
0xF4
VCS = 0 ⇒ ttransition = 2048/FS
VCS = 1 ⇒ ttransition = 4096/FS
xxxxxxxx
0xF3
Mute / Unmute Command
0xF0
I2C Slave Mode
I2C Master Mode
Volume Commands − GPIO Terminals
GPIO0 − Volume Up − CH1 / CH2
GPIO1 − Volume Down − CH1 / CH2
GPIO2 − Volume Up − CH3
GPIO3 − Volume Down − CH1 / CH2
S
Volume Command
Parameter
Subaddress
CH2
CH3
Volume Command
S Mute/Unmute = 0xF0
S Volume Slew Command = 0xF1
Soft Volume
28
VCS
S
28
Ack xxxxxxxx
m
0000 s xxx Ack xxxxxxxx
b
a2
28
28
a1
Ack xxxxxxxx
Ack xxxxxxxx
m
0000 s xxx Ack xxxxxxxx
b
m
0000 s xxx Ack xxxxxxxx
b
Ack xxxxxxxx
Ack xxxxxxxx
( ) LG
m
0000 s xxx Ack xxxxxxxx
b
Σ
48
Σ
48
48
Ack
l
xxxxxx s Ack
b
LG
b2
l
xxxxxx s Ack
b
z−1
z−1
b1
l
xxxxxx s Ack
b
b0
a2
l
xxxxxx s Ack
b
l
xxxxxx s Ack
b
a1
l
xxxxxx s Ack
b
AUDIO OUT
Loudness Compensation
Ack
Ack
Ack
Ack
Ack
Σ
O
G
2LO
S
S
S
Slave Addr
Slave Addr
Slave Addr
xxxxxxxx
00000000
Ack Sub-Addr Ack
xxxxxxxx
00000000
0xB0
0xAF
0xAE
0xAD
0xAC
Ack xxxxxxxx
l
Ack xxxxxxx s Ack
b
LO LSBs
LO MSBs
l
Ack xxxxxxx s Ack
b
Ack xxxxxxxx
l
Ack xxxxxxx s Ack
b
O Is A 25.23 Format Number
Ack xxxxxxxx
m
Ack 00000000 Ack s xxxxxxx Ack xxxxxxxx Ack
b
CH 1 = 0xA5
CH 2 = 0xAA
CH 3 = 0xAF
G Is A 5.23 Format Number
Ack xxxxxxxx
CH 1 = 0xA4
CH 2 = 0xA9
CH 3 = 0xAE
0 LSBs
0 MSBs
G
LO Is A 25.23 Format Number
Ack xxxxxxxx
m
Ack 00000000 Ack s xxxxxxx Ack xxxxxxxx Ack
b
CH 1 = 0xA3
CH 2 = 0xA8
CH 3 = 0xAD
0xAB
0xAA
0xA9
0xA8
0xA7
Subaddress
CH2
CH3
m
Ack Sub-Addr Ack 0000 s xxx Ack xxxxxxxx
b
Ack Sub-Addr Ack
0xA5
0xA6
BiQuad
0xA4
0xA3
0xA2
CH1
Loudness
O
G
LO
LG
Parameter
LOUDNESS
LG Is A 5.23 Format Number
Ack xxxxxxxx
All biquad gain coefficients 5.23 numbers.
28
b2
b1
28
b0
Commanded 5.23
Volume Command
z−1
z−1
BiQuad Coefficients
CH 1 = 0xA6
CH 2 = 0xAB
CH 3 = 0xB0
m
Ack Sub-Addr Ack 0000 s xxx Ack xxxxxxxx
b
m
Ack Sub-Addr Ack 0000 s xxx Ack xxxxxxxx
b
Slave Addr
Slave Addr
S
CH 1 = 0xA2
CH 2 = 0xA7
CH 3 = 0xAC
Soft Volume and Loudness Subaddress
1−15
Audio Input
CH2
Audio Input
CH1 or CH3
5.23 Format
RMS
Voltage
Estimator
32
Comparator
T2-LSBits
25.23
Format
1−ae
ae
O1
{
T1
O2
T2
{
tWindow ≈ −1/[FS x ln(1−ae)] Where FS = Audio Sample Frequency
CH1/CH2 = 0xB4
CH3 = 0xB9
CH1/CH2 = 0xB3
CH3 = 0xB8
25.23
Format
CH1/CH2 = 0xB5
CH3 = 0xBA
td
ta ≈ −1/[FS x ln(1−aa)]
td ≈ −1/[FS x ln(1−ad)]
Cut
Attack / Decay Control
ta
l
Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxx s
b
l
Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxx s
b
l
Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxx s
b
l
Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxx s
b
Volume
5.23 Format
m
S Slave Addr Ack Sub-Addr Ack 0000 s xxx
b
m
0000 s xxx
b
m
0000 s xxx
b
m
0000 s xxx
b
5.23 Format
K2
K1
K0
m
S Slave Addr Ack Sub-Addr Ack 00000000 Ack 00000000 Ack s xxxxxxx Ack xxxxxxxx
b
l
xxxxxxxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxx s
b
m
00000000 Ack 00000000 Ack s xxxxxxx Ack xxxxxxxx
b
l
xxxxxxxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxx s
b
NOTE: Compression / Expansion / Compression Displayed
K0
K1
Compression / Expansion
Coefficient Computation
K2
ae and (1−ae) Set Time Window Over Which RMS Value is Computed
Applies to DRC Servicing CH1/CH2 Only
RMS
Voltage
Estimator
32
Ack
T2-MSBits
T1-LSBits
Ack
Ack
T1-MSBits
Ack
m
l
S Slave Addr Ack Sub-Addr Ack 0000 s xxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxx s Ack
b
b
m
l
0000 s xxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxx s Ack
b
b
m
l
0000 s xxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxx s Ack
b
b
m
S Slave Addr Ack Sub-Addr Ack 00000000 Ack 00000000 Ack s xxxxxxx Ack xxxxxxxx Ack
b
l
xxxxxxxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxx s Ack
b
CH1/CH2 = 0xB1
CH3 = 0xB6
xxxxxxxx Ack
00000000 Ack
xxxxxxxx Ack
S Slave Addr Ack Sub-Addr Ack 00000000 Ack
m
00000000 Ack s xxxxxxx Ack xxxxxxxx
b
l
xxxxxxxx Ack xxxxxxxx Ack xxxxxxx s
b
m
00000000 Ack s xxxxxxx Ack xxxxxxxx
b
l
xxxxxxxx Ack xxxxxxxx Ack xxxxxxx s
b
CH1/CH2 = 0xB2
CH3 = 0xB7
Subaddress — Dynamic Range Control (DRC) Block
1−aa
ad
1−ad
Ack
Ack
5.23 Format
28
aa
Ack
DRC-Derived
Gain Coefficient
O2-LSBits
O2-MSBits
O1-LSBits
O1-MSBits
Ack
Ack
Ack
Ack
Ack
1−16
Slave Addr
m
Ack Sub-Addr Ack 0000 s
b
m
0000 s
b
m
0000 s
b
m
0000 s
b
m
0000 s
b
xxxxxxxx
xxxxxxxx
xxxxxxxx
xxxxxxxx
xxxxxxxx
xxx Ack
xxx Ack
xxx Ack
xxx Ack
xxx Ack
Slave Addr
m
Ack Sub-Addr Ack 0000 s xxx Ack
b
m
0000 s xxx Ack
b
S
VU Meter Output = 0xFE
S
Slave Addr
Slave Addr
Ack Sub-Addr Ack
Ack Sub-Addr Ack
Ack
Ack
Ack
Ack
Ack
xxxxx.xxx
xxxxx.xxx
xxxxx.xxx
xxxxx.xxx
xxxxx.xxx
Ack
Ack
xxxxx.xxx
xxxxx.xxx
Ack
xxxxx.xxx
Ack
Ack
xxxxx.xxx
xxxxx.xxx
Ack
xxxxx.xxx
xxxxxxxx
Ack
Ack
xxxxxxxx
xxxxxxxx
xxxxxxxx
xxxxxxxx
xxxxxxxx
xxxxxxxx
xxxxxxxx
xxxxx.xxx
Ack
Ack
Ack
Ack
Ack
Ack
xxxxxxxx
Spectrum Analyzer Output Subaddress = 0xFD
S
RMS Window Time Constant Subaddress = 0xBB
S
Spectrum Analyzer/VU Meter
BiQuad 1 to 10 Subaddresses = 0xBC to 0xC5
Ack
Ack
Ack
Ack
Ack
VU Meter Output 1
(BiQuad 6)
VU Meter Output 1
(BiQuad 5)
BiQuad 10
BiQuad 9
BiQuad 8
BiQuad 7
BiQuad 6
BiQuad 5
BiQuad 4
RMS Voltage
Estimator
RMS Voltage
Estimator
RMS Voltage
Estimator
Log
Log
Log
Log
RMS Voltage
Estimator
tWindow ≈ −1/[FS x ln(1−asa)] Where FS = Audio Sample Frequency
asa and (1−asa) Set Time Window Over Which RMS Value Is
Computed
BiQuad 10
BiQuad 9
BiQuad 8
BiQuad 7
Log
Log
Log
Log
Log
Log
RMS Voltage
Estimator
RMS Voltage
Estimator
RMS Voltage
Estimator
BiQuad 4
BiQuad 5
RMS Voltage
Estimator
BiQuad 3
BiQuad 3
t
s
RMS Voltage
Estimator
RMS Voltage
Estimator
BiQuad 2
BiQuad 1
Spectrum Analyzer / VU Meter
BiQuad 6
1−asa
asa
b2
b1
b0
a2
a1
BiQuad 2
BiQuad 1
l
Ack xxxxxxx s Ack
b
l
Ack xxxxxxx s Ack
b
l
Ack xxxxxxx s
b
l
Ack xxxxxxx s
b
l
Ack xxxxxxx s
b
l
Ack xxxxxxx s
b
l
Ack xxxxxxx s
b
Spectrum Analyzer/VU Meter
Sub-Address Decod
coderr
I2C Bus
1−17
Σ
Σ
Σ
Dither 1
Dither 2
Dither 3
Slave Addr
Slave Addr
Ack
Ack
xxxxxxxx
xxxxxxxx
Ack Sub-Addr Ack
St6
St5
St4
St3
St2
St 1
p
00000000
Sampler
xxxxxxxx
xxxxxxxx
0.5
Ack
0xC7
0.25
00000000
L
O
G
I
C
Condensed
LFSR1
Seed
LFSR2
Dither Seed
Distribution 2 Mix
Distribution 1 Mix
m
l
l
m
Ack s xxxxxx s Ack s xxxxxx s Ack
b
b
b
b
Condensed
LFSR2
Seed
Seed Build Logic
− W 0 +W
Output
LFSR1
l
Ack xxxxxxx s Ack
b
l
Ack xxxxxxx s Ack
b
Linear Feedback Shift Register Block
LFSR1 Mix and LFSR2 Mix Are 5.23 Format Coefficients
NOTE: W = 16.0 => 0x000008000000 in 25.23 Format
S
S
m
Ack Sub-Addr Ack 0000 s xxx Ack
b
m
0000 s xxx Ack
b
0xC6
Dither Block
1−18
GPIO3
GPIO2
GPIO1
GPIO0
PWRDN
S Slave Addr Ack
LRCLK
Sub-Addr
24
0xEF
16
15
8
7
0
D
D
D
D
Q
Q
Q
Q
Sample
Logic
LD
Reset
S Slave Addr Ack
Reset
31
24
23
0xEE
16
15
Microprocessor
Control
Decode 216
8
23
16
15
8
7
0
7
0
3 2 1 0 Ack
4 3
GPIO_in_out
Microprocessor
Clock
1 (Default State)
Disables Watchdog
Timer
Ack 00000000 Ack 00000000 Ack 00000000 Ack 0000000x Ack
Watchdog
Counter
Sub-Addr
0xEB
31
24
Determines How Many Consecutive Logic 0 Samples
(Where Each Sample Is Spaced by GPIOFSCOUNT LRCLKs)
Are Required to Read a Logic 0 on a GPIO Input Port
Ack 00000000 Ack 00000000 Ack 00000000 Ack 0000
I2C Master
Mode Read
I2C Slave Mode
and
I2C Master Mode
Write
DATA PATH SWITCH
Microprocessor
Firmware
Sub-Addr
Decode 0
Down
Counter
S Slave Addr Ack
READ Microprocessor
EN Bus
Microprocessor
GPIODIR Ack GPIOFSCOUNT Ack GPIO_samp_int Ack
3 2 1 0
23 20 19
Reserved
Ack 00000000 Ack 0000
31
CH1/CH2 to CH3 After Effects Mixers
0xCA−0xCF
0xD0−0xD1
SDIN4 Input Mixers
0xC8−0xC9
0xD2−0xEA
PARAMETER(s)
Factory Test Subaddresses
SUBADDRESS(es)
GPIO and Watchdog Timer Subaddresses
1−19
S
Slave Addr
Bass
Filter Set N
Ack Sub-Addr Ack
0xF2−0xF4—See Subaddress 0xA2
tTransition = TBLC[7:0] x 1/LRCLK
0xF1
24
00000000
31
Ack
16
8
xxxxxxxx
0
Ack
CH1−CH3 Volume CMDS
tTransition = TBLC[7:0] x 1/LRCLK
Treble and Bass
Slew Rate
TBLC[7:0]
v
Ack 0000000 s Ack
c
15
Treble
Filter Set N
00000000
23
7
Master Mute/Un-Mute
0xF0—See Subaddress 0xA2
0xF1—Also See Subaddress 0xA2 and Subaddress 0xF5
Reserved/Factory Test Subaddresses
GPIO Port I/O Values and GPIO Parameters
0xEC−0xED
PARAMETER(s)
0xEE−0xEF—See Subaddress 0xEB
SUBADDRESS(es)
Subaddress—Bass and Treble Shelf Filter Parameters
Treble/Bass Slew Rate Selection
S
0xF1
Slave Addr
7
Ack Sub-Addr Ack
CH3
CH2
00000000
Ack
00000000
V
Ack 0000000 C Ack
S
S
Slave Addr
Ack Sub-Addr Ack
00000000
Ack
00000xxx
Ack
00000xxx
Ack
00000xxx
xxxxxxxx
Ack
Treble/Bass Slew Rate = TBLC
(Slew Rate = TBLC/FS,
Where FS = Audio Sample Rate)
CH1
Bass Filter Set Selection
0xF5
0
Ack
CH2
CH3
CH1
Treble Filter Set Selection
S
0xF7
Slave Addr
Ack Sub-Addr Ack
CH3
CH2
00000000
Ack
00000xxx
Ack
00000xxx
Ack
00000xxx
Ack
CH1
Bass Shelf Selection (Filter Index)
0xF6
S
Slave Addr
Ack Sub-Addr Ack
00000000
Ack
xxxxxxxx
Ack
xxxxxxxx
Ack
xxxxxxxx
Ack
CH3
CH1
CH2
Treble Shelf Selection (Filter Index)
0xF8
BASS
FILTER 5
BASS
FILTER 3
S
Slave Addr
Ack Sub-Addr Ack
BASS
FILTER 1
MAX BOOST
SHELF
00000000
TREBLE
FILTER 5
Ack
xxxxxxxx
Ack
xxxxxxxx
TREBLE
FILTER 3
Ack
xxxxxxxx
Ack
TREBLE
FILTER 1
MID-BAND
BASS
FILTER 4
MAX CUT
SHELF
BASS
FILTER 2
TREBLE
FILTER 4
TREBLE
FILTER 2
FREQUENCY
Treble & Bass Filter Set Commands
0 => No Change
1 − 5 => Filter Sets 1 − 5
6 − 7 => Illegal (Behavior Indeterminate)
Treble & Bass Filter Shelf Commands
0 => Illegal (Behavior Indeterminate)
1 − 150 => Filter Shelves 1 − 150
1
=> +18-dB Boost
150 => −18-dB Cut
151 − 255 => Illegal (Behavior Indeterminate)
3-dB CORNERS (kHz)
FS
(LRCLK)
FILTER SET 5
FILTER SET 4
FILTER SET 3
FILTER SET 2
FILTER SET 1
BASS
TREBLE
BASS
TREBLE
BASS
TREBLE
BASS
TREBLE
BASS
96 kHz
0.25
6
0.5
12
0.75
18
1
24
1.5
36
88.4 kHz
0.23
5.525
0.46
11.05
0.691
16.575
0.921
22.1
1.381
33.15
64 kHz
0.167
4
0.333
8
0.5
12
0.667
16
1
24
48 kHz
0.125
3
0.25
6
0.375
9
0.5
12
0.75
18
44.1 kHz
0.115
2.756
0.23
5.513
0.345
8.269
0.459
11.025
0.689
16.538
32 kHz
0.083
2
0.167
4
0.25
6
0.333
8
0.5
12
24 kHz
0.063
1.5
0.125
3
0.188
4.5
0.25
6
0.375
9
22.05 kHz
0.057
1.378
0.115
2.756
0.172
4.134
0.23
5.513
0.345
8.269
16 kHz
0.042
1
0.083
2
0.125
3
0.167
4
0.25
6
12 kHz
0.031
0.75
0.063
1.5
0.094
2.25
0.125
3
0.188
4.5
11.025 kHz
0.029
0.689
0.057
1.378
0.086
2.067
0.115
2.756
0.172
4.134
1−20
TREBLE
Master
SCL
I2C
÷10
1/(M+1)
3
2
÷4
0
m[3:0] n[2:0]
6
MUX
1
0
MICROCLK_DIV
MUX
0
1
2
3
MUX
7
6
5
÷512
SCLKOUT2
MUX
1
SCLKOUT1
MUX
0
MUX
7
6
5
÷16
÷32
÷8
4
÷4
÷256
÷384
3
4
2
÷2
1
0
÷192
1
27 26
24
÷4
÷2
XTALI
SCLKIN
MUX
0
1
ICS
23
0xF9
w[1:0] y[2:0] Ack
28
÷128
0
÷2
÷4
0
1
2
3
MUX
MCLKO
3
÷32
÷64
29
MUX
0
1
2
PLL[1:0]
PLL0
000
31
PLL
BYPASS
x11
PLL
PLL1
2
1
0
NOTE: F9 must not be updated without first muting all three monaural channels in the TAS3103.
See Section 2.1.1 for a detailed discussion of this restriction.
LRCLK
Digital Audio
Processor
Clock
S Slave Addr Ack Sub-Addr Ack
S Slave Addr Ack Sub-Addr Ack 00000000 Ack 00000000 Ack 00000000 Ack 0xxxxxxx Ack
0xFB
I2C
Module
÷2N
Microprocessor
Clock
I2C
Sampling
Clock
I2C_SCL I2C_SDA
÷2
0
0
1
1
0
0
1
1
0
0
0
0
1
1
1
1
independent.
16
0
0
0
1
1
1
1
0
0
0
0
1
1
1
1
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
15
8
8
OW[2:0]
11 10
IW[2:0]
13
7
0
1
1
1
0
0
1
0
0
1
0
1
1
1
0
0
IM1/OM1
0
1
0
1
0
1
0
0
1
1
1
0
0
1
0
1
IM0/OM0
0
7
4
OM[3:0]
Mode
3
0
6 Ch, Single Chip, 20 − Bit
6 Ch, Single Chip, Crystal (LJ)
6 Ch, Single Chip (LJ)
6 Ch, Single Chip, Crystal (I 2 S)
TDM_20Bit_6
TDM_I2S_8
TDM_I2S_6
TDM_I2S_4
Discrete, 16 − Bit Packed
TDM_LJ_8
TDM_LJ_6
TDM_LJ_4
Discrete, Left Justified
Discrete, Right Justified
2
Discrete, I S
Discrete, Left Justified
IM[3:0]
‡Input and output mode selections are independent.
IM2/OM2
0
32 Bit
32 Bit
32 Bit
20 Bit
24 Bit
18 Bit
16 Bit
32 Bit
Word Size
z[2:0] Ack DWFMT Ack IOM Ack
AB
14
DWFMT (Data Word Format)
15
1
1
0
0
1
0
1
0
IW0/OW0
†Input and output word sizes are
IW1/OW1
IW2/OW2
Word Size Code†
AB assigns TDM time slots for those TDM
outputs involving two TAS3103s. For these
output formats, one of the TAS3103 chips
must be defined as AB = 0. The other
TAS3103 chip must be defined as AB = 1.
Serial Audio Port (AP) Mode Code‡
÷16
÷32
÷8
÷4
19 18
x[2:0]
21
MCLKI
IM3/OM3
0
MUX
7
6
5
4
2
3
1
0
IMS
22
OSC
XTALO
CRYSTAL
I2S FORMAT, CLOCK MANAGEMENT, AND I2C M AND N ASSIGNMENTS
ÎÎ
1−21
Delay/Reverb Assignments
0xFA
Delay
S
Delay Channel 1 = 2 x {D1[11:0] + 1}
Slave Addr
Reverb
m
l
m
l
Ack Sub-Addr Ack 0000 s xxx Ack xxxxxxx s Ack 0000 s xxx Ack xxxxxxx s Ack D1 and R1
b
b
b
b
l
m
l
m
0000 s xxx Ack xxxxxxx s Ack 0000 s xxx Ack xxxxxxx s Ack D2 and R2
b
b
b
b
l
m
l
m
0000 s xxx Ack xxxxxxx s Ack 0000 s xxx Ack xxxxxxx s Ack D3 and R3
b
b
b
b
Note: 2 x (D1 + D2 + D3) + 3 x (R1 +R2 +R3) ≤ 4076
Delay Channel 2 = 2 x {D2[11:0] + 1}
Delay Channel 3 = 2 x {D3[11:0] + 1}
Reserved
Reverb Channel 1 = 2 x {R1[11:0] + 1}
Reverb Channel 2 = 2 x {R2[11:0] + 1}
Reverb Channel 3 = 2 x {R3[11:0] + 1}
Reserved
1−22
NOTE: Changes in reverb and delay assignments can result in unplesasant and extended audio artifacts.
It is recommended that the TAS3103 always be muted before making reverb and delay changes.
See section 3.6.3 for a detailed discussion of this restriction.
SUB-ADDRESS(ES)
0xFC—See Subaddress 0x00
PARAMETER(S)
Ending I2C Check Word
0xFD−0xFE—See Subaddress 0xBB
Spectrum Analyzer/VU Meter Ouputs
0xFF—Volume Busy Flag
Volume
Flag
S
Slave Addr
Ack Sub-Addr Ack
0000000x
S
S
Volume Flag = 0 ⇒ No volume commands are active.
Volume Flag = 1 ⇒ One or more volume commands are
active.
Ack
1−23
1−24
2 Hardware Architecture
Figure 2−1 depicts the hardware architecture of the chip. The architecture consists of five major blocks:
•
Input Serial Audio Port (SAP)
•
Output Serial Audio Port (SAP)
•
DPLL and Clock Management
•
Controller
•
Digital Audio Processor (DAP)
2−1
SCLKIN
MCLKI
XTALI
XTALO
MCLKO
÷2
PLL1 PLL0 SCLKOUT1
÷2
M
U
X
MCLK
÷Z
M
U
X
÷2
M
U
X
PLL
(x11)
÷4
M
U
X
M
U
X
÷X
M
U
X
÷Y
SDIN1
MICROCLK_DIV SCLKOUT2
M
U
X
OSC
÷2
LRCLK
ORIN
PLL and Clock Management
32 Bits
SDOUT1
32 Bits
4K x 16
Delay Line
RAM
32 Bits
Dual Port
Data RAM
Coefficient
RAM
Program
ROM
256 Bits
32 Bits
Arithmetic
Engine
SDIN2
DAP
Instruction Decoder/Sequencer
SDOUT2
76-Bit Adder
32 Bits
SDIN3
Regs
Regs
Digital Audio Processor
(DAP)
Arithmetic Unit
64 Bits
SDOUT3
32 Bits
Controller
SDIN4
64 Bits
Volume
Update
2K x 8
Data
RAM
32 Bits
16K x 8
Program
ROM
Reg
Reg
INPUT SAP
OUTPUT SAP
1/2N
I2C
Master/Slave
Controller
I2C_SDA I2C_SCL CS0
CS1
Oversample Clock
Master
SCL
I2CM_S
÷10
1/(M+1)
256 x 8
Data
RAM
8-Bit
WARP
8051 Microprocessor
GPIO0
Figure 2−1. TAS3103 Detailed Hardware Block Diagram
2−2
GPIO1
GPIO2
GPIO3
2.1 Input and Output Serial Audio Ports (SAPs)
The TAS3103 accepts data in various serial data formats including left/right justified and I2S, 16 through 32 bits,
discrete, or TDM. Sample rates from 8 kHz through 96 kHz are supported. Each TAS3103 has four input serial ports
and three output serial ports, labeled SDIN[4:1] and SDOUT[3:1] respectively. All ports accommodate stereo data
formats, and SDIN1 and SDOUT1 also accommodate time-division multiplex (TDM) data formats. The formats are
selectable via I2C commands. All input channels are assigned the same format and all output channels are assigned
the same format; and the two formats need not be the same. The TAS3103 can accommodate system architectures
that require data format conversions without the need for additional glue logic. If a TDM format is selected for the input
port, only SDIN1 is active; the other three input channels cannot be used. If a TDM format is selected for the output
port, only SDOUT1 is active; the other two channels cannot be used.
2.1.1
SAP Configuration Options
The TAS3103 serial interface data format options for discrete (stereo) data are detailed in Figure 2−2.
Left / Right
Justified
LRCLK
I2S
SCLK
Bit 0
Bit 31
Left Justified
MSB
LSB+1
MSB−1
LSB
Bit 31
MSB
Bit 0
LSB
MSB−1
Bit 31
MSB
MSB−1
SCLK
Bit 31
Bit 0
Right Justified
MSB
MSB−1
LSB+1
Bit 31
Bit 0
LSB
MSB
MSB−1
LSB+1
Bit 31
LSB
SCLK
Bit 0
I2S
Bit 31
MSB
Bit 0
MSB−1
LSB
LSB+1
Bit 31
MSB
Bit 0
MSB−1
LSB
LSB+1
Bit 31
MSB
Figure 2−2. Discrete Serial Data Formats
When the TAS3103 is transmitting serial data, it uses the negative edge of SCLK to output a new data bit. The
TAS3103 samples incoming serial data on the rising edge of SCLK.
The TDM modes on the TAS3103 only provide left justified and I2S formats, and each word in the TDM data stream
adheres to the bit placement shown in Figure 2−2. Figure 2−3 illustrates the output data stream for a 4-channel TDM
mode. Two cases are illustrated; an I2S data format case (SAP output mode 1010) and a left-justified data format case
(SAP output mode 0111).
LRCLK
Left Channel 1
Left
Justified
Data
MSB
MSB−1
Left Channel 2
LSB
MSB
Right Channel 1
LSB
MSB−1
MSB
Right Channel 2
LSB
MSB−1
MSB
LSB
MSB−1
MSB
LRCLK
I2S
Left Channel 1
Data
MSB
MSB−1
LSB
Left Channel 2
MSB
MSB−1
Right Channel 1
LSB
MSB
MSB−1
Right Channel 2
LSB
MSB
MSB−1
LSB
Figure 2−3. Four-Channel TDM Serial Data Formats
2−3
A 16-bit field contained in the 32-bit word located at I2C subaddress 0xF9 configures both the input and output serial
audio ports. Figure 2−4 illustrates the format of this 16-bit field. The data is shown in the transmitted I2C protocol
format, and thus, in addition to the data, the start bit S, the slave address, the subaddress, and the acknowledges
required by every byte are also shown.
Output Port
TDM Alignment
Î
0xF9
15
14
AB
31
24
23
16
Input Port
Word
Size
13
Output Port
Word
Size
11 10
8
IW[2:0]
OW[2:0]
15
8
DWFMT (Data Word Format)
7
0
S Slave Addr Ack Sub-Addr Ack xxxxxxxx Ack xxxxxxxx Ack DWFMT Ack IOM Ack
7
4
3
0
IM[3:0]
OM[3:0]
Input
Port
Format
Output
Port
Format
Figure 2−4. SAP Configuration Subaddress Fields
Commands to reconfigure the SAP cannot be issued as standalone commands, but must accompany mute and
unmute commands. The reason for this is that an SAP configuration change while a volume or bass or treble update
is taking place can cause the update to not properly be completed. Figure 2−5 shows the recommended procedure
for issuing SAP configuration update commands.
After a reset, ensure that 0xF9 is written before volume, treble, or bass.
2−4
Enter
Yes
Vol
Busy
No
Issue Mute Command
Yes
Vol
Busy
No
Issue SAP Configuration
Change Command
Issue Un-Mute Command
Yes
Vol
Busy
No
Mute Command = 0x00000007 at subaddress x0F0
Un-mute command = 0x00000000 at subaddress 0xF0
SAP configuration subaddress = 0x59
Volume busy flag = LSB of subaddress 0xFF. Logic 1 = busy
Exit
Figure 2−5. Recommended Procedure for Issuing SAP Configuration Updates
Figure 2−6, Figure 2−7, and Figure 2−8 tabularize the formatting and word size options available for the input SAP
and output SAP. In these figures, data formats are paired when the only difference between the pair is whether the
word placement within the LRCLK period is left justified or I2S. The TDM formats available include single chip TDM
output formats (SDOUT1chipA or SDOUT1chipB) and two chip TDM output formats (SDOUT1chipA OR’ed with
SDOUT1chipB). For two chip TDM output formats, the OR’ing operation is accomplished by routing SDOUT1 from
one of the two TAS3103 chips to ORIN of the other TAS3103 chip. The AB bit also comes into play for two chip TDM
formats; AB must be set to 1 on one of the two TAS3103 chips and set to 0 on the other TAS3103 chip. For a given
connection of SDOUT1 to ORIN, it does not matter which TAS3103 is set up as chip AB = 1 and which chip is set
up as chip AB = 0. However, the routing of the processed data to the output registers in the TAS3103 is dependent
on which chip is chip AB = 0 and which chip is chip AB = 1. Figure 2−10 and Figure 2−11 illustrate this dependence.
2−5
In Figure 2−10 and Figure 2−11, the paired TDM output formats 0101 and 1000 are unique in that each format, in
effect, services two distinct industry formats. For these two modes, if register Y in chip AB = 1 is set to zero (by
appropriate output mixer coefficient settings), the resulting format is a standard 8 CH TDM format. This option is
illustrated in Figure 2−7.
2−6
2−7
All options valid
All options valid except 32 bit
8 CH transfer, left
justified
8 CH transfer, I2S
6 CH, left justified
0101
1000
0110/1101
(2)(3)
All options valid
All options valid except 32 bit
IW[2:0] = 011
All options valid
All options valid except 32 bit
4 CH, left justified
4 CH, I2S
6 CH, 20 bit
6 CH data, 8 CH
transfer, left justified
6 CH data, 8 CH
transfer, I2S
1010
1011/1111(4)
1100
1110
All options valid except 32 bit
0111
6 CH, I2S
IW[2:0] = 001
16-bit packed
0100
1001(3)
All options valid except 32 bit
I2S
0011
All options valid
All options valid
Right justified
0010
All options valid
WORDSIZE
Left justified
FORMAT
000X(1)
IM[3:0]
E
C
32
E
32
20
20
L
C
32
E
E
32
G
32
L
C
32
A
L
32
A
20
A
32
4
A
32
C
L
32
32
32
C
L
32
32
F
32
H
16
16
A
B
R
B
R
20
F
32
D
32
F
32
A
L
32
A
L
SDIN1
F
32
D
R
32
R
20
D
R
32
B
32
D
R
R
D
32
B
32
B
20
B
32
B
32
4
32
32
D
R
Not available
Not available
Not available
Not available
Not available
Not available
Not available
Not available
Not available
Not available
32
C
L
SDIN2
32
F
R
Not available
Not available
Not available
Not available
Not available
Not available
Not available
Not available
Not available
Not available
32
E
L
SDIN3
DATA DISTRIBUTION: A, B, C, D, E, F, G, H = INPUT MIXER INPUTS
Figure 2−6. Format Options: Input Serial Audio Port
NOTES: 1. Left justified, stereo is the default input format.
2. IM[3:0] modes 0110 and 1101 are identical for the input SAP. OM[3:0] modes 0110 and 1101 do produce different results in the output SAP (see Figure 2−7).
3. If a 6 CH input format is selected, the output format must also be set to 6 CH. When in a 6 CH mode, data format selections—I2S and left justified—for the 6 Ch input
SAP can be made independent of the data format selections—I2S and left justified—made for the 6 CH output SAP.
4. IM[3:0] modes 1011 and 1111 are identical for the input SAP. OM[3:0] modes 1011 and 1111 do produce different results in the output SAP (see Figure 2−7).
TIME
DIVISION
MULTIPLEX
(TDM)
DISCRETE
INPUT
TYPE
2−8
All options valid
All options valid
All options valid except 32 bit
6 CH, 2 chip, left
justified
6 CH, 2 chip, I2S
4 CH, left justified
4 CH, I2S
0110(1)
1001(1)
0111
1010
All options valid
All options valid except 32 bit
All options valid
6 CH data,
8 CH transfer,
left justified
6 CH data,
8 CH transfer,
I2S
6 CH,
left justified
6 CH, 20 bit
1100
1110
1101(1)
1111
OW[2:0] = 011
OW[2:0] = 011
6 CH, 2 chip,
20 bit
1011
All options valid except 32 bit
8 CH, 2 chip, I2S
1000
All options valid except 32 bit
All options valid
8 CH, 2 chip, left
justified
0101
WORDSIZE
FORMAT
OM[3:0]
AB = 0
L
L
20
V
20
20
20
W
32
32
V
V
L
32
U
32
20
U
L
4
32
W
32
W
20
V
20
20
V
32
U
32
32
L
32
U
L
32
32
U
32
4
4
W
20
X
W
32
32
W
20
32
32
R
Y
R
X
Y
X
R
32
Y
20
Z
R
R
32
Z
32
Y
20
V
20
32
W
32
32
32
32
X
20
20
W
R
32
32
32
32
32
V
32
32
V
32
U
32
20
U
32
32
U
32
32
V
U
U
AB = 1
AB = 0
AB = 1
AB = 0
AB = 1
AB = 0
L
SDOUT1
R
32
Z
20
W
20
4
32
W
32
32
Y
32
X
4
4
32
32
X
32
Not available
Not available
Not available
Not available
Not available
Not available
Not available
Not available
Not available
Not available
Not available
SDOUT2
Not available
Not available
Not available
Not available
Not available
Not available
Not available
Not available
Not available
Not available
Not available
SDOUT3
DATA DISTRIBUTION: U, V, W, X, Y, Z = OUTPUT MIXER OUTPUTS
Figure 2−7. TDM Format Options: Output Serial Audio Port
NOTE 1: If a 6 CH output format is selected, the input format must also be set to 6 CH. When in a 6 CH mode, data format selections—I2S and left justified—for the output SAP can
be made independent of the data format selections—I2S and left justified—can be made for the input SAP.
TIME
DIVISION
MULTIPLEX
(TDM)
OUTPUT
TYPE
2−9
OW[2:0] = 001
0011
16-bit packed
All options valid except 32 bit
I2S
0100
All options valid
Right justified
0010
All options valid
WORDSIZE
Left justified
FORMAT
000X(1)
OM[3:0]
V
16
16
R
L
U
32
V
R
32
U
L
SDOUT1
0
0
1
1
1
1
18 Bit
20 Bit
24 Bit
32 Bit
(32)
(32)
0
0
(32)
16 Bit
1
1
0
0
1
1
0
0
IW1
1
0
1
0
1
0
1
0
IW0
INPUT IW[2:0]
IW2
1
1
1
1
0
0
0
0
1
1
0
0
1
1
0
0
OW1
1
0
1
0
1
0
1
0
OW0
OUTPUT OW[2:0]
OW2
Figure 2−9. Word Size Settings
(32) ⇒ Reserved for future family members. Selection of 000, 110, or 111 in the
TAS3103 selects a 32-bit sample size.
DEFAULT →
SAMPLE
SIZE
32
X
R
Not available
32
W
L
SDOUT2
32
Z
R
Not available
32
Y
L
SDOUT3
DATA DISTRIBUTION: U, V, W, X, Y, Z = OUTPUT MIXER OUTPUTS
Figure 2−8. Discrete Format Options: Output Serial Audio Port
NOTE 1: Left justified, stereo is the default input format.
DISCRETE
OUTPUT
TYPE
TAS3103
SDOUT1
U1
(AB = ’1’)
LRCLK
L
LRCLK
R
L
R
ORIN
U2
U
32
V
32
32
W
32
32
X
SDOUT1U1
32
32
32
W
0
X
32
32
32
UU2
UU1
VU2
VU1
WU2
WU1
XU2
XU1
32
32
32
32
32
32
32
32
TAS3103
SDOUT1
U1
U
32
V
32
32
32
32
U2
(AB = ’0’)
ORIN
Figure 2−10. 8 CH TDM Format Using SAP Modes 0101 and 1000
TAS3103
SDOUT1
U1
(AB = 0)
LRCLK
L
LRCLK
R
L
R
ORIN
U1
U
V
32
32
32
32
W
32
32
32
32
0
32
SDOUT1U1
32
32
32
W
Y
0
32
32
32
32
UU1
UU2
VU1
32
32
32
32
WU1
WU2
YU2
32
32
32
32
TAS3103
SDOUT1
U2
U
0
U2
(AB = 1)
ORIN
Figure 2−11. 6 CH Data, 8 CH Transfer TDM Format Using SAP Modes 0101 and 1000
For these same two modes, if register X in chip AB = 0 is set to zero, and registers V and X in chip AB = 1 are set
to zero, the resulting format is a 6 CH data, 8 CH transfer format. This option is shown in Figure 2−11.
The data output format in Figure 2−11 is identical to that realized using data output formats 1100 and 1110 in
Figure 2−7. The difference is that SAP modes 1010 and 1000 provide six independent monaural channels to process
the data, whereas SAP modes 1100 and 1110 provide only three independent monaural channels to process the data.
2.1.2
Processing Flow—SAP Input to SAP Output
All SAP data format options other than I2S result in a two-sample delay from input to output, as illustrated in
Figure 2−12. Figure 2−12 is also relevant if I2S formatting is used for both the input SAP and the output SAP (the
polarity of LRCLK in Figure 2−12 has to be inverted in this case). However, if I2S format conversions are performed
between input and output, the delay becomes either 1.5 samples or 2.5 samples, depending on the processing clock
frequency selected for the digital audio processor (DAP) relative to the sample rate of the incoming data. The input
to output delay for an I2S input format and a non-I2S output format is illustrated in Figure 2−13(a), and Figure 2−13(b)
illustrates the delay for a non-I2S input format and an I2S output format. In each case, two distinct input to output delay
times are shown: a 1.5 sample delay time if the processing time in the DAP is less than half the sample period, and
a 2.5 sample delay time if the processing time in the DAP is greater than half the sample period.
The departure from the two-sample input to output processing delay when I2S format conversions are performed is
due to the use of a common LRCLK. The I2S format uses the falling edge of LRCLK to begin a sample period, whereas
all other formats use the rising edge of LRCLK to begin a sample period. This means that the input SAP and digital
audio processor (DAP) operate on sample windows that are 180° out of phase with respect to the sample window
used by the output SAP. This phase difference results in the output SAP outputting a new data sample at the midpoint
of the sample period used by the DAP to process the data. If the processing cycle completes all processing tasks
before the midpoint of the processing sample period, the output SAP outputs this processed data. However, if the
processing time extends past the midpoint of the processing sample period, the output SAP outputs the data
processed during the previous processing sample period. In the former case, the delay from input to output is 1.5
samples. In the latter case, the delay from input to output is 2.5 samples.
2−10
2−11
SDIN4
SDIN3
SDIN2
SDIN1
SDIN4
SDIN3
SDIN2
SDIN1
G
E
C
A
Serial Input
Rx
Holding
Regs
Regs
H
F
D
B
Input
Holding
Regs
H
F
D
B
Input
Holding
Regs
Sample Time N
G
E
C
A
Serial Input
Rx
Holding
Regs
Regs
Sample Time N
SDIN1
Sample Time N
G
E
C
H
F
D
Channel 3
Channel 2
Channel 3
Channel 2
Channel 1
Z
Y
X
W
V
SDIN4
SDIN3
SDIN2
G
E
C
H
F
D
B
Input
Holding
Regs
Figure 2−12. SAP Input-to-Output Latency
SDOUT3
SDOUT2
A
Serial Input
Rx
Holding
Regs
Regs
Channel 3
Channel 2
Channel 1
Sample Time N + 1
SDOUT1
SDIN4
SDIN3
SDIN2
B
Channel 1
Sample Time N + 2
Sample Time N + 2
SDOUT3
SDOUT2
A
Sample Time N + 1
U
Z
Y
X
W
V
U
SDIN1
Sample Time N + 1
Channel 3
Channel 2
Channel 1
SDOUT1
Input
Holding
Regs
Sample Time N + 1
Serial Input
Rx
Holding
Regs
Regs
2nd Half − Sample Time N
Sample Time N
Sample Time N + 1
Sample Time N + 2
1st Half − Sample Time N
Z
Y
X
W
V
U
Z
Y
X
W
V
U
Sample Time N + 2
Sample Time N + 2
SDOUT3
SDOUT2
SDOUT1
SDOUT3
SDOUT2
SDOUT1
LRCLK
SDIN
L2
R3
L4
L2, R2
R4
L3, R3
Load Output Holding Registers
Holding Register ⇒ Output Serial Registers
L0
SDOUT
Processing Cycle
1.5
Cycle
Delay
L3
L1, R1
Processing Cycle
2.5
Cycle
Delay
R2
L1, R1
R0
L1
R1
L2
L3, R3
L2, R2
Load Output Holding Registers
Holding Register ⇒ Output Serial Registers
SDOUT
R0
L1
R1
L2
R2
L3
(a) Left-Justified Input / I2S Output
LRCLK
SDIN
Processing Cycle
2.5
Cycle
Delay
L2
L3
L1, R1
R3
L2, R2
L4
R4
L3, R3
Load Output Holding Registers
Holding Register ⇒ Output Serial Registers
L0
SDOUT
Processing Cycle
1.5
Cycle
Delay
R2
L1, R1
R0
L1
R1
L2
L3, R3
L2, R2
Load Output Holding Registers
Holding Register ⇒ Output Serial Registers
SDOUT
R0
L1
R1
L2
R2
L3
(b) I2S Input / Left-Justified Output
Figure 2−13. SAP Input-to-Output Latency for I2S Format Conversions
The delay from input to output can thus be either 1.5 or 2.5 sample times when data format conversions are performed
that involves the I2S format. However, which delay time is obtained for a particular application is determinable and
fixed for that application, providing care is taken in the selection of MCLKI/XTALI with respect to the incoming sample
clock LRCLK.
2−12
Table 2−1 lists all viable clock selections for a given audio sample rate (LRCLK). The table only includes those clock
choices that allow enough processing throughput to accomplish all tasks within a given sample time (Ts = 1/LRCLK).
For each entry in the table, the DAP processing time is given in terms of whether the time is greater than 0.5 Ts
(resulting in an input to output delay of 2.5 Ts ), or less than 0.5 Ts (resulting in an input to output delay of 1.5 Ts ).
Table 2−1 is valid for both master and slave I2S modes (bit IMS at subaddress 0xF9 determines I2S master/slave
selection—see the DPLL and Clock Management section that follows). For all applications, MCLK must be ≥128
LRCLK (FS). In the I2S master mode, MCLK, SCLK (I2S bit clock) LRCLK are all harmonically related. Furthermore,
in the I2S master mode, if a master clock value given in Table 2−1 is used, the latency realized in performing I2S format
conversions, 1.5 samples or 2.5 samples, is stable and fixed over the duration of operation. However, greater care
must be taken for the I2S slave mode. In this mode, the device has the proper operational throughput to perform all
required computations as long as MCLK is ≥128 LRCLK. But there is no longer the requirement that MCLK be
harmonically related to SCLK and LRCLK. Values of MCLK could be chosen such that the output dithers between
latencies of 1.5 and 2.5 sample times. There may be cases where part of the data stream output exhibits sample time
latencies of 1.5 Ts and the other portion of the output data stream exhibits sample time latencies of 2.5 Ts . To assure
that such cases do not happen in the I2S slave mode, the relationships between MCLK and LRCLK given in Table 2−1
should be followed for data format conversions involving the I2S format. The MCLKI/XTALI frequencies given in
Table 2−1 (if set to within ±5% of the nominal value shown) assure that the DAP processing time falls above 0.5 Ts
or below 0.5 Ts with enough margin to assure that there is no race condition between the outputting of data and the
completion of the processing tasks.
Table 2−1. TAS3103 Throughput Latencies vs MCLK and LRCLK
AUDIO
SAMPLE RATE
(LRCLK)
96 kHz
88.2 kHz
48 kHz
44.1 kHz
32 kHz
24 kHz
22.05 kHz
8 kHz
DAP(1) CLOCK
(PLL_OUTPUT)
DAP CLOCK
CYCLES/LRC
LK
DAP
PROCESSING
TIME
THROUGHPU
T DELAY
24.576 MHz, 12.288 MHz
135.168 MHz
1408
> Ts/2
2.5 Ts
22.5792 MHz, 11.2896 MHz
124.1856 MHz
1408
> Ts/2
2.5 Ts
24.576 MHz, 12.288 MHz
135.168 MHz
2816
< Ts/2
1.5 Ts
MASTER CLOCK(2)
(MCLKI/XTALI)
24.576 MHz, 12.288 MHz, 6.144 MHz
67.584 MHz
1408
> Ts/2
2.5 Ts
22.5792 MHz, 11.2896 MHz
124.1856 MHz
2816
< Ts/2
1.5 Ts
22.5792 MHz, 11.2896 MHz, 5.6448 MHz
62.0928 MHz
1408
> Ts/2
2.5 Ts
16.384 MHz, 8.192 MHz
90.112 MHz
2816
< Ts/2
1.5 Ts
16.384 MHz, 8.192 MHz, 4.096 MHz
45.056 MHz
1408
> Ts/2
2.5 Ts
24.576 MHz, 12.2858 MHz
135.168 MHz
5632
< Ts/2
1.5 Ts
24.576 MHz, 12.2858 MHz, 6.144 MHz
67.584 MHz
2816
< Ts/2
1.5 Ts
12.288 MHz, 6.144 MHz, 3.072 MHz
33.792 MHz
1408
> Ts/2
2.5 Ts
22.5792 MHz, 11.2896 MHz
124.1856 MHz
5632
< Ts/2
1.5 Ts
22.5792 MHz, 11.2896 MHz, 5.6448 MHz
62.0928 MHz
2816
< Ts/2
1.5 Ts
11.2896 MHz, 5.6448 MHz, 2.8224 MHz
31.0464 MHz
1408
> Ts/2
2.5 Ts
24.576 MHz, 12.288 MHz
135.168 MHz
16896
< Ts/2
1.5 Ts
24.576 MHz, 12.288 MHz, 6.144 MHz
67.584 MHz
8448
< Ts/2
1.5 Ts
12.288 MHz, 6.144 MHz, 3.072 MHz
33.792 MHz
4224
< Ts/2
1.5 Ts
6.144 MHz, 3.072 MHz, 1.536 MHz
16.896 MHz
2112
> Ts/2
2.5 Ts
NOTES: 1. DAP clock is the internal digital audio processor clock. It is equal to 11 × MCLK1/XTALI, 11/2 × MCLKI/XTALI, or 11/4 × MCLKI/XTALI
(as determined by a bit field in I2C subaddress 0xF9). The DAP clock must always be greater than or equal to 1400 FS (LRCLK).
2. Unless in PLL bypass, MCLKI must be ≤ 20 MHz.
3. XTALI must always be ≤ 20 MHz.
2−13
2.2 DPLL and Clock Management
Clock management for the TAS3103 consists of two control structures:
•
Master clock management: oversees the selection of the clock frequencies for the microprocessor, the I2C
controller, and the digital audio processor (DAP). The master clock (MCLKI or XTALI) serves as the source
for these clocks. In most applications, the master clock is input to an on-chip digital phase lock loop (DPLL),
and the DPLL output is used to drive the microprocessor and DAP clocks. A DPLL bypass mode can also
be used, in which case the master clock is used to drive the microprocessor and DAP clocks.
•
Serial audio port (SAP) clock management: oversees SAP master/slave mode, the settings of SCLKOUT1
and SCLKOUT2, and the setting of LRCLK in the SAP master mode.
Figure 2−14 illustrates the clock circuitry in the TAS3103. The bold lines in Figure 2−14 highlight the default settings
at power turn on, or after a reset. Inputs MCLKI and XTALI source the master clock for the TAS3103. Within the
TAS3103, these two inputs are combined by an OR gate, and thus only one of these two sources can be active at
any one time. The source that is not active must be set to logic 0. In normal operation, the master clock is divided
by 1, 2, or 4 (as determined by the logic levels set at input pins PLL0 and PLL1) and then multiplied by 11 in frequency
by the on-chip DPLL. The DPLL output (or MCLKI/XTALI if the DPLL is bypassed) is the processing clock used by
the digital audio processor (DAP).
The DAP processing clock can also serve as the clock for the on-chip microprocessor, or the DAP clock can be divided
by four prior to sending it to the microprocessor. The input pin MICROCLK_DIV makes this clock choice. A logic 1
input level on this pin selects the DAP clock for the microprocessor clock; a logic 0 input level on this pin selects the
DAP clock/four for the microprocessor clock.
The selected microprocessor clock is also used to drive the clocks used by the I2C control block. Two parameters,
N and M, define the clocks used by the I2C control block. The I2C control block sampling frequency is set by 1/2N,
where N can range in value from 0 to 7. A 1/(1 + M) divisor followed by a 1/10 divisor generates the data bit clock
(SCL). This drived SCL clock is only used when the I2C control block is set to master mode (input pin I2CM_S = 1).
The default value for the I2C parameter N depends on whether the I2C controller is in a slave mode (I2CM_S = 0)
or a master mode (I2CM_S = 1). In the I2C master mode N = 2 (2N = 4), which assures that a 100-kHz I2C data clock
(SCL) can be generated when the digital audio processor (DAP) is running at its maximum frequency of 135 MHz.
In the I2C slave mode N = 1 (2N = 2), which assures the I2C controller an adequate over-sampling clock when the
DAP is running at the minimum clock frequency required to process 8-kHz audio data (approximately 11.2 MHz). In
I2C master mode, the values for M and N are fixed and cannot be changed.
2−14
SCLKIN
MCLKI
XTALI
MCLKO
XTALO
÷2
PLL1 PLL0 SCLKOUT1
÷2
MCLK
÷2
MICROCLK_DIV SCLKOUT2
M
U
X
OSC
÷ Z = 2DEFAULT
LRCLK
M
U
X
÷2
M
U
X
PLL
(x11)
M
U
X
÷4
M
U
X
M
U
X
÷ X = 1DEFAULT
M
U
X
÷Y = 64DEFAULT
PLL and Clock Management
Input
SAP
Digital Audio Processor
(DAP)
Output
SAP
Microprocessor
and
I2C Bus Controller
I2C
Master/Slave
Controller
I2C_SDA I2C_SCL
N = 1 (I2C Slave Default)
= 2 (I2C Master Default)
1/2N
Oversample Clock
Master
SCL
÷10
1/(M+1)
8-Bit
WARP
8051 Microprocessor
I2CM_S
Figure 2−14. DPLL and Clock Management Block Diagram
When the SAP is in the master mode, the serial audio port (SAP) uses the MCLKI/XTALI master clock to drive the
serial port clocks SCLKOUT1, SLCKOUT2, and LRCLK. When the SAP is in the slave mode, LRCLK is an input and
SCLKOUT2 and SCLKOUT1 are derived from SCLKIN. As shown in Figure 2−14, SCLKOUT1 clocks data into the
input SAP and SCLKOUT2 clocks data from the output SAP. Two distinct clocks are required to support TDM to
discrete and discrete to TDM data format conversions. Such format conversions also require that SCLKIN be the
higher valued bit clock frequency. For TDM in/discrete out format conversions, SCLKIN must be equal to the input
bit clock. For discrete in/TDM out format conversions, SCLKIN must be equal to the output bit clock. The frequency
settings for SCLKOUT1, SCLKOUT2, and LRCLK in the SAP master mode, as well as the SAP master/slave mode
selection, are all controlled by I2C commands.
Table 2−2 lists the default settings at power turn on or after a received reset.
2−15
Table 2−2. TAS3103 Clock Default Settings
CLOCK
DEFAULT SETTING
SCLKOUT1
SCLKIN
SCLKOUT2
SCLKIN
LRCLK
Input
MCLKO
MCLKI or XTALI
DAP processing clock
Set by pins PLL0 and PLL1
Microprocessor clock
I2C sampling clock
Set by pin MICROCLK_DIV
I2C master mode
Microprocessor clock/4
I2C slave mode
Microprocessor clock/2
I2C sampling clock/90
I2C master SCL
The selections provided by the dedicated TAS3103 input pins and the programmable settings provided by I2C
subaddress commands give the TAS3103 a wealth of clocking options. Table 2−1, in the section describing the serial
audio port (SAP), lists typical clocking selections for different audio sampling rates. However, the following clocking
restrictions must be adhered to:
•
MCLKI or XTALI ≥ 128 FS (NOTE: For some TDM modes, MCLKI or XTALI must be ≥ 256 FS)
•
DAP clock ≥ 1400 x FS
•
DAP clock < 136 MHz
•
Microprocessor clock/20 ≥ I2C SCL clock
•
Microprocessor clock ≤ 35 MHz
•
I2C oversample clock/10 ≥ I2C SCL clock
•
XTALI ≤ 12.288 MHz
•
MCLKI ≤ 25 MHz, unless PLL is bypassed
As long as these restrictions are met, all other clocking options are allowed.
2.3 Controller
The controller serves as the interface between the digital audio processor (DAP), the asynchronous I2C bus interface,
and the four general-purpose I/O (GPIO) pins. Included in the controller block is an industry-standard 8051
microprocessor and an I2C master/slave bus controller.
2.3.1
8051 Microprocessor
The 8051 microprocessor receives and distributes I2C write data, retrieves and outputs to the I2C bus controller the
required I2C read data, and participates in most processing tasks requiring multiframe processing cycles. The
microprocessor also controls the flow of data into and out of the GPIO pins, which includes volume control when in
the I2C master mode The microprocessor has its own data RAM for storing intermediate values and queuing I2C
commands, and a fixed program ROM. The microprocessor’s program cannot be altered.
2.3.2
I2C Bus Controller
The TAS3103 has a bidirectional, two-wire, I2C-compatible interface. Both 100K-bps and 400K-bps data transfer
rates are supported, and the TAS3103 controller can serve as either a master I2C device or a slave I2C device.
Master/slave operation is defined by the logic level input into pin I2CM_S (logic 1 = master mode, logic 0 = slave
mode).
If this input level is changed, the TAS3103 must be reset.
2−16
In the I2C master mode, data rate transfer is fixed at 100 kHz, assuming MCLKI or XTALI = 12.288 MHz, PLL0 = PLL1
= 0, and MICROCLK_DIV = 0. In the I2C slave mode, data rate transfer is determined by the master device. However,
the setting of I2C parameter N at subaddress 0xFB (see the PLL and Clock Management section) does play a role
in setting the data transfer rate. In the I2C slave mode, bit rates other than (and including) the I2C-specific 100K-bps
and 400K-bps bit rates can be obtained, but N must always be set so that the over-sample clock into the I2C
master/slave controller is at least a factor of 10 higher in frequency than SCL.
The I2C communication protocol for the I2C slave mode is shown in Figure 2−15.
Start
(By Master)
Read or Write
(By Master)
Stop
(By Master)
Slave Address
(By Master)
S
0
1
1
0
1
Data Byte
(By Transmitter)
C C R A
S S
/ C
1† 0† W K
M
S
B
Acknowledge
(By TAS3103)
I2C_SDA
MSB
MSB−1 MSB−2
Data Byte
(By Transmitter)
L
S
B
A
C
K
M
S
B
Acknowledge
(By Receiver)
L
S
B
A
C
K
S
Acknowledge
(By Receiver)
LSB
I2C_SCL
Start Condition
I2C_SDA ↓ While I2C_SCL = 1
Stop Condition
I2C_SDA ↑ While I2C_SCL = 1
† Bits CS1 and CS0 in the TAS3103 slave address are compared to the logic levels on pins CS0 and CS1 for address verification. This provides
the ability to address up to four TAS3103 chips on the same I2C bus.
Figure 2−15. I2C Slave Mode Communication Protocol
In the slave mode, the I2C bus is used to:
•
Update coefficient values and output data to those GPIO ports configured as output.
•
Read status flags, input data from those GPIO ports configured as inputs and retrieve spectrum
analyzer/VU meter data.
In the master mode, the I2C bus is used to download a user-specific configuration from an I2C compatible EEPROM.
In the slave mode only, specific registers and memory locations in the TAS3103 are accessible with the use of I2C
subaddresses. There are 256 such I2C subaddresses. The protocol required to access a specific subaddress is
presented in Figure 2−16.
As shown in Figure 2−16, a read transaction requires that the master device first issue a write transaction to give the
TAS3103 the subaddress to be used in the read transaction that follows. This subaddress assignment write
transaction is then followed by the read transaction. For write transactions, the subaddress is supplied in the first byte
of data written, and this byte is followed by the data to be written. For write transactions, the subaddress must always
be included in the data written. There cannot be a separate write transaction to supply the subaddress, as was
required for read transactions. If a subaddress assignment only write transaction is followed by a second write
transaction supplying the data, erroneous behavior results. The first byte in the second write transaction is interpreted
by the TAS3103 as another subaddress replacing the one previously written.
2−17
I2C Read Transaction
Start
(By Master)
S
TAS3103
Subaddress
(By Master)
Write
(By Master)
TAS3103
Address
7-Bit Slave
Address
(By Master)
W
ACK
Stop
Start
(By Master) (By Master)
ACK
Subaddress
Acknowledge
(By TAS3103)
S
S
Acknowledge
(By TAS3103)
Read
(By Master)
TAS3103
Address
7-Bit Slave
Address
(By Master)
R
Data
(By TAS3103)
ACK
Data
Acknowledge
(By TAS3103)
Stop
(By Master)
Data
(By TAS3103)
ACK
Data
Acknowledge
(By Master)
ACK
NAK
S
Acknowledge
(By Master)
No Acknowledge
(By Master)
I2C Write Transaction
Start
(By Master)
S
Write
(By Master)
TAS3103
Address
7-Bit Slave
Address
(By Master)
W
ACK
TAS3103
Subaddress
(By Master)
Subaddress
Acknowledge
(By TAS3103)
Data
(By Master)
ACK
Acknowledge
(By TAS3103)
Data
Stop
(By Master)
Data
(By Master)
ACK
Acknowledge
(By TAS3103)
Data
ACK
ACK
S
Acknowledge
(By TAS3103)
Acknowledge
(By TAS3103)
Figure 2−16. I2C Subaddress Access Protocol
2.3.2.1 I2C Master Mode Operation
The TAS3103 uses the master mode to download an operational configuration. The configuration downloaded must
contain data for all 256 subaddresses, with spacer data supplied for those subaddresses that are GPIO
subaddresses, read-only subaddresses, factory-test subaddresses, or unused (reserved) subaddresses. The spacer
data must always be assigned the value zero. Table 2−3 organizes the 256 subaddresses (and their corresponding
EEPROM addresses) into sequential blocks, with each block containing either valid data or spacer data.
Table 2−3 also illustrates that the subaddresses and their corresponding EEPROM memory addresses do not directly
correlate. This is because many subaddresses are assigned more than one 32-bit word. For example, there is a
unique subaddress for each biquad filter in the TAS3103, but each subaddress is assigned five 32-bit
coefficients—resulting in 20 bytes of memory being assigned to each biquad subaddress.
The TAS3103, in the I2C master mode, can execute a complete download without requiring any wait states. After the
TAS3103 has downloaded all 2367 bytes of coefficient and spacer data, the I2C bus is disabled and cannot be used
to update coefficient values or retrieve status or spectrum/VU meter data. Volume control is available in the master
mode via the four GPIO pins.
In I2C master mode, the watchdog timer must not be enabled.
When programming the EEPROM, make sure that the starting I2C check word (subaddres 0x00) and ending I2C
check word (subaddress 0xFC) are identical.
2−18
Table 2−3. I2C EEPROM Data
DATA TYPE
Valid data
EEPROM BYTE
ADDRESSES
SUBADDRESS(es)
Starting I2C check word—must match ending check word
0x000−0x003
0x00
Input mixers—set 1
0x004−0x0CF
0x01−0x33
Effects block biquads
0x0D0−0x2AF
0x34−0x4B
Reverb block mixers
0x2B0−0x2C7
0x4C−0x4E
CH1 biquads
0x2C8−0x3B7
0x4F−0x5A
CH2 biquads
0x3B8−0x4A7
0x5B−0x66
CH3 biquads
0x4A8−0x597
0x67−0x72
Bass and treble inline/bypass mixers
0x590−0x5AF
0x73−0x75
DRC mixers
0x5B0−0x5DF
0x76−0x7E
Dither input mixers
0x5E0−0x5EB
0x7F−0x81
CH3 (sub-woofer) to CH 1/2 (L/R) mixers
0x5EC−0x5F3
0x82−0x83
Spectrum analyzer/VU meter mixers
0x5F4−0x60B
0x84−0x89
Output mixers
0x60C−0x66B
0x8A−0xA1
CH1 loudness parameters
0x66C−0x697
0xA2−0xA6
CH2 loudness parameters
0x698−0x6C3
0xA7−0xAB
CH3 loudness parameters
0x6C4−0x6EF
0xAC−0xB0
CH 1/2 DRC parameters
0x6F0−0x733
0xB1−0xB5
CH3 DRC parameters
0x734−0x777
0xB6−0xBA
Spectrum analyzer parameters
0x778−0x847
0xBB−0xC5
Dither output mixers
0x848−0x84F
0xC6
Dither speed
0x850−0x853
0xC7
0x854−0x85F
0xC8−0xC9
0x860−0x87F
0xCA−0xD1
Factory Test Data (EEPROM Spacer Data) − Zeros
Valid data
Input mixers—set 2
0x880−0x8E3
0xD2−0xEA
Valid data
Watchdog timer enable—must be disabled = 00 00 00 01
0x8E4−0x8E7
0xEB
Factory Test Data (EEPROM Spacer Data) − Zeros
0x8E8−0x8EF
0xEC−0xED
GPIO port parameters
0x8F0−0x8F7
0xEE−0xEF
Volume parameters
0x8F8−0x90B
0xF0−0xF4
Bass/treble filter selections
I2S command word
0x90C−0x91R
0xF5−0xF8
0x91C−0x91F
0xF9
Delay/reverb settings
I2C M and N
0x920−0x92B
0xFA
0x92C−0x92F
0xFB
Ending I2C check word—must match starting check word
0x930−0x933
0xFC
0x934−0x941
0xFD−0xFF
Spacer Data
Valid data
Read Only Data (EEPROM Spacer Data)
NOTE: EEPROM organization must be big Endian−MS byte of data word allocated to the lowest address in memory.
The I2C master mode also utilizes the starting and ending I2C check words to verify a proper EEPROM download.
The first 32-bit data word received from the EEPROM, the starting I2C check word at subaddress 0x00, is stored and
compared against the 32-bit data word received for subaddress 0xFC, the ending I2C check word. These two data
words must be equal as stored in the EEPROM. If the two words do not match when compared in the TAS3103, the
TAS3103 conducts another parameter download from the EEPROM. If the comparison check again fails, the
TAS3103 discards all downloaded parameters and set all parameters to the default values listed in the subaddress
table presented in the Appendix. In the I2C slave mode, these default values are used to initialize the TAS3103 at
power turnon or after a reset.
2−19
2.3.2.2 I2C Slave Mode Operation
The I2C slave mode is the mode that must be used if it is required to change configuration parameters (other than
volume via the GPIO pins for the I2C master mode) during operation. The I2C slave mode is also the only I2C mode
that provides access to the spectrum analyzer and VU meter outputs. Configuration downloads from a master device
can be used to replace the I2C master mode EEPROM download.
For I2C read commands, the TAS3103 responds with data, a byte at a time, starting at the subaddress assigned, as
long as the master device continues to respond with acknowledges. If a given subaddress does not use all 32 bits,
the unused bits are read as logic 0. I2C write commands, however, are treated in accordance with the data assignment
for that address space. If a write command is received for a biquad subaddress, the TAS3103 expects to see five
32-bit words. If fewer than five data words have been received when a stop command (or another start command)
is received, the data received is discarded. If a write command is received for a mixer coefficient, the TAS3103
expects to see only one 32-bit word.
Supplying a subaddress for each subaddress transaction is referred to as random I2C addressing. The TAS3103 also
supports sequential I2C addressing. For write transactions, if a subaddress is issued followed by data for that
subaddress and the fifteen subaddresses that follow, a sequential I2C write transaction has taken place, and the data
for all 16 subaddresses is successfully received by the TAS3103. For I2C sequential write transactions, the
subaddress then serves as the start address and the amount of data subsequently transmitted, before a stop or start
is transmitted, determines how many subaddresses are written to. As was true for random addressing, sequential
addressing requires that a complete set of data be transmitted. If only a partial set of data is written to the last
subaddress, the data for the last subaddress is discarded. However, all other data written is accepted; just the
incomplete data is discarded.
The GPIO subaddresses and most reserved read-only and factory test subaddresses require the downloading of four
bytes of zero-valued spacer data in order to proceed to the next subaddress. However, there are five exceptions to
this rule and Table 2−4 lists the subaddresses of these fexceptions and the number of zero-valued bytes that must
be written.
Table 2−4. Four Byte Write Exceptions—Reserved and Factory-Test I2C Subaddresses
SUB-ADDRESS
NUMBER OF ZERO-VALUED BYTES
THAT MUST BE WRITTEN
0xC9
8
0xED
8
0xFD
10 (0xA)
0xFE
2
0xFF
1
The TAS3103 can always receive sequential I2C addressing write data without issuing wait states. If it is desired to
download data to all subaddresses using one sequential write transaction, spacer data for the reserved, GPIO,
read-only, and factory-test subaddresses must be supplied as per Table 2−3 and Table 2−4.
The TAS3103 also supports sequential read transactions. When an I2C subaddress assignment write transaction is
followed by a read transaction, the TAS3103 outputs the data for that subaddress, and then continue to output data
for the subaddresses that follow as long as the master continues to issue data received acknowledges. Except for
two exceptions, the TAS3103 outputs four bytes of zero-valued data for reserved and factory-test subaddresses. The
subaddresses of the exceptions and the number of bytes supplied by the TAS3103 for each exception are given in
Table 2−5. If a GPIO port is assigned as an output port, a logic 0 bit value is supplied by the TAS3103 for this GPIO
port in response to a read transaction at subaddress 0xEE.
CAUTION: Sequential write transactions must be in ascending subaddress order. The
TAS3103 does not wrap around from subaddress 0xFF to 0x00.
Sequential read transactions wrap around from subaddress 0xFF to 0x00.
2−20
Table 2−5. Four Byte Read Exceptions—Reserved and Factory-Test I2C Subaddresses
SUB-ADDRESS
NUMBER BYTES SUPPLIED BY TAS3103
0xC9
8
0xED
8
NOTE: Table 2−5 does not include read-only subaddresses and thus does
not include subaddresses 0xFD, 0xFE, and 0xFF. When read, these
read-only subaddresses output 10, 2, and 1 byte respectively.
Thus, for all reserved and factory-test subaddresses, except subaddresses OxC9 and 0xED, the master device must
issue four data received acknowledges for the four bytes of zero-valued data. For subaddresses OxC9 and 0xED,
the master device must issue eight data received acknowledges for the eight bytes of zero-valued data.
Sequential read transactions do not have restrictions on outputting only complete subaddress data sets. If the master
does not issue enough data received acknowledges to receive all the data for a given subaddress, the master device
simply does not receive all the data. If the master device issues more data received acknowledges than required to
receive the data for a given subaddress, the master device simply receives complete or partial sets of data, depending
on how many data received acknowledges are issued from the subaddress(es) that follow.
I2C read transactions, both sequential and random, can impose wait states. For the standard I2C mode
(SCL = 100 kHz), worst-case wait state times for an 8-MHz microprocessor clock is on the order of 2 µs. Nominal
wait state times for the same 8-MHz microprocessor clock is on the order of 1 µs. For the fast I2C mode (SCL =
400 kHz) and the same 8-MHz microprocessor clock, worst-case wait state times can extend up to 10.5 µs in duration.
Nominal wait state times for this same case lie in a range from 2 µs to 4.6 µs. Increasing the microprocessor clock
frequency lowers the wait state times and for the standard I2C mode, a higher microprocessor clock can totally
eliminate the presence of wait states. For example, increasing the microprocessor clock to 16 MHz results in no wait
states. For the fast I2C mode, higher microprocessor clocks shortens the wait state times encountered, but does not
totally eliminate their presence.
2.4 Digital Audio Processor (DAP) Arithmetic Unit
The digital audio processor (DAP) arithmetic unit is a fixed-point computational engine consisting of an arithmetic unit
and data and coefficient memory blocks. Figure 2−17 is a block diagram of the arithmetic unit.
4K x 16
Delay Line
RAM
Dual Port
Data RAM
Coefficient
RAM
Program
ROM
Arithmetic
Engine
DAP
Instruction Decoder/Sequencer
76-Bit Adder
Regs
Regs
Digital Audio Processor
(DAP)
Arithmetic Unit
Figure 2−17. Digital Audio Processor Arithmetic Unit Block Diagram
The DAP arithmetic unit is used to implement all firmware functions—soft volume, loudness compensation, bass and
treble processing, dynamic range control, channel filtering, 3D effects, input and output mixing, spectrum analyzer,
VU meter, and dither.
Figure 2−18 shows the data word structure of the DAP arithmetic unit. Eight bits of overhead or guard bits are provided
at the upper end of the 48-bit DAP word and 8 bits of computational precision or noise bits are provided at the lower
end of the 48-bit word. The incoming digital audio words are all positioned with the most significant bit abutting the
8-bit overhead/guard boundary. The sign bit in bit 39 indicates that all incoming audio samples are treated as signed
data samples.
2−21
CAUTION: Audio data into the TAS3103 is always treated as signed data.
47 S
S
S
S
40 S
39 S
32
31
Overhead/Guard Bits
16-Bit
Audio
18-Bit
Audio
20-Bit
Audio
24-Bit
Audio
24
23
22
21
20
19
32-Bit
Audio
16
15
8
7
Precision/Noise Bits
0
Figure 2−18. DAP Arithmetic Unit Data Word Structure
The arithmetic engine is a 48-bit (25.23 format) processor consisting of a general-purpose 76-bit arithmetic logic unit
and function-specific arithmetic blocks. Multiply operations (excluding the function-specific arithmetic blocks) always
involve 48-bit DAP words and 28-bit coefficients (usually I2C programmable coefficients). If a group of products are
to be added together, the 76-bit product of each multiplication is applied to a 76-bit adder, where a DSP-like
multiply-accumulate (MAC) operation takes place. Biquad filter computations use the MAC operation to maintain
precision in the intermediate computational stages.
To maximize the linear range of the 76-bit ALU, saturation logic is not used. Intermediate overflows are then permitted
in multiply-accumulate operations, but it is assumed that subsequent terms in the multiply-accumulate computation
flow corrects the overflow condition. The biquad filter structure used in the TAS3103 is the direct form I structure and
has only one accumulation node. With this type of structure, intermediate overflow is allowed as long as the designer
of the filters has assured that the final output is bounded and does not overflow. Figure 2−19 shows a bounded
computation that experiences intermediate overflow condition. 8-bit arithmetic is used for ease of illustration.
The DAP memory banks include a dual port data RAM for storing intermediate results, a coefficient RAM, a 4K x 16
RAM for implementing the delay stages, and a fixed program ROM. Only the coefficient RAM, assessable via the I2C
bus, is available to the user.
8-Bit ALU Operation
(Without Saturation)
Rollover
10110111
(−73)
−73
+ 11001101
(−51)
+ −51
10000100
(−124)
−124
+ 11010011
(−45)
+ −45
01010111
(57)
+ 00111011
(59)
10010010
(−110)
−169
+
59
−110
Figure 2−19. DAP ALU Operation With Intermediate Overflow
2−22
The DAP processing clock is set by pins PLL0 and PLL1, in conjunction with the source clock XTALI or MCLKI. The
DAP operates at speeds up to 136 MHz, which is sufficient to process 96-kHz audio.
2.5 Reset
The reset circuitry in the TAS3103 is shown in Figure 2−20. A reset is initiated by inputting logic 0 on the reset pin
RST.. A reset is also issued at power turnon by the internal 1.8-V regulator subsystem.
VDSS
A_VDSS
1.8-V Regulator Subsystem
PWR GOOD
RST
CLR
Enable
MCLKI
Reset Timer
Chip Reset
Lock
DPLL
XTALI
dpll_clk
Figure 2−20. TAS3103 Reset Circuitry
At power turnon, the internal 1.8-V regulator subsystem issues an internal reset that remains active until regulation
is reached. The duration of this signal assures that all reset activities are conducted at power turnon. This means that
the external reset pin RST does not require an RC time constant derived external reset to assure that a reset is applied
at power turnon. The reset pin RST can then be used exclusively for exception resets, saving the cost and size impact
of additional RC components. However, since RST is an asynchronous clear, it can respond to narrow negative signal
transitions. Some applications, therefore, might require a high-frequency capacitor on the RST pin in order to remove
unwanted noise excursions.
2.6 Power Down
Setting the PWRDN pin to logic 1 enables power down. Power down stops all clocks in the TAS3103, but preserves
the state of the TAS3103. When PWRDN is deactivated (set to logic 0) after a period of activation, the TAS3103
resumes the processing of audio data upon receiving the next LRCLK (indicating a new sample of audio data is
available for processing). The configuration of the TAS3103 and all programmable parameters are retained during
power down.
There is a time lag between setting PWRDN to logic 1 and entering the power down state. PWRDN is sampled every
GPIOFSCOUNT LRCLK periods (see the subaddress 0xEF and the watchdog timer and GPIO ports sections). This
means that a time lag as great as GPIOFSCOUNT(1/LRCLK) could exist between the activation of PWRDN (setting
to logic 1) and the time at which the microprocessor recognizes that the PWRDN pin has been activated. Normally,
upon recognizing that the PWRDN pin has been activated, the TAS3103 enters the power-down state approximately
80 microprocessor clock cycles later. However, if a soft volume update is in progress, the TAS3103 waits until the
soft volume update is complete before entering the power down state. For this case then, the worst case time lag
2−23
between recognizing the activation of pin PWRDN and entering the power down would be 4096 LRCLK periods,
assuming a volume slew rate selection (bit VSC of I2C subaddress 0xF1) of 4096 and the issuance of a volume
update immediately preceding the reading of pin PWRDN. The worst case time lag between setting PWRDN to logic
1 and entering the power down state is then:
80
power – down – time lag Worst-Case + 4096 ) GPIOFSCOUNT )
LRCLK
Microprocessor-Clock
There is also a time lag between deactivating PWRDN (setting PWRDN to logic 0) and exiting the power down state.
This time lag is set by the time it takes the internal digital PLL to stabilize, and this time, in turn, is set by the master
clock frequency (MCLKI or XTALI) and the PLL output clock frequency. For a 135-MHz PLL output clock and a 24.576
MCLKI, the time lag is approximately 25 µs. For a 11.264-MHz PLL output clock and a 1.024-MHz MCLKI, the time
lag is approximately 360 µs.
Power consumption in the power-down state is approximately 12 mW.
2.7 Watchdog Timer
There is a watchdog timer in the TAS3103 that monitors the microprocessor activity. If the microprocessor ever
ceases to execute its stored program, the watchdog timer fires and resets the TAS3103. This capability was included
in the TAS3103 for factory test purposes and has little use in applications. The program structure used in the
microprocessor assures that the microprocessor always executes its stored program unless a hardware failure
occurs.
The watchdog timer is governed by the parameter GPIOFSCOUNT in subaddress 0xEF and the LSB of the 32-bit
word at subaddress 0xEB. The default value of the LSB of the 32-bit word at subaddress 0xEB is 1 and this value
disables the watchdog timer. The GPIOFSCOUNT is also used in other functions and balancing the needs of these
other functions regarding GPIOFSCOUNT with the requirements of the watchdog timer is an involved process. For
this reason, it is strongly recommended that the LSB of the 32-bit word at subaddress 0xEB remain a 1. If an
application does require use of the watchdog timer, it is requested that the user contact an application engineer in
the Digital Audio Department of Texas Instruments for details in properly using this feature.
2.8 General-Purpose I/O (GPIO) Ports
The TAS3103 has four general-purpose I/O (GPIO) ports. Figure 2−21 is a block diagram of the GPIO circuitry in the
TAS3103.
2−24
0xEF
31
S Slave Addr Ack
Sub-Addr
24
23 20 19
Ack 00000000 Ack 0000
16
15
8
7
0
GPIODIR Ack GPIOFSCOUNT Ack GPIO_samp_int Ack
3 2 1 0
Determines How Many Consecutive Logic 0 Samples
(Where Each Sample Is Spaced by GPIOFSCOUNT
LRCLKs) are Required to Read a Logic 0 on a
GPIO Input Port
Microprocessor
Microprocessor
Firmware
Down
Counter
LRCLK
LD
Decode 0
DATA PATH SWITCH
GPIO0
D
GPIO1
D
GPIO2
D
GPIO3
D
Q
I2C Slave Mode
and
I2C Master Mode
Write
Q
Sample
Logic
I2C Master
Mode Read
Q
Microprocessor
Control
Q
S Slave Addr Ack
Sub-Addr
Ack 00000000 Ack 00000000 Ack 00000000 Ack 0000
31
24
23
16
15
8
7
3 2 1 0 Ack
GPIO_in_out
4 3
0
0xEE
Figure 2−21. GPIO Port Circuitry
GPIO Functionality—I2C Master Mode
2.8.1
In the I2C master mode, the GPIO ports are strictly input ports and are used to control volume. Table 2−6 lists the
functionality of each GPIO port in the I2C master mode. Bit field GPIOFSCOUNT (15:8) of I2C subaddress 0xEF
governs the rate at which the GPIO pins are sampled for a volume update. The sample rate is:
ƒ
GPIO_Port
+
LRCLK
GPIOFSCOUNT
Table 2−6. GPIO Port Functionality—I2C Master Mode
GPIO PORT
FUNCTION
GPIO0 (pin 18)
Volume up—CH1 and CH2
GPIO1 (pin 19)
Volume down—CH1 and CH2
GPIO2 (pin 20)
Volume up—CH3
GPIO3 (pin 21)
Volume down—CH3
GPIOFSCOUNT also governs the rate at which the power down pin PWRDN is sampled and the rate at which the
watchdog counter is reset. GPIOFSCOUNT then cannot be independently used to tune the volume adjustment. For
this reason, bit field GPIO_samp_int of the same I2C subaddress (0xEF) is included to provide the ability to adjust
the responsiveness (or sluggishness) of the volume switches.
Each GPIO port has a weak pullup to VDDS. A volume control switch then typically switches the signal line to the GPIO
port between ground and an open circuit. The parameter GPIO_samp_int sets how many consecutive GPIO port
2−25
samples must be logic 0 before a logic 0 is read. A read logic 0 on a given GPIO port is interpreted as a command
to increase or decrease volume. If a logic 0 is read, and the signal level into the GPIO port remains at logic 0 for another
GPIO_samp_int consecutive samples, a second logic 0 value is read.
For each logic 0 read, the volume is increased or decreased 0.5 dB. After two consecutive logic 0 readings, each logic
0 reading that follows results in the volume level increasing or decreasing 5 dB instead of 0.5 dB. Figure 2−22 shows
an example of activating a volume switch. For the example in Figure 2−22, GPIOFSCOUNT is set to 3 and
GPIO_samp_int is set to 2. It is also noted in Figure 2−22 that the parameter GPIO_samp_int only comes into play
on logic 0 valued samples. As soon as the GPIO sample goes to logic 1, the audio updating ceases.
2.8.2
GPIO Functionality—I2C Slave Mode
In the I2C slave mode, the GPIO ports can be used as true general-purpose ports. Each port can be individually
programmed, via the I2C bus, to be either an input or an output port. The default assignment for all GPIO ports, in
the I2C slave mode, is an input port.
When a given GPIO port is programmed as an output port, by setting the appropriate bit in the bit field GPIODIR
(19:16) of subaddress 0xEF to logic 1, the logic level output is set by the logic level programmed into the appropriate
bit in bit field GPIO_in_out (3:0) of subaddress 0xEE. The I2C bus then controls the logic output level for those GPIO
ports assigned as output ports.
When a given GPIO port is programmed as an input port by setting the appropriate bit in bit field GPIODIR (19:16)
of subaddress 0xEF to logic 0, the logic input level into the GPIO port is written to the appropriate bit in bit field
GPIO_in_out (3:0) of subaddress 0xEE. The I2C bus can then be used to read bit field GPIO_in_out to determine
the logic levels at the input GPIO ports. Whether a given bit in the bit field GPIO_in_out is a bit to be read via the I2C
bus or a bit to be written to via the I2C bus is strictly determined by the corresponding bit setting in bit field GPIODIR.
In the I2C slave mode, the GPIO input ports are read every GPIOFSCOUNT LRCLKs, as was the case in the I2C
master mode. However, parameter GPIO_samp_int does not have a role in the I2C slave mode. If a GPIO port is
assigned as an output port, a logic 0 bit value is supplied by the TAS3103 for this GPIO port in response to a read
transaction at subaddress 0xEE.
If the GPIO ports are left in their power turnon state default state, they are input ports with a weak pullup on the input
to VDSS.
2−26
2−27
GPIO Reads
GPIO Data
Samples
LRCLK
GPIO Pin
Input
Read = 1
Read = 1
Adjust 0.5 dB
Adjust 5 dB
Read = 0
GPIO_samp_int = 2
Read = 0
Adjust 5 dB
Read = 0
Read = 1
Read = 1
Figure 2−22. Volume Adjustment Timing—Master I2C Mode
Adjust 0.5 dB
Read = 0
GPIOFSCOUNT = 3
Adjust 0.5 dB
Read = 0
Adjust 0.5 dB
Read = 0
2−28
3 Firmware Architecture
3.1 I2C Coefficient Number Formats
The firmware for the TAS3103 is housed in ROM resources within the TAS3103 and cannot be altered. However,
mixer gain, level offset, and filter tap coefficients, which can be entered via the I2C bus interface, provide a user with
the flexibility to set the TAS3103 to a configuration that achieves the system level goals.
The firmware is executed in a 48-bit signed fixed-point arithmetic machine. The most significant bit of the 48-bit data
path is a sign bit, and the 47 lower bits are data bits. Mixer gain operations are implemented by multiplying a 48-bit
signed data value by a 28-bit signed gain coefficient. The 76-bit signed output product is then truncated to a signed
48-bit number. Level offset operations are implemented by adding a 48-bit signed offset coefficient to a 48-bit signed
data value. In most cases, if the addition results in overflowing the 48-bit signed number format, saturation logic is
used. This means that if the summation results in a positive number that is greater than 0x7FFF_FFFF_FFFF (the
spaces are used to ease the reading of the hexadecimal number), the number is set to 0x7FFF_FFFF_FFFF. If the
summation results in a negative number that is less than 0x8000_0000_0000 0000, the number is set to
0x8000_0000_0000 0000. There are exceptions to the use of saturation logic for summations that overflow—see the
section Digital Audio Processor (DAP) Arithmetic Unit.
3.1.1
28-Bit 5.23 Number Format
All mixer gain coefficients are 28-bit coefficients using a 5.23 number format. Numbers formatted as 5.23 numbers
means that there are 5 bits to the left of the decimal point and 23 bits to the right of the decimal point. This is shown
in the Figure 3−1.
2−23 Bit
2−4 Bit
2−1 Bit
20 Bit
23 Bit
Sign Bit
S_xxxx.xxxx_xxxx_xxxx_xxxx_xxx
Figure 3−1. 5.23 Format
The decimal value of a 5.23 format number can be found by following the weighting shown in Figure 3−2. If the most
significant bit is logic 0, the number is a positive number, and the weighting shown yields the correct number. If the
most significant bit is a logic 1, then the number is a negative number. In this case every bit must be inverted, a 1 added
to the result, and then the weighting shown in Figure 3−2 applied to obtain the magnitude of the negative number.
23 Bit
22 Bit
20 Bit
2−1 Bit
2−4 Bit
2−23 Bit
(1 or 0) x 23 + (1 or 0) x 22 + … + (1 or 0) x 20 + (1 or 0) x 2−1 + … + (1 or 0) x 2−4 + … + (1 or 0) x 2−23
Figure 3−2. Conversion Weighting Factors—5.23 Format to Floating Point
3−1
Gain coefficients, entered via the I2C bus, must be entered as 32-bit binary numbers. The format of the 32-bit number
(4-byte or 8-digit hexadecimal number) is shown in Figure 3−3.
Fraction
Digit 6
Sign
Bit
Integer
Digit 1
u
u
u
u
Coefficient
Digit 8
S
x
x
x
Coefficient
Digit 7
Fraction
Digit 1
x. x
x
x
Coefficient
Digit 6
Fraction
Digit 2
x
x
x
x
Coefficient
Digit 5
Fraction
Digit 3
x
x
x
x
Coefficient
Digit 4
Fraction
Digit 4
x
x
x
x
Coefficient
Digit 3
Fraction
Digit 5
x
x
x
x
Coefficient
Digit 2
0
x
x
x
x
Coefficient
Digit 1
u = unused or don’t care bits
Digit = hexadecimal digit
Figure 3−3. Alignment of 5.23 Coefficient in 32-Bit I2C Word
As Figure 3−3 shows, the hex value of the integer part of the gain coefficient cannot be concatenated with the hex
value of the fractional part of the gain coefficient to form the 32-bit I2C coefficient. The reason is that the 28-bit
coefficient contains 5 bits of integer, and thus the integer part of the coefficient occupies all of one hex digit and the
most significant bit of the second hex digit. In the same way, the fractional part occupies the lower 3 bits of the second
hex digit, and then occupies the other five hex digits (with the eighth digit being the zero-valued most significant hex
digit).
3.1.2
48-Bit 25.23 Number Format
All level adjustment and threshold coefficients are 48-bit coefficients using a 25.23 number format. Numbers
formatted as 25.23 numbers means that there are 25 bits to the left of the decimal point and 23 bits to the right of the
decimal point. This is shown in Figure 3−4.
2−23 Bit
2−10 Bit
2−1 Bit
20 Bit
216 Bit
222 Bit
223 Bit
Sign Bit
S_xxxx_xxxx_xxxx_xxxx_xxxx_xxxx.xxxx_xxxx_xxxx_xxxx_xxxx_xxx
Figure 3−4. 25.23 Format
3−2
Figure 3−5 shows the derivation of the decimal value of a 48-bit 25.23 format number.
223 Bit
222 Bit
20 Bit
2−1 Bit
2−23 Bit
(1 or 0) x 223 + (1 or 0) x 222 + … + (1 or 0) x 20 + (1 or 0) x 2−1 + … + (1 or 0) x 2−23
Figure 3−5. Alignment of 5.23 Coefficient in 32-Bit I2C Word
Two 32-bit words must be sent over the I2C bus to download a level or threshold coefficient into the TAS3103. The
alignment of the 48-bit, 25.23 formatted coefficient in the 8-byte (two 32-bit words) I2C word is shown in Figure 3−6.
Integer
Digit 4
(Bits 23 − 21)
Sign
Bit
Integer
Digit 1
u
u
u
u
Coefficient
Digit 16
u
u
u
u
Coefficient
Digit 15
u
u
u
u
Coefficient
Digit 14
u
u
u
u
Coefficient
Digit 13
S
x
x
x
Coefficient
Digit 12
Integer
Digit 2
x
x
x
x
Coefficient
Digit 11
Integer
Digit 3
x
x
x
x
Coefficient
Digit 10
x
Word 1
(Most
Significant
Word)
x
Fraction
Digit 6
Integer
Digit 5
x
x
Coefficient
Digit 9
Integer
Digit 4
(Bit 20)
x
x
x
x
Coefficient
Digit 8
Integer
Digit 6
x
x
x
x
Coefficient
Digit 7
Fraction
Digit 1
x. x
x
x
Coefficient
Digit 6
Fraction
Digit 2
x
x
x
x
Coefficient
Digit 5
Fraction
Digit 3
x
x
x
x
Coefficient
Digit 4
Fraction
Digit 4
x
x
x
x
Coefficient
Digit 3
Fraction
Digit 5
x
x
x
x
Coefficient
Digit 2
0
x
x
x
x
Word 2
(Least
Significant
Word)
Coefficient
Digit 1
u = unused or don’t care bits
Digit = hexadecimal digit
Figure 3−6. Alignment of 25.23 Coefficient in Two 32-Bit I2C Words
3−3
3.2 Input Crossbar Mixers
The TAS3103 has four serial input ports—SDIN1, SDIN2, SDIN3 and SDIN4. SDIN1, SDIN2, and SDIN3 provide the
input resources to process 5.1 channel audio in two TAS3103 chips. SDIN4 provides the capability to multiplex
between a full 5.1 channel system and a stereo source or an information/warning audio message as might be found
in an automotive application.
Each serial input port is assigned two internal processing nodes. The mixers following these internal processing
nodes serve to distribute the input audio data to various processing nodes within the TAS3103. Figure 3−7 shows
the assignment of the internal processing nodes to the serial input ports. Two cases are shown in Figure 3−7—
discrete mode and TDM mode.
The input crossbar mixer topology for internal processing nodes A, B, C, D, E and F is shown in Figure 3−8. Each
of the six nodes is assigned six mixers. These six mixers provide the ability to route the incoming serial port data on
SDIN1, SDIN2, and SDIN3 to:
•
Processing node d—bypassing effects block and directly feeding monaural CH1
•
Processing node e—bypassing effects block and directly feeding monaural CH2
•
Processing node f—directly feeding the section of the effects block assigned to monaural CH3
•
Processing nodes a and b—directly feeding paths that contain the reverb delay elements assigned to CH1
and CH2
•
Processing node c—directly feeding an effects block path assigned to CH1 and CH2 that bypasses all
reverb delay elements.
The ability to route all input nodes to the same set of processing nodes fully decouples the input order (what audio
components are wired to which serial input ports) from the processing flow. As is seen in the discussion of the output
crossbar mixers, the output serial ports are fully decoupled from the three monaural channels (any monaural channel
output can be routed to either the left or the right side of any output port). The TAS3103 thus provides full flexibility
in the routing of audio data into and out of the chip.
The mixer topology for internal processing nodes g and h is shown in Figure 3−9. Nodes g and h are each assigned
three mixers. The mixers provide the ability to route the incoming data on serial port SDIN4 to:
3−4
•
Output processing nodes to facilitate input to output pass through
•
CH1/CH2 effects block input nodes that bypass reverb delay
•
CH3 effects block input node
SDIN1
LRCLK
L
Internal
Processing
Nodes
Internal
Processing
Nodes
R
A
A
L
B
B
R
time
SDIN2
LRCLK
L
Internal
Processing
Nodes
R
time
R
C
C
L
D
D
Internal
Processing
Nodes
Internal
Processing
Nodes
R
E
E
L
F
F
Internal
Processing
Nodes
Internal
Processing
Nodes
R
G
G
L
H
H
SDIN3
LRCLK
L
R
time
SDIN4
LRCLK
L
time
Internal
Processing
Nodes
SDIN1
R
(a) Discrete Mode − For I2S Format, Polarity
of LRCLK Opposite That Shown
(b) TDM Mode
Figure 3−7. Serial Input Port to Processing Node Topology
3−5
3−6
Input Crossbar
Mixers
f
b
h
c
g
a
4
BiQuad
Filters
4
BiQuad
Filters
4
BiQuad
Filters
4
BiQuad
Filters
4
BiQuad
Filters
4
BiQuad
Filters
Reverb
Delay
Reverb
Delay
Reverb
Delay
3-D Effects Block
e
d
aa
Figure 3−8. Input Mixer and Effects Block Topology—Internal Processing Nodes A, B, C, D, E, and F
Processing
Node F
Processing
Node E
Processing
Node D
Processing
Node C
Processing
Node B
Processing
Node A
Monaural
CH 3
Monaural
CH 2
Monaural
CH 1
3−7
f
h
Monaural
CH 3†
Monaural
CH 2†
Monaural
CH 1†
Figure 3−9. Input Mixer Topology—Internal Processing Nodes G and H
† Monaural channels consist of 12 biquad filters, followed by bass and treble processing, followed by
bolume and loudness processing, followed by dynamic range control, followed by fither processing.
See the TAS3103 Firmware Block Diagram in the Appendix.
Processing
Node H
Processing
Node G
g
Effects Block
(See Figure 3−10)
Output Crossbar
Mixer
(See Figure 3−33)
Z
Y
X
W
V
U
All input crossbar mixers use signed 5.23 format mixer gain coefficients and all are programmable via the I2C bus.
The 5.23 format provides a range of gain adjustment from 2−23 (−138 dB) to 24 – 1 (23.5 dB).
3.3 3D Effects Block
The 3D effects block, shown in Figure 3−10, performs the first suite of processing tasks conducted on the incoming
serial audio data streams. The TAS3103 has three monaural channels—CH1, CH2, and CH3. CH1 and CH2 share
the same effects block, as well as the same dynamic range compression block. CH 3 has its own effects block and
its own dynamic range compression block. In typical two TAS3103 chip configurations for processing 5.1 audio,
monaural channels CH1 and CH2 are used to process left/right front and left/right surround audio components, and
CH3 is used to process the subwoofer and center audio components. To support such a processing structure, the
effects block for CH1/CH2 offers more option for inserting audio effects into the audio data stream than does the
effects block for CH3.
3.3.1
CH1/CH2 Effects Block
This block consists of five signal flow paths, starting at processing nodes a, b, c, g, and h. All five paths contain four
programmable biquad filters, and paths a and b contain reverb delay lines as well. Nodes a, b and c can be sourced
by any of the input nodes A, B, C, D, E, and F (SDIN1, SDIN2, SDIN3). Nodes g and h can be sourced by input nodes
G and H respectively (SDIN4) and/or by weighted replicas of the data on nodes A and B respectively. Nodes g and
h can also be sourced by the same weighted replica of the data on node f. Node c can also be sourced by weighted
replicas of the data on both node a and node b.
3D effects processing typically consist of installing sound direction effects. Sound direction effects are typically
created by the use of three major components.
•
Time differentiation
•
Loudness differentiation
•
Spectral differentiation
Time differentiation is achieved by using the paths containing the reverb delay elements. Loudness differentiation
is achieved both by the mixers feeding the five paths and the two mixers located at the output of each of the five paths.
Spectral differentiation is achieved using the four biquad filters located in each path.
3.3.2
CH3 Effects Block
This block, starting at node f, typically processes center and sub-woofer audio components. Time and spectral
differentiation with respect of CH1 and CH2 can be realized via the reverb delay line and the four biquad filters.
Loudness differentiation can be achieved by adjusting the volume level of CH3. Weighted replicas of the effects block
output for CH1 and CH2 can also be summed into the output of the CH3 effects block. This capability is typically used
when processing the center channel audio component.
3−8
3−9
Mixer Gain Coefficient Sub-Address Format
Node H
Processing
Node G
Processing
A, B, C, D, E, F
Processing Nodes
Reverb
Mixers
f
0x33
b
0x28
h
0x26
c
0x25
g
0x27
a
4
BiQuad
Filters
0x48 − 0x4B
4
BiQuad
Filters
0x38 − 0x3B
4
BiQuad
Filters
0x40 − 0x43
4
BiQuad
Filters
0x44 − 0x47
4
BiQuad
Filters
0x3C − 0x3F
4
BiQuad
Filters
0x34 − 0x37
0x30
0x2F
0x32
0x31
0x2E
0x2D
g1
Reverb
Delay
0xFA
Delay
Line
g0
g0/g1 = 0x4E
0xFA
Delay
Line
g0
g1
Reverb
Delay
g1
Reverb
Delay
g0/g1 = 0x4C
0xFA
Delay
Line
g0
g0/g1 = 0x4D
3−D Effects Block
Figure 3−10. TAS3103 3D Effects Processing Block
m
l
S Slave Addr Ack Sub-Addr Ack 0000s xxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxs Ack g0
b
b
m
l
0000s xxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxs Ack g1
b
b
Reverb Mixer Coefficients Sub-Address Format
l
m
l
m
0xFA S Slave Addr Ack Sub-Addr Ack 0000s xxx Ack xxxxxxs Ack 0000s xxx Ack xxxxxxs Ack D1 & R1
b
b
b
b
l
m
l
m
0000s xxx Ack xxxxxxs Ack 0000s xxx Ack xxxxxxs Ack D2 & R2
b
b
b
b
l
m
l
m
0000s xxx Ack xxxxxxs Ack 0000s xxx Ack xxxxxxs Ack D3 & R3
b
b
b
b
Delay
Ack b2
Ack b1
Ack b0
Ack a2
Ack a1
Input Crossbar
l
Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxs
b
l
Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxs
b
l
Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxs
b
l
Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxs
b
l
Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxs
b
Reverb and Delay Assignments Sub-Address Format
m
S Slave Addr Ack Sub-Addr Ack 0000s xxx
b
m
0000s xxx
b
m
0000s xxx
b
m
0000s xxx
b
m
0000s xxx
b
BiQuad Filter Coefficients Sub-Address Format
m
l
S Slave Addr Ack Sub-Addr Ack 0000s xxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxs Ack
b
b
0x2C
0x2B
0x2A
0x29
e
d
0xD1
aa
0xD0
CH 3
Monaural
CH 2
Monaural
CH 1
Monaural
3.4 Biquad Filters
There are a total of 73 biquad filters in the TAS3103. The breakout of the biquad filters per functional element is given
in Table 3−1.
Table 3−1. Biquad Filter Breakout
FUNCTION
BIQUAD FILTERS
SUBADDRESS
3D effects block
24
0x34−0x4B
Monaural channel CH1
12
0x4F−0x5A
Monaural channel CH2
12
0x5B−0x66
Monaural channel CH3
12
0x67−0x72
0xA6−CH1
Loudness processing
3
0xAB−CH2
Spectrum analyzer/VU meter
10
0xBC−0xC5
0xB0−CH3
All 73 biquad filters are second order direct form I structures. A block diagram of the structure of the biquad filter is
shown in Figure 3−11.
Biquad Filter Structure
b0
28
48
76
Σ
76
Magnitude
Truncation
a1
b1
28
z−1
48
48
28
76
b2
z−1
48
76
a2
28
z−1
48
28
76
76
z−1
48
NOTE: All gain coefficients 5.23 numbers.
Biquad Filter Coefficient Subaddress Format
S
Slave Addr
m
Ack Sub-Addr Ack 0000 s xxx
b
m
0000 s xxx
b
m
0000 s xxx
b
m
0000 s xxx
b
m
0000 s xxx
b
Ack
xxxxxxxx
Ack
xxxxxxxx
Ack
xxxxxxxx
Ack
xxxxxxxx
Ack
xxxxxxxx
Ack
xxxxxxxx
Ack
xxxxxxxx
Ack
xxxxxxxx
Ack
xxxxxxxx
Ack
xxxxxxxx
l
Ack xxxxxx s
b
l
Ack xxxxxx s
b
l
Ack xxxxxx s
b
l
Ack xxxxxx s
b
l
Ack xxxxxx s
b
Ack
a1
Ack
a2
Ack
b0
Ack
b1
Ack
b2
NOTE: Each Biquad filter has one subaddress which contains the mixer gain coefficients a1, a2, b0, b1, b2
Figure 3−11. Biquad Filter Structure and Coefficient Subaddress Format
The transfer function for the biquad filter is given by:
*1 ) b z *2
(z)
OUT + b 0 ) b 1z
2
*1
V (z)
1 * a 1z
* a 2z *2
IN
V
3−10
The direct form I structure provides a separate delay element and mixer (gain coefficient) for each node in the biquad
filter. Each mixer output is a signed 76-bit product of a signed 48-bit data sample (25.23 format number) and a signed
28-bit coefficient (5.23 format number). A 76-bit ALU in the TAS3103 allows the 76-bit resolution to be retained when
summing the mixer outputs (filter products). This is an important factor as it removes the need to carefully tailor the
order of addition for each filter implementation to minimize the effects of finite-precision arithmetic. Intermediate
overflows are allowed while summing the biquad terms to further minimize the effects of finite-precision arithmetic
(see the Digital Audio Processor – DAP – Arithmetic Unit section for more discussion on intermediate overflow).
3.5 Bass and Treble Processing
The TAS3103 has three fully independent bass and treble adjustment blocks—one for each of the three monaural
channels. Adjustments in bass and treble are accomplished by selecting a bass filter set and a treble filter set, and
then selecting a shelf filter within each filter set. The filter set selected, of which there are five sets for treble and five
sets for bass to select from, determines the frequency at which the bass and treble adjustments take effect. The shelf
filters determine the gain to be applied to the bass and treble components of the incoming audio. All selections are
independent of one another—any bass filter set can be combined with any treble filter set and any shelf filter can be
selected in a given filter set.
Figure 3−12 shows the bass and treble selections available, their I2C subaddresses, and the data fields in each
subaddress used to make the selections. Each bass filter set has 150 low pass shelving filters to choose from, with
the shelves ranging from a cut (attenuation) of 18 dB to a boost (gain) of 18 dB. A shelf selection of 0 dB effectively
removes bass processing. All 150 filters in a given filter set have the same 3-dB frequency, as measured from the
shelf. The only difference in the 150 filters in a given bass filter set is the gain of the shelf.
Each treble filter set has 150 high-pass shelving filters to choose from, with the shelves again ranging from a cut
(attenuation) of 18 dB to a boost (gain) of 18 dB. A shelf selection of 0 dB effectively removes treble processing. All
150 filters in a given filter set have the same 3-dB frequency, as measured from the shelf. The only difference in the
150 filters in a given treble filter set is the gain of the shelf.
Commands to adjust the bass and treble levels within a given filter set (by commanding the selection of different
shelving filters) results in a soft adjustment to the newly commanded levels. The filters are labeled 1 through 150,
with filter #1 implementing the maximum cut (18 dB) and filter #150 implementing the maximum boost (18 dB). If a
command is received to change a shelf setting, the transition is made by stepping through each filter, one at a time,
until the shelf filter commanded is reached. A soft transition is achieved by residing at each step (or shelf filter) for
a period determined by the programmable parameter TBLC (bit field 7:0 of subaddress 0xF1). The time period set
by TBLC is TBLC/FS (where FS is the audio sample rate). For an audio sample rate of 48 kHz and a TBLC setting
of 64, the dwell time on each shelf filter is approximately 1.33 ms. The transition time then depends on the number
of shelves separating the current and commanded shelf values. For 48-kHz audio and a TBLC setting of 64, a
maximum transition time of approximately 198 ms is required to transition from shelf #1 to shelf #150 and a minimum
time of approximately 1.33 ms is required to transition from shelf x to shelf x + 1.
3−11
Treble/Bass Slew Rate Selection
S
0xF1
Slave Addr
7
Ack Sub-Addr Ack
CH2
CH3
00000000
Ack
00000000
V
Ack 0000000 C Ack
S
S
Slave Addr
Ack Sub-Addr Ack
00000000
Ack
00000xxx
Ack
00000xxx
Ack
Ack
Treble/Bass Slew Rate = TBLC
(Slew Rate = TBLC/FS,
Where FS = Audio Sample Rate)
CH1
Bass Filter Set Selection
0xF5
0
xxxxxxxx
00000xxx
Ack
CH2
CH3
CH1
Treble Filter Set Selection
S
0xF7
Slave Addr
Ack Sub-Addr Ack
00000000
CH2
CH1
CH3
Ack
00000xxx
Ack
00000xxx
Ack
00000xxx
Ack
Bass Shelf Selection (Filter Index)
0xF6
S
Slave Addr
Ack Sub-Addr Ack
00000000
Ack
xxxxxxxx
Ack
xxxxxxxx
Ack
xxxxxxxx
Ack
CH2
CH3
CH1
Treble Shelf Selection (Filter Index)
0xF8
BASS
FILTER 5
BASS
FILTER 3
S
Slave Addr
Ack Sub-Addr Ack
BASS
FILTER 1
MAX BOOST
SHELF
00000000
Ack
TREBLE
FILTER 5
xxxxxxxx
Ack
xxxxxxxx
TREBLE
FILTER 3
Ack
xxxxxxxx
Ack
TREBLE
FILTER 1
MID-BAND
BASS
FILTER 4
MAX CUT
SHELF
BASS
FILTER 2
TREBLE
FILTER 4
TREBLE
FILTER 2
FREQUENCY
Treble & Bass Filter Set Commands
0 => No Change
1 − 5 => Filter Sets 1 − 5
6 − 7 => Illegal (Behavior Indeterminate)
Treble & Bass Filter Shelf Commands
0 => Illegal (Behavior Indeterminate)
1 − 150 => Filter Shelves 1 − 150
1
=> +18-dB Boost
150 => −18-dB Cut
151 − 255 => Illegal (Behavior Indeterminate)
3-dB CORNERS (kHz)
FS
(LRCLK)
FILTER SET 5
FILTER SET 4
FILTER SET 3
FILTER SET 2
FILTER SET 1
BASS
TREBLE
BASS
TREBLE
BASS
TREBLE
BASS
TREBLE
BASS
TREBLE
0.25
6
0.5
12
0.75
18
1
24
1.5
36
88.4 kHz
0.23
5.525
0.46
11.05
0.691
16.575
0.921
22.1
1.381
33.15
64 kHz
0.167
4
0.333
8
0.5
12
0.667
16
1
24
96 kHz
48 kHz
0.125
3
0.25
6
0.375
9
0.5
12
0.75
18
44.1 kHz
0.115
2.756
0.23
5.513
0.345
8.269
0.459
11.025
0.689
16.538
38 kHz
0.099
2.375
0.198
4.75
0.297
7.125
0.396
9.5
0.594
14.25
32 kHz
0.083
2
0.167
4
0.25
6
0.333
8
0.5
12
24 kHz
0.063
1.5
0.125
3
0.188
4.5
0.25
6
0.375
9
22.05 kHz
0.057
1.378
0.115
2.756
0.172
4.134
0.23
5.513
0.345
8.269
16 kHz
0.042
1
0.083
2
0.125
3
0.167
4
0.25
6
12 kHz
0.031
0.75
0.063
1.5
0.094
2.25
0.125
3
0.188
4.5
11.025 kHz
0.029
0.689
0.057
1.378
0.086
2.067
0.115
2.756
0.172
4.134
Figure 3−12. Bass and Treble Filter Selections
3−12
CAUTION: There is no soft transition implemented when changing bass and treble filter
sets; soft transitions only apply when adjusting gains (shelves) within a given filter set.
The variable TBLC should be set so that the dwell time at each shelf is never less than
32 audio sample periods; otherwise audio artifacts could be introduced into the audio
data stream.
Figure 3−12 summarizes the bass and treble adjustments available within each monaural channel. As noted in
Figure 3−12, the 3-dB frequency for the bass filter sets decreases in value as the filter set number is increased,
whereas the 3-dB frequency for the treble filter set increases in value as the filter set number is increased. The valid
selection for bass and treble sets ranges from 1 to 5.
Table 3−2 and Table 3−3 list, respectively, the bass and treble filter shelf selections for all 1/2 dB settings between
18-dB cut and 18-dB boost. The treble and bass selections are not the same, and the delta in the selection values
between 1/2 dB points are not constant across the 36-dB range. Table 3−2 and Table 3−3 do not list all 150 filter shelf
selections, but all 150 selection values for both bass and treble are valid, allowing the use of linear potentiometer or
GUI-based sliders. Table 3−2 and Table 3−3 are provided for those applications requiring the adjustment of bass and
treble in 1/2 dB steps.
CAUTION: Filter set selections 6 and 7 are illegal. Filter shelf selections 0 and 151
through 255 are illegal. Programming an illegal value could result in erratic and
erroneous behavior.
As an example, consider the case of a 44.1-kHz audio sample rate. For this audio rate it is desired to have, for all
three monaural channels
•
•
•
•
A 3-dB bass shelf corner frequency of 100 Hz
A bass shelf volume boost of 9 dB
A 3-dB treble shelf corner frequency of 8.1 kHz
A treble shelf volume cut of 4 dB
Bass and treble filter set selections can be made by referring to Figure 3−12. For a 44.1-kHz audio sample rate, filter
set 5 provides a 3-dB bass corner frequency at 115 Hz, and filter set 3 provides a 3-dB treble corner frequency at
8.269 kHz. These corner frequencies are the closest realizable corner frequencies to the specified 100-Hz bass and
8.1-kHz treble corner frequencies.
Table 3−2 provides the indices for achieving specified bass volume levels. An index of 0x55 yields a bass shelf gain
of 9 dB, which matches the specified shelf volume boost of 9 dB. Table 3−3 provides the indices for achieving specified
treble volume levels. An index of 0x7A yields a treble shelf cut of 4 dB (−4-dB gain), which matches the specified shelf
volume cut of 4 dB.
Figure 3−13 presents the resulting subaddress entries required to implement the parameters specified in the bass
and treble example.
3−13
CH3
CH2
CH1
Bass Filter Set Selection
0xF5
S
Slave Addr
Ack Sub-Addr Ack
00000000
Ack
00000101
Ack
CH3
00000101
Ack
CH2
00000101
Ack
Filter Set 5 Selected
Ack
Filter Set 3 Selected
Ack
Filter Shelf 0x55 Selected
CH1
Treble Filter Set Selection
0xF7
S
Slave Addr
Ack Sub-Addr Ack
00000000
Ack
00000011
Ack
CH3
00000011
Ack
CH2
00000011
CH1
Bass Shelf Selection (Filter Index)
0xF6
S
Slave Addr
Ack Sub-Addr Ack
00000000
Ack
01010101
Ack
CH3
01010101
Ack
CH2
01010101
CH1
Treble Shelf Selection (Filter Index)
0xF8
S
Slave Addr
Ack Sub-Addr Ack
00000000
Ack
01111010
Ack
01111010
Ack
01111010
Ack
Filter Shelf 0x7A Selected
Figure 3−13. Bass and Treble Application Example—Subaddress Parameters
Table 3−2. Bass Shelf Filter Indices for 1/2-dB Adjustments
ADJUSTMENT
(DB)
INDEX(1)
ADJUSTMENT
(DB)
INDEX(1)
ADJUSTMENT
(DB)
INDEX(1)
18
0x01
5.5
0x63
−7
0x80
17.5
0x08
5
0x64
−7.5
0x81
17
0x10
4.5
0x66
−8
0x82
16.5
0x16
4
0x67
−8.5
0x83
16
0x1D
3.5
0x69
−9
0x84
15.5
0x23
3
0x6A
−9.5
0x85
15
0x28
2.5
0x6B
−10
0x86
14.5
0x2D
2
0x6D
−10.5
0x87
14
0x32
1.5
0x6E
−11
0x88
13.5
0x37
1
0x6F
−11.5
0x89
13
0x3B
0.5
0x71
−12
0x8A
12.5
0x3F
0
0x72
−12.5
0x8B
12
0x42
−0.5
0x73
−13
0x8C
11.5
0x46
−1
0x74
−13.5
0x8D
11
0x49
−1.5
0x75
−14
0x8E
10.5
0x4C
−2
0x76
−14.5
0x8F
10
0x4F
−2.5
0x77
−15
0x90
9.5
0x52
−3
0x78
−15.5
0x91
9
0x55
−3.5
0x79
−16
0x92
8.5
0x58
−4
0x7A
−16.5
0x93
8
0x5A
−4.5
0x7B
−17
0x94
7.5
0x5C
−5
0x7C
−17.5
0x95
−18
0x96
7
0x5E
−5.5
0x7D
6.5
0x60
−6
0x7E
6
0x62
−6.5
0x7F
(1) CH1 Index is Subaddress 0xF6, Bit Field 7:0. CH2 Index is Subaddress 0xF6, Bit Field 15:8. CH3
Index is Subaddress 0xF6, Bit Field 23:16.
3−14
Table 3−3. Treble Shelf Filter Indices for 1/2-dB Adjustments
ADJUSTMENT
(DB)
INDEX(1)
ADJUSTMENT
(DB)
INDEX(1)
ADJUSTMENT
(DB)
INDEX(1)
18
0x01
5.5
0x63
−7
0x80
17.5
0x09
5
0x65
−7.5
0x81
17
0x10
4.5
0x66
−8
0x82
16.5
0x16
40
0x68
−8.5
0x83
16
0x1C
3.5
0x69
−9
0x84
15.5
0x22
3
0x6B
−9.5
0x85
15
0x28
2.5
0x6C
−10
0x86
14.5
0x2D
2
0x6D
−10.5
0x87
0x88
14
0x31
1.5
0x6F
−11
13.5
0x35
1
0x70
−11.5
0x89
13
0x3A
0.5
0x71
−12
0x8A
12.5
0x3E
0
0x72
−12.5
0x8B
12
0x42
−0.5
0x73
−13
0x8C
11.5
0x45
−1
0x74
−13.5
0x8D
11
0x49
−1.5
0x75
−14
0x8E
10.5
0x4C
−2
0x76
−14.5
0x8F
10
0x4F
−2.5
0x77
−15
0x90
9.5
0x52
−3
0x78
−15.5
0x91
9
0x55
−3.5
0x79
−16
0x92
8.5
0x57
−4
0x7A
−16.5
0x93
8
0x5A
−4.5
0x7B
−17
0x94
7.5
0x5C
−5
0x7C
−17.5
0x95
7
0x5E
−5.5
0x7D
−18
0x96
6.5
0x60
−6
0x7E
6
0x62
−6.5
0x7F
(1) CH1 Index is Subaddress 0xF8, Bit Field 7:0. CH2 Index is Subaddress 0xF8, Bit Field 15:8. CH3
Index is Subaddress 0xF8, Bit Field 23:16.
3.5.1
Treble and Bass Processing and Concurrent I2C Read Transactions
I2C read transactions at subaddresses 0x01 through 0xD1 are not allowed during: (1) transitions between filter
shelves within a given bass or treble filter set or (2) transitions between filter sets. This means that after issuing an
I2C command to change treble or bass filter shelves or filter sets, time must be allowed to complete the bass/treble
activity before proceeding to read subaddresses 0x01 through 0xD1. There is no subaddress available to directly
monitor bass/treble activity. However, there is a means of probing the state of bass/treble activity via the factory test
subaddresses 0xEC and 0xED.
If it is required to read subaddress data that falls in the subaddress range of 0x01 through 0xD1 after issuing an I2C
command to change treble or bass filter shelves or filter sets, the procedure presented in Figure 3−14 can be used
to monitor bass/treble activity. As Figure 3−14 shows, six readings must be taken to verify that none of the three
monaural channels have on-going bass or treble activity. If all six readings show the value zero in the 8th or least
significant byte of the 8-byte data word output at subaddress 0xED, then all commanded bass/treble activity has
completed and it is safe to resume I2C read transactions at subaddresses 0x01 through 0xD1.
3−15
Start
Write 0x00, 0x06, 0x00, 0xCD
subaddress 0xEC
Read 8 byte output from
subaddress 0xED
N
8th Byte = 0
CH 1 Treble
Y
Write 0x00, 0x06, 0x00, 0xD1
subaddress 0xEC
Read 8 byte output from
subaddress 0xED
N
8th Byte = 0
CH 1 Bass
Y
Write 0x00, 0x06, 0x00, 0xCE
subaddress 0xEC
Read 8 byte output from
subaddress 0xED
N
8th Byte = 0
CH 2 Treble
Y
Write 0x00, 0x06, 0x00, 0xD2
subaddress 0xEC
Read 8 byte output from
subaddress 0xED
N
8th Byte = 0
Y
Write 0x00, 0x06, 0x00, 0xD5
subaddress 0xEC
Read 8 byte output from
subaddress 0xED
N
8th Byte = 0
Read 8 byte output from
subaddress 0xED
Bass / Treble Processing Active
8th Byte = 0
CH 3 Treble
Y
Write 0x00, 0x06, 0x00, 0xD3
subaddress 0xEC
N
CH 2 Bass
CH 3 Bass
Y
Bass / Treble Processing Inactive
Figure 3−14. I2C Bass/Treble Activity Monitor Procedure
3−16
3.6 Soft Volume/Loudness Processing
Each of the three monaural channels in the TAS3103 has dedicated soft volume control and loudness compensation.
Volume level changes are issued by I2C bus commands in the I2C slave mode and by setting the appropriate GPIO
pin to logic 0 in the I2C master mode. Commanded changes in volume are implemented softly, using a smooth S-curve
trajectory to transition the volume to the newly commanded level.
Volume commands are formatted as signed 5.23 numbers. The maximum volume boost then is 24 dB
(4 bits x 6 dB/bit); the maximum volume cut is ∞ with a volume command of zero. The maximum finite volume cut
is 138 dB (23 bits x 6 dB/bit). The resolution of the volume adjustment is not fixed over all gain settings. For large gains
the resolution of the volume adjustment is very fine, and the resolution of the volume adjustment decreases as the
volume gain decreases. As an example, at maximum gain, the volume level can be adjusted to a resolution of
0.000001 dB [(138 + 24) dB adjustment range/227 adjustment steps]. At the other end of the gain scale, if the volume
setting is at the maximum finite cut (volume command = 0x0000001) and is increased by one count (volume command
= 0x0000002), a 6-dB adjustment is realized.
Each monaural channel volume control is also assigned a separate mute command, which has the same effect as
issuing a zero-valued volume command. If loudness is enabled, disabling it by setting the parameter G to zero is
necessary to obtain a total cut (−∞ dB). This requirement is further discussed in the paragraphs that follow.
Loudness compensation tracks the volume control setting to allow spectral compensation. An example of loudness
compensation would be a boost in bass frequencies to compensate for weak perceived bass at low volume levels.
Both linear and log control laws can be implemented for volume gain tracking, and a dedicated biquad filter can be
used to achieve spectral discernment.
3.6.1
Soft Volume
Figure 3−15 is a simplified block diagram of the implementation of soft volume and loudness compensation. A volume
level change (either via an I2C bus issued command in the I2C slave mode, or a GPIO-issued command in the I2C
master mode) initiates a transition process that assures a smooth transition to the newly commanded volume level
without producing artifacts such as pops, clicks, and zipper noise. The transition time, or volume slew rate, can be
selected to occupy a time window of either 2048 or 4096 audio sample periods. For 48-kHz audio, for example, this
equates to a transition time of 42.67 ms or 85.34 ms. It is anticipated that 42.67 ms is the transition time of choice
for most applications, and the 4096 sample transition option is primarily included to yield the same 42.67-ms transition
time for 96-kHz audio. The slope of the S-curve (and its implementation) is proprietary and cannot be altered.
If additional volume commands for a given monaural channel are received, while a previously commanded volume
change is still active, the last command received over-writes previous commands and is used. When the previously
commanded transition completes, the volume command last received while the transition was taking place is acted
upon and a new transition to this volume level initiated. For example, assume three volume commands are
sequentially issued, via the I2C bus, to adjust the volume levels of the three monaural channels in the TAS3103. The
first command received immediately triggers the start of a soft volume transition to the newly commanded level. The
other two volume commands are received and queued to await the completion of the currently active soft volume
transaction. When the first soft volume transition completes, and assuming no further volume commands have been
received to replace the other two volume commands received, the next two volume commands are activated, and
soft volume transitions on both monaural channels takes place. The total time then, for a 48-kHz audio sample rate
and a transition selection of 2048 FS cycles is 2 x 42.67 ms or 85.34 ms. If, during the first soft volume transition, a
second volume command is received for this same monaural channel, the second soft volume transition period would
have soft volume transitions taking place on all three monaural channels. It is also noted that the soft volume transition
time is independent of the magnitude of the adjustment. All volume commands take 2048 or 4096 FS cycles,
regardless of the magnitude of the change.
In the I2C slave mode, the status of a commanded soft volume transition can be found by reading bit 0 of the 32-bit
data word retrieved at subaddress 0xFF. If this bit is set to logic 1, one or more monaural channels are actively
transitioning their volume setting. If a volume transition is taking place on one of the monaural channels in the
TAS3103, volume commands received for the other two monaural channels are not acted upon until the active volume
transition completes. When the active volume transition does complete, the latest volume commands received for
the three monaural channels during the previous soft volume transition time are serviced.
3−17
Soft Volume
2048 Sample
Transition
4096 Sample
Transition
Microprocessor
I2C
Volume
Commands
GPIO
Volume
Commands
I2C Slave Mode
VSCSubaddress 0xF1 = 0
VSCSubaddress 0xF1 = 1
Soft Volume
Gain Control
volume_setting
I2C Master Mode
Loudness Compensation
f (Volume)
Programmable
Biquad Filter
Channel-Processed
Audio
Volume-Adjusted
Audio
Figure 3−15. Soft Volume and Loudness Compensation Block Diagram
Figure 3−15 is a more detailed block diagram of soft volume and loudness compensation, and includes the I2C
subaddress commands that control volume and loudness compensation. Volume control is accomplished using three
I2C subaddresses—volume control, mute/unmute control, and volume slew rate control.
Volume control in the TAS3103 applies a linear gain. The volume commands issued via I2C subaddresses 0xF2,
0xF3, and 0xF4 (monaural channels 1, 2, and 3 respectively) are signed 5.23 format numbers. These commands are
applied to mixers, whose other input port is the audio data stream. The mixer output is the product of the audio data
stream and the volume command. Examples of volume command settings are given below.
Volume Command = 0x3580B07 = 0011_0.101_1000_0000_1011_0000_0111 = 6.6878365
Volume Command = 0x01F0000 = 0000_0.001_1111_0000_0000_0000_0000_0000 = 0.2421875
Volume Command = 0xCD6EFFE = 1100_1.101_0110_1110_1111_1111_1110 = −6.3208010
The volume control range is 0 = −∞ to 2−23 = −138.47 dB to 24 − 2−23 = 24.08 dB. Volume control is achieved by means
of a 5.23 format gain coefficient that is applied to a linear mixer. The volume gain setting realized, for a given volume
gain coefficient is:
Gain = 20log (Volume_Gain_Coefficient)
3−18
There are several techniques of volume management for a linear volume control process.
•
Precise calculations involving logarithms can be employed.
•
A high-resolution gain table, with entries for every 0.5-dB step, can be employed.
•
A more coarse gain table (entries in 3 to 6-dB steps with linear interpolation between entries) can be
employed.
•
Or approximations involving very simple calculations can be employed.
As an example of using approximations, equations for increasing a linear 5.23 gain setting X by 0.5 dB that involve
only simple binary shift and add operations and the accuracy of these equations are given below.
20log10X + 0.5 dB ≈ X + 2−4X gives 0.52657877-dB steps
20log10X + 0.5 dB ≈ X + 2−4X − 2−8X gives 0.49458651-dB steps
20log10X + 0.5 dB ≈ X + 2−4X − 2−8X + 2−11X gives 0.49859199-dB steps
20log10X + 0.5 dB ≈ X[1 + 2−4 − 2−8 + 2−11 + 2−12 − 2−13 + 2−14 − 2−16 + 2−18] gives
0.4999997332 dB−steps
Approximations can also be found for decreasing a linear 5.23 gain setting X by 0.5 dB that involve using only simple
binary shift and add operations, but the equations differ slightly from those used to increase the gain in 0.5-dB steps.
The approximations to decrease the volume by 0.5 dB and the accuracy of these approximations are given below.
20log10X − 0.5 dB ≈ X − 2−4X gives −0.56057447 dB-steps
20log10X − 0.5 dB ≈ X − 2−4X + 2−7X gives −0.48849199-dB steps
20log10X − 0.5 dB ≈ X − 2−4X + 2−7X − 2−10X gives −0.49746965-dB steps
20log10X − 0.5 dB ≈ X[1 − 2−4 + 2−7 − 2−10 − 2−12 − 2−15 − 2−21] gives −0.5000006792 dB-steps
Repeated use of a set of the above approximations results in an accumulation of the errors in the approximations.
For example, if an application started at 0-dB volume, and repeatedly used the approximations to increase and
decrease the volume, the exact reference point of 0 dB would be lost. If an application does require the maintenance
of accurate reference points, it is necessary for the application to establish a set of exact gain reference settings and
command these exact settings in place of a computed gain setting whenever the current gain setting and the next
computed gain setting straddle an exact gain setting.
Table 3−4 lists the I2C coefficient settings to adjust volume from 24 dB to −136 dB in 0.5-dB steps. For each volume
setting, the gain in dB is presented in one column, the same gain in a floating point number
ǒfloat + 10
Gain
Ǔ is
dB
20
presented in the adjacent column, and the same gain formatted in the 32-bit hexadecimal gain coefficient format
required to enter the value into the TAS3103 via the I2C bus is presented in a third column.
3−19
Table 3−4. Volume Adjustment Gain Coefficients
3−20
GAIN (dB)
GAIN (FLOAT)
GAIN (COEFFICIENT)
GAIN (dB)
GAIN (FLOAT)
GAIN (COEFFICIENT)
24
15.84893192
07ECA9CD
2
1.25892541
00A12477
23.5
14.96235656
077B2E7F
1.5
1.18850223
009820D7
23
14.12537545
07100C4D
1
1.12201845
008F9E4C
22.5
13.33521432
06AAE84D
0.5
1.05925373
008795A0
22
12.58925412
064B6CAD
0
1
00800000
21.5
11.88502227
05F14868
−0.5
0.94406088
0078D6FC
21
11.22018454
059C2F01
−1
0.89125094
00721482
20.5
10.59253725
054BD842
−1.5
0.84139514
006BB2D6
20
10
05000000
−2
0.79432823
0065AC8C
19.5
9.44060876
04B865DE
−2.5
0.74989421
005FFC88
19
8.91250938
0474CD1B
−3
0.70794578
005A9DF7
18.5
8.41395142
0434FC5C
−3.5
0.66834392
00558C4B
18
7.94328235
03F8BD79
−4
0.63095734
0050C335
17.5
7.49894209
03BFDD55
−4.5
0.59566214
004C3EA8
17
7.07945784
038A2BAC
−5
0.56234133
0047FACC
16.5
6.68343918
03577AEF
−5.5
0.53088444
0043F405
16
6.30957344
0327A01A
−6
0.50118723
004026E7
15.5
5.95662144
02FA7292
−6.5
0.47315126
003C9038
15
5.62341325
02CFCC01
−7
0.44668359
00392CED
14.5
5.30884444
02A78836
−7.5
0.42169650
0035FA26
14
5.01187234
02818508
−8
0.39810717
0032F52C
13.5
4.73151259
025DA234
−8.5
0.37583740
00301B70
13
4.46683592
023BC147
−9
0.35481339
002D6A86
12.5
4.21696503
021BC582
−9.5
0.33496544
002AE025
12
3.98107171
01FD93C1
−10
0.31622777
00287A26
11.5
3.75837404
01E11266
−10.5
0.29853826
00263680
11
3.54813389
01C62940
−11
0.28183829
00241346
10.5
3.34965439
01ACC179
−11.5
0.26607251
00220EA9
10
3.16227766
0194C583
−12
0.25118864
002026F3
9.5
2.98538262
017E2104
−12.5
0.23713737
001E5A84
9
2.81838293
0168C0C5
−13
0.22387211
001CA7D7
8.5
2.66072506
015492A3
−13.5
0.21134890
001B0D7B
8
2.51188643
0141857E
−14
0.19952623
00198A13
7.5
2.37137371
012F892C
−14.5
0.18836491
00181C57
7
2.23872114
011E8E6A
−15
0.17782794
0016C310
6.5
2.11348904
010E86CF
−15.5
0.16788040
00157D1A
6
1.99526231
00FF64C1
−16
0.15848932
00144960
5.5
1.88364909
00F11B69
−16.5
0.14962357
001326DD
5
1.77827941
00E39EA8
−17
0.14125375
0012149A
4.5
1.67880402
00D6E30C
−17.5
0.13335214
001111AE
4
1.58489319
00CADDC7
−18
0.12589254
00101D3F
3.5
1.49623566
00BF84A6
−18.5
0.11885022
000F367B
3
1.41253754
00B4CE07
−19
0.11220185
000E5CA1
2.5
1.33352143
00AAB0D4
−19.5
0.10592537
000D8EF6
Table 3−4. Volume Adjust Gain Coefficient (Continued)
GAIN (dB)
GAIN (FLOAT)
GAIN (COEFFICIENT)
GAIN (dB)
GAIN (FLOAT)
GAIN (COEFFICIENT)
−20
0.1
000CCCCC
−42
0.00794328
00010449
−20.5
0.09440609
000C157F
−42.5
0.00749894
0000F5B9
−21
0.08912509
000B6873
−43
0.00707946
0000E7FA
−21.5
0.08413951
000AC515
−43.5
0.00668344
0000DB00
−22
0.07943282
000A2ADA
−44
0.00630957
0000CEC0
−22.5
0.07498942
00099940
−44.5
0.00595662
0000C32F
−23
0.07079458
00090FCB
−45
0.00562341
0000B844
−23.5
0.06683439
00088E07
−45.5
0.00530884
0000ADF5
−24
0.06309573
00081385
−46
0.00501187
0000A43A
−24.5
0.05956621
00079FDD
−46.5
0.00473151
00009B0A
−25
0.05623413
000732AE
−47
0.00446684
0000925E
−25.5
0.05308844
0006CB9A
−47.5
0.00421697
00008A2E
−26
0.05011872
00066A4A
−48
0.00398107
00008273
−26.5
0.04731513
00060E6C
−48.5
0.00375837
00007B27
−27
0.04466836
0005B7B1
−49
0.00354813
00007443
−27.5
0.04216965
000565D0
−49.5
0.00334965
00006DC2
−28
0.03981072
00051884
−50
0.00316228
0000679F
−28.5
0.03758374
0004CF8B
−50.5
0.00298538
000061D3
−29
0.03548134
00048AA7
−51
0.00281838
00005C5A
−29.5
0.03349654
0004499D
−51.5
0.00266073
0000572F
−30
0.03162278
00040C37
−52
0.00251189
0000524F
−30.5
0.02985383
0003D240
−52.5
0.00237137
00004DB4
−31
0.02818383
00039B87
−53
0.00223872
0000495B
−31.5
0.02660725
000367DD
−53.5
0.00211349
00004541
−32
0.02511886
00033718
−54
0.00199526
00004161
−32.5
0.02371374
0003090D
−54.5
0.00188365
00003DB9
−33
0.02238721
0002DD95
−55
0.00177828
00003A45
−33.5
0.02113489
0002B48C
−55.5
0.00167880
00003702
−34
0.01995262
00028DCE
−56
0.00158489
000033EF
−34.5
0.01883649
0002693B
−56.5
0.00149624
00003107
−35
0.01778279
000246B4
−57
0.00141254
00002E49
−35.5
0.01678804
0002261C
−57.5
0.00133352
00002BB2
−36
0.01584893
00020756
−58
0.00125893
00002940
−36.5
0.01496236
0001EA49
−58.5
0.00118850
000026F1
−37
0.01412538
0001CEDC
−59
0.00112202
000024C4
−37.5
0.01333521
0001B4F7
−59.5
0.00105925
000022B5
−38
0.01258925
00019C86
−60
0.001
000020C4
−38.5
0.01188502
00018572
−60.5
0.00094406
00001EEF
−39
0.01122018
00016FA9
−61
0.00089125
00001D34
−39.5
0.01059254
00015B18
−61.5
0.00084140
00001B92
−40
0.01
000147AE
−62
0.00079433
00001A07
−40.5
0.00944061
00013559
−62.5
0.00074989
00001892
−41
0.00891251
0001240B
−63
0.00070795
00001732
−41.5
0.00841395
000113B5
−63.5
0.00066834
000015E6
3−21
Table 3−4. Volume Adjust Gain Coefficient (Continued)
3−22
GAIN (dB)
GAIN (FLOAT)
GAIN (COEFFICIENT)
GAIN (dB)
GAIN (FLOAT)
GAIN (COEFFICIENT)
−64
0.00063096
000014AC
−86
5.01188E−05
000001A4
−64.5
0.00059566
00001384
−86.5
4.73152E−05
0000018C
−65
0.00056234
0000126D
−87
4.46684E−05
00000176
−65.5
0.00053088
00001165
−87.5
4.21696E−05
00000161
−66
0.00050119
0000106C
−88
3.98108E−05
0000014D
−66.5
0.00047315
00000F81
−88.5
3.75838E−05
0000013B
−67
0.00044668
00000EA3
−89
3.54814E−05
00000129
−67.5
0.00042170
00000DD1
−89.5
3.34966E−05
00000118
−68
0.00039811
00000D0B
−90
3.16228E−05
00000109
−68.5
0.00037584
00000C50
−90.5
2.98538E−05
000000FA
−69
0.00035481
00000BA0
−91
2.81838E−05
000000EC
−69.5
0.00033497
00000AF9
−91.5
2.66072E−05
000000DF
−70
0.00031623
00000A5C
−101
8.91250E−06
0000004A
−70.5
0.00029854
000009C8
−101.5
8.41396E−06
00000046
−71
0.00028184
0000093C
−102
7.94328E−06
00000042
−71.5
0.00026607
000008B7
−102.5
7.49894E−06
0000003E
−72
0.00025119
0000083B
−103
7.07946E−06
0000003B
−72.5
0.00023714
000007C5
−103.5
6.68344E−06
00000038
−73
0.00022387
00000755
−104
6.30958E−06
00000034
−73.5
0.00021135
000006EC
−104.5
5.95662E−06
00000031
−74
0.00019953
00000689
−105
5.62342E−06
0000002F
−74.5
0.00018837
0000062C
−105.5
5.30884E−06
0000002C
−75
0.00017783
000005D3
−106
5.01188E−06
0000002A
−75.5
0.00016788
00000580
−106.5
4.73152E−06
00000027
−76
0.00015849
00000531
−107
4.46684E−06
00000025
−76.5
0.00014962
000004E7
−107.5
4.21696E−06
00000023
−77
0.00014125
000004A0
−108
3.98108E−06
00000021
−77.5
0.00013335
0000045E
−108.5
3.75838E−06
0000001F
−78
0.00012589
00000420
−109
3.54814E−06
0000001D
−78.5
0.00011885
000003E4
−109.5
3.34966E−06
0000001C
−79
0.00011220
000003AD
−110
3.16228E−06
0000001A
−79.5
0.00010593
00000378
−110.5
2.98538E−06
00000019
−80
0.0001
00000346
−111
2.81838E−06
00000017
−80.5
9.44060E−05
00000317
−111.5
2.66072E−06
00000016
−81
8.91250E−05
000002EB
−112
2.51188E−06
00000015
−81.5
8.41396E−05
000002C1
−112.5
2.37138E−06
00000013
−82
7.94328E−05
0000029A
−113
2.23872E−06
00000012
−82.5
7.49894E−05
00000275
−113.5
2.11348E−06
00000011
−83
7.07946E−05
00000251
−92
2.51188E−05
000000D2
−83.5
6.68344E−05
00000230
−92.5
2.37138E−05
000000C6
−84
6.30958E−05
00000211
−93
2.23872E−05
000000BB
−84.5
5.95662E−05
000001F3
−93.5
2.11348E−05
000000B1
−85
5.62342E−05
000001D7
−94
1.99526E−05
000000A7
−85.5
5.30884E−05
000001BD
−94.5
1.88365E−05
0000009E
Table 3−4. Volume Adjust Gain Coefficient (Continued)
GAIN (dB)
GAIN (FLOAT)
GAIN (COEFFICIENT)
GAIN (dB)
GAIN (GLOAT)
GAIN (COEFFICIENT)
−95
1.77828E−05
00000095
−122.5
7.49894E−07
00000006
−95.5
1.67880E−05
0000008C
−123
7.07946E−07
00000005
−96
1.58489E−05
00000084
−123.5
6.68344E−07
00000005
−96.5
1.49624E−06
0000007D
−124
6.30958E−07
00000005
−97
1.41254E−05
00000076
−124.5
5.95662E−07
00000004
−97.5
1.33352E−05
0000006F
−125
5.62342E−07
00000004
−98
1.25893E−05
00000069
−125.5
5.30884E−07
00000004
−98.5
1.18850E−05
00000063
−126
5.01188E−07
00000004
−99
1.12202E−05
0000005E
−126.5
4.73152E−07
00000003
−99.5
1.05925E−05
00000058
−127
4.46684E−07
00000003
−100
0.00001
00000053
−127.5
4.21696E−07
00000003
−100.5
9.44060E−06
0000004F
−128
3.98108E−07
00000003
−114
1.99526E−06
00000010
−128.5
3.75838E−07
00000003
−114.5
1.88365E−06
0000000F
−129
3.54814E−07
00000002
−115
1.77828E−06
0000000E
−129.5
3.34966E−07
00000002
−115.5
1.67880E−06
0000000E
−130
3.16228E−07
00000002
−116
1.58489E−06
0000000D
−130.5
2.98538E−07
00000002
−116.5
1.49624E−06
0000000C
−131
2.81838E−07
00000002
−117
1.41254E−06
0000000B
−131.5
2.66072E−07
00000002
−117.5
1.33352E−06
0000000B
−132
2.51188E−07
00000002
−118
1.25893E−06
0000000A
−132.5
2.37138E−07
00000001
−118.5
1.18850E−06
00000009
−133
2.23872E−07
00000001
−119
1.12202E−06
00000009
−133.5
2.11348E−07
00000001
−119.5
1.05925E−06
00000008
−134
1.99526E−07
00000001
−120
0.000001
00000008
−134.5
1.88365E−07
00000001
−120.5
9.44080E−07
00000007
−135
1.77828E−07
00000001
−121
8.91250E−07
00000007
−135.5
1.67880E−07
00000001
−121.5
8.41396E−07
00000007
−136
1.58489E−07
00000001
−122
7.94328E−07
00000006
3.6.1.1 Soft Volume Adjustment Range Limitations
When the 2048 sample transition time is selected for the S-curve volume transition, the full −∞ to 24-dB adjustment
range is available. All possible values but one in the signed 28-bit (5.23 format) volume level command range are
valid volume level selections. The one exception is the maximum negative volume level command 0x08000000. This
value is illegal and if used, could result in erratic behavior that requires a reset to the part to correct.
When the 4096 sample transition time is selected, the upper two bits of the 28-bit volume command cannot be used.
This means that the valid volume level command range is reduced to a maximum positive value of 0x1FFFFFF
(12 dB) and a maximum negative value of 0x0E000000 (also 12 dB, but with a 180° phase inversion of the audio
signal). Values above these maximum levels could result in erratic behavior that requires a reset to the part to correct.
Although the volume boost for the 4096 sample transition time is limited to 12 dB, volume boost can be realized
elsewhere in the processing signal flow, such as in the input crossbar mixers, the biquad filters in the three monaural
channels, or the biquad filters in the effects block.
3.6.1.2 Soft Volume Transitions and Concurrent I2C Read Transactions
I2C read transactions at subaddresses 0x01 through 0xD1 are not allowed during volume level transitions, as read
activity during volume level transitions could result in erroneous data being output. If it is required to read subaddress
3−23
data that falls in the subaddress range 0x01 through 0xD1 after issuing an I2C command to change the volume level
on one of the three monaural channels, the busy bit at subaddress 0xFF must be monitored to determine when the
volume activity has ceased and it is safe to resume I2C read activity at subaddresses 0x01 through 0xD1. A value
of 0 in the least significant bit of the byte output upon reading subaddress 0xFF signifies that all volume transition
activity has completed.
3.6.2
Loudness Compensation
Loudness compensation employs a single, coefficient-programmable, biquad filter and a function block
f(volume_setting), whose output is only a function of the volume control setting. The biquad filter input is the channel
processed audio data stream that also feeds the volume gain-control mixer. The biquad output feeds a gain control
mixer whose other input is the volume control setting after processing by the function block f(volume setting).
Loudness compensation then allows a given spectral segment of the audio data stream (as determined by the biquad
filter coefficients) to be given a delta adjustment in volume as determined by the programmable function block
f(volume_setting).
The output of the function block f(volume_setting) can be expressed in terms of the programmable I2C coefficients
as:
f(volume_setting) = [(volume_setting)LG x2LO xG] +O
where:
LG = logarithmic gain = 5.23 format number
LO = logarithmic offset = 25.23 format number
G = gain = 5.23 format number
O = offset = 25.23 format number
If LG = 0.5, LO = 0.0, G = 1.0, and O = 0.0, f(volume setting) becomes:
f(volume_setting) = √volume_setting
The output of the soft volume/loudness compensation block for the above case would be:
audioOut = [audioIn × volume_setting] + [audioIn × F(s)Biquad × √volume_setting ]
If a given monaural channel is set up as in the above example, a true mute is not obtained when a mute command
is issued for the monaural channel. A mute command results in a volume control setting of 0.0. LG and LO operate
in logarithmic space and the logarithm of 0.0 is − ∞. Therefore (volume_setting)−∞ and 2−∞ should both be of value
0. However, the function block f(volume_setting) approximates logarithmic arithmetic and a consequence of the
mathematical approximations used is that the function block can output a non-zero value for an input volume setting
of 0.0. For f(volume_setting) = √volume_setting, a volume_setting value of 0.0 results in the function block outputting
a gain coefficient that cuts the output of the biquad filter by approximately 90 dB. If true muting is required, it can be
achieved by setting G = 0.0 after the volume control setting has softly transitioned to 0.0.
3−24
3−25
Volume
Commands
I2C Bus
Slave Addr
Ack Sub-Addr Ack
S
S
xxxxxxxx
Ack xxxxxxxx
CCC
Ack xxxxxxxx Ack xxxxx HHH Ack
321
m
Ack Sub-Addr Ack 0000 s xxx Ack xxxxxxxx
b
Ack xxxxxxxx
l
Ack xxxxxxx s Ack
b
ttransition
Commanded
Volume
48
28
VCS
S
Ack xxxxxxxx
m
0000 s xxx Ack xxxxxxxx
b
a2
28
28
a1
Ack xxxxxxxx
Ack xxxxxxxx
m
0000 s xxx Ack xxxxxxxx
b
m
0000 s xxx Ack xxxxxxxx
b
Ack xxxxxxxx
Ack xxxxxxxx
( ) LG
m
0000 s xxx Ack xxxxxxxx
b
Σ
48
Σ
48
48
Ack
l
xxxxxx s Ack
b
LG
z−1
z−1
b2
b1
l
xxxxxx s Ack
b
l
xxxxxx s Ack
b
b0
a2
a1
l
xxxxxx s Ack
b
l
xxxxxx s Ack
b
l
xxxxxx s Ack
b
AUDIO OUT
Loudness Compensation
Ack
Ack
Ack
Ack
Ack
Σ
O
G
2LO
S
S
S
Slave Addr
Slave Addr
Slave Addr
LOUDNESS
LG Is A 5.23 Format Number
Ack xxxxxxxx
All biquad gain coefficients 5.23 numbers.
28
28
b2
b1
28
b0
Commanded 5.23
Volume Command
z−1
z−1
BiQuad Coefficients
CH 1 = 0xA6
CH 2 = 0xAB
CH 3 = 0xB0
m
Ack Sub-Addr Ack 0000 s xxx Ack xxxxxxxx
b
m
Ack Sub-Addr Ack 0000 s xxx Ack xxxxxxxx
b
Slave Addr
Slave Addr
S
CH 1 = 0xA2
CH 2 = 0xA7
CH 3 = 0xAC
xxxxxxxx
Ack Sub-Addr Ack
Ack xxxxxxxx
l
Ack xxxxxxx s Ack
b
l
Ack xxxxxxx s Ack
b
CH 1 = 0xA5
CH 2 = 0xAA
CH 3 = 0xAF
G Is A 5.23 Format Number
Ack xxxxxxxx
CH 1 = 0xA4
CH 2 = 0xA9
CH 3 = 0xAE
Ack xxxxxxxx
l
Ack xxxxxxx s Ack
b
O Is A 25.23 Format Number
Ack xxxxxxxx
m
00000000 Ack 00000000 Ack s xxxxxxx Ack xxxxxxxx Ack
b
xxxxxxxx
LO LSBs
LO MSBs
0 LSBs
0 MSBs
G
LO Is A 25.23 Format Number
Ack xxxxxxxx
m
00000000 Ack 00000000 Ack s xxxxxxx Ack xxxxxxxx Ack
b
CH 1 = 0xA3
CH 2 = 0xA8
CH 3 = 0xAD
m
Ack Sub-Addr Ack 0000 s xxx Ack xxxxxxxx
b
Ack Sub-Addr Ack
Figure 3−16. Detailed Block Diagram—Soft Volume and Loudness Compensation
AUDIO IN
= x16 BoostMAX
Volume Command = 1/223 Cut
MAX (LSB)
(5.23 Precision) = Zero Output For 0x0000000 Volume Control
Slave Addr
Mute Command = 1 => 0x0000000 Volume Control
CH 1 = 0xF2
CH 2 = 0xF3
Volume
CH 3 = 0xF4
Command
Ack Sub-Addr Ack
Original
Volume
Note: Negative Volume Commands Result In Audio Polarity Inversion
Slave Addr
SOFT VOLUME
v
Ack xxxxxxx c Ack xxxxxxxx Ack
s
0xF1
Ack xxxxxxxx
VCS = 0 ⇒ ttransition = 2048/FS
VCS = 1 ⇒ ttransition = 4096/FS
xxxxxxxx
Mute / Unmute Command
0xF0
I2C Slave Mode
I2C Master Mode
Volume Commands − GPIO Terminals
GPIO0 − Volume Up − CH1 / CH2
GPIO1 − Volume Down − CH1 / CH2
GPIO2 − Volume Up − CH3
GPIO3 − Volume Down − CH1 / CH2
S
If G is set to 0.0 and O is set to 0.0, loudness compensation is disabled. If G is set to 0.0 and O is set to 1.0, the
biquad-filtered audio is directly added to the volume level adjusted audio. Typically, LG and LO are used to derive
the desired loudness compensation function, G is used to turn loudness compensation on and off, and O is used to
enable and disable the biquad filter output when automatic volume tracking is turned off.
3.6.3
Time Alignment and Reverb Delay Processing
The TAS3103 provides delay line facilities at two locations in the TAS3103—in the 3D effects block (reverb delay),
and at the output mix block (delay). There are three reverb delay blocks, one for each monaural channel, and these
delay elements are typically used in implementing sound spatiality, a single mix reverberation, and other sound
effects. There are also three delay blocks, again one for each monaural channel, and these delay elements are
typically used for temporal channel alignment. The delay line facilities are implemented using a single 4K (4096) x
16-bit RAM resource. Each delay element implemented provides a one-sample delay (1/FS). The size of each delay
line can be programmed via the I2C bus, or set by the EEPROM download in the I2C master mode. The only restriction
is that the total delay line resources programmed cannot exceed the capacity of the 4K x 16-bit memory bank.
Figure 3−17 illustrates how delay line structures are established within the 4K RAM memory. As seen in Figure 3−17,
each delay line immediately begins where the previously implemented delay line leaves off. The actual placement
of the pointers is computed by the resident microprocessor; the user need only enter the delay value required.
However, the following four points must be kept in mind in programming the lengths of the delay lines.
1. Each delay line of length L requires L+1 memory sample spaces
2. Reverb delay lines require three memory words (48 bits) to implement a single delay element as the reverb
delay line operates in the 48-bit word structure of the digital audio processor (DAP).
3. Delay lines require two memory words (32 bits) to implement a single delay element as the delay lines
operate on the mixer outputs after 32-bit truncation has been applied,
4. There are five words of reserved memory space that must be preserved.
In Figure 3−17, the terms PCHx refer to delay line size assignments for the delay lines. The terms PRx see the delay
line size assignments for the reverb delay lines. For the example shown in Figure 3−17, the delay for channel 3 is
set to 0 and the reverberation delay for channel 2 is set to 0. From these zero-valued settings, it is seen that a delay
of 0 requires the use of one delay element. For the case of a delay of 0, the write transaction into the single delay
element takes place before the read from the single delay element, thereby achieving a net delay of 0.
In making the delay line length assignments, the only restriction is that the 4K memory resource not be exceeded.
Figure 3−18 illustrates the computations required to determine the maximum delay line length obtainable for five
cases:
1. One reverb delay line
2. One delay line
3. Three equal length reverb delay lines but no delay lines
4. Three equal length delay lines but no reverb delay lines
5. Three equal length delay lines and three equal length reverb delay lines.
3−26
Start
0
Delay Memory Allocation − CH1
(CH 1 Delay Assignment = PCH1)
Delay
Channel 1
Stop
Start
2(PCH1 + 1) − 1
2(PCH1 + 1)
Delay
Channel 2
Delay Memory Allocation − CH2
(CH 2 Delay Assignment = PCH2)
Stop
Start
Delay Memory Allocation − CH3
(CH 3 Delay Assignment = PCH3 = 0)
Stop
Start
Delay Memory Allocation − Reserved
(Reserved Delay Assignment = 0)
Stop
Start
2(PCH1 + PCH2 + 2) − 1
2(PCH1 + PCH2 + 2)
2(PCH1 + PCH2 + 2) + 1
2(PCH1 + PCH2 + 3)
2(PCH1 + PCH2 + 3) + 1
2(PCH1 + PCH2 + 4)
ÎÎ
ÎÎ
ÎÎ
Stop
Start
Stop
Start
2(PCH1 + PCH2 + 4) + 3(PR1 + 1) − 1
2(PCH1 + PCH2 + 4) + 3(PR1 + 1)
Reverb
Channel 2 (PR2 = 0)
2(PCH1 + PCH2 + 4) + 3(PR1 + 1) + 2
2(PCH1 + PCH2 + 4) + 3(PR1 + 2)
Reverb
Channel 3
Reverb Delay Memory Allocation − CH3
(CH 3 Reverb Delay Assignment = PR3)
Stop
Start
Reverb Delay Memory Allocation − Reserved
(Reserved Reverb Delay Assignment = 0)
Reserved
Reverb
Channel 1
Reverb Delay Memory Allocation − CH1
(CH 1 Reverb Delay Assignment = PR1)
Reverb Delay Memory Allocation − CH2
(CH 2 Reverb Delay Assignment = PR2 = 0)
Delay Channel 3 (PCH3 = 0)
Stop
2(PCH1 + PCH2 + 4) + 3(PR1 + PR3 + 3) − 1
2(PCH1 + PCH2 + 4) + 3(PR1 + PR3 + 3)
2(PCH1 + PCH2 + 4) + 3(PR1 + PR3 + 4) − 1
ÎÎ
ÎÎ
ÎÎ
ÎÎ
Reserved
Figure 3−17. Delay Line Memory Implementation
3−27
CASE 1: Maximum Length − One Reverb Delay Line
P Reverb_max_CH2 + [(4096 * 5 * 2 * 2 * 2 * 3 * 3) B 3] * 1 + 1358 2 ³ 1358
3
CH1 CH2 CH3 CH1 CH3
Delay
Reverb
Reserved
L length
requires L + 1
delay elements
CASE 2: Maximum Length − One Delay Line
P Delay_max_CH3 + [(4096 * 5 * 2 * 2 * 3 * 3 * 3) B 2] * 1 + 2038
CH1 CH2 CH1 CH2 CH3
Delay
Reverb
Reserved
L length
requires L + 1
delay elements
CASE 3: Maximum Length − Three Equal Length Reverb Delay Lines
P Reverb_max_CH1, CH2, CH3 + |{[(4096 * 5 * 2 * 2 * 2) B 3] + 1361 2 ³ 1361} B 3| + 453 2 * 1 + 452 2 ³ 452
3
3
3
CH1 CH2 CH3
Delay
Reserved
L length
requires L + 1
delay elements
CASE 4: Maximum Length − Three Equal Length Delay Lines
P Delay_max_CH1, CH2, CH3 + |{[(4096 * 5 * 3 * 3 * 3) B 2] + 2041} B 3| * 1 + 679 1 ³ 679
3
CH1 CH2 CH3
Reverb
Reserved
L length
requires L + 1
delay elements
CASE 5: Maximum Length − Three Equal Length Delay Lines and Three Equal Length Reverb Delay Lines
P DelayńReverb_max_CH1, CH2, CH3 + (4096 * 5) + 4091 + 3[2(D ) 1)] ) 3[3(R ) 1)] Where D and R + No. delay and reverb elementsńChannel
3 Channels
Reserved
R takes
No. 16-bit words
per delay element
L length
requires L + 1
delay elements
3
the memory D does, so there should be three D elements for every two R elements.
2
Therefore, D +
3
R
2
3
4091 + 3[2( R ) 1)] ) 3[3(R ) 1)] ³ R + 226 4 ³ 226
2
9
D +
3
R ³ D + 339
2
Figure 3−18. Maximum Delay Line Lengths
3−28
Commands to reconfigure the reverb delay and delay lines should not be issued as standalone commands. When
new delay assignments are issued, the content of the 4K memory resource used to implement the delay lines is not
flushed. It takes a finite time for the memory to refill with samples in correspondence with its new assignments, and
until this time has elapsed, audio samples can be output on the wrong channel. For this reason, it is recommended
that all delay line assignment commands be preceded by a mute command, and followed by an unmute command.
CAUTION: There are no error flags issued if the delay line assignments exceed the
capacity of the 4K memory resource, but undefined and erratic behavior results if the
delay line capacity is exceeded.
3.7 Dynamic Range Control (DRC)
The DRC provides both compression and expansion capabilities over three separate and definable regions of audio
signal levels. Programmable threshold levels set the boundaries of the three regions. Within each of the three regions
a distinct compression or expansion transfer function can be established and the slope of each transfer function is
determined by programmable parameters. The offset (boost or cut) at the two boundaries defining the three regions
can also be set by programmable offset coefficients. The DRC implements the composite transfer function by
computing a 5.23 format gain coefficient from each sample output from the rms estimator. This gain coefficient is then
applied to a mixer element, whose other input is the audio data stream. The mixer output is the DRC-adjusted audio
data.
There are two distinct DRC blocks in the TAS3103. One DRC services two monaural channels—CH1 and CH2. This
DRC computes rms estimates of the audio data streams on both CH1 and CH2. The two estimates are then compared
on a sample-by-sample basis, and the larger of the two is used to compute the compression/expansion gain
coefficient. The gain coefficient is then applied to both CH1 and CH2 audio. The other DRC services only monaural
channel CH3. This DRC also computes an rms estimate of the signal level on CH3, and this estimate is used to
compute the compression/expansion gain coefficient applied to CH3 audio.
Figure 3−19 shows the positioning of the DRC block in the TAS3103 processing flow. As seen, the DRC input can
come from either before or after soft volume control and loudness processing, or can be a weighted combination of
both. The mixers feeding the DRC control the selection of which audio data stream or combination thereof, is input
into the DRC. The mixers also provide a means of gaining or attenuating the signal level into the DRC. If the DRC
setup is referenced to the 0-dB level at the TAS3103 input, the coefficient values for these mixers must be taken into
account. Discussions and examples that follow further explore the role the mixers play in setting up the transfer
function of the DRC.
Base and Treble
Bypass
Soft
Volume
DRC Bypass
Loudness
From
Effects
Block
Channel
BiQuad
Filter Bank
Base
and
Treble
Σ
Σ
Σ
DRC-Derived
Gain Coefficient
Σ
DRC
Input Mixer
DRC
Input Mixer
DRC
CH 1/2 DRC Only
Figure 3−19. DRC Positioning in TAS3103 Processing Flow
3−29
Figure 3−20 illustrates a typical DRC transfer function.
DRC − Compensated Output
Region
0
Region
1
Region
2
k2
k1
1:1 Transfer Function
Implemented Transfer Fucntion
k0
O2
O1
T1
T2
DRC Input Level
Figure 3−20. Dynamic Range Compression (DRC) Transfer Function Structure
The three regions shown in Figure 3−20 are defined by three sets of programmable coefficients:
•
Thresholds T1 and T2—define region boundaries.
•
Offsets O1 and O2—define the DRC gain coefficient settings at thresholds T1 and T2 respectively.
•
Slopes k0, k1, and k2—define whether compression or expansion is to be performed within a given region.
The magnitudes of the slopes define the degree of compression or expansion to be performed.
The three sets of parameters are all defined in logarithmic space and adhere to the following rules:
•
The maximum input sample into the DRC is referenced at 0 dB. All values below this maximum value then
have negative values in logarithmic (dB) space.
•
The samples input into the DRC are 32-bit words and consist of the upper 32 bits of the 48-bit word format
used by the digital audio processor (DAP). The 48-bit DAP word is derived from the 32-bit serial data
received at the serial audio receive port by adding 8 bits of headroom above the 32-bit word and 8 bits of
computational precision below the 32-bit word. If the audio processing steps between the SAP input and
the DRC input result in no accumulative boost or cut, the DRC would operate on the 8 bits of headroom and
the 24 MSBs of the audio sample. Under these conditions, a 0-dB (maximum value) audio sample
(0x7FFFFFFF) is seen at the DRC input as a –48-dB sample (8 bits x −6.02 dB/bit = −48 dB).
•
Thresholds T1 and T2 define, in dB, the boundaries of the three regions of the DRC, as referenced to the
rms value of the data into the DRC. Zero valued threshold settings reference the maximum valued rms input
into the DRC and negative valued thresholds reference all other rms input levels. Positive valued thresholds
have no physical meaning and are not allowed. In addition, zero valued threshold settings are not allowed.
Although the DRC input is limited to 32-bit words, the DRC itself operates using the 48-bit word format of the DAP.
The 32-bit samples input into the DRC are placed in the upper 32 bits of this 48-bit word space. This means that the
threshold settings must be programmed as 48-bit (25.23 format) numbers.
CAUTION: Zero valued and positive valued threshold settings are not
allowed and cause unpredictable behavior if used.
3−30
•
Offsets O1 and O2 define, in dB, the attenuation (cut) or gain (boost) applied by the DRC-derived gain
coefficient at the threshold points T1 and T2 respectively. Positive offsets are defined as cuts, and thus boost
or gain selections are negative numbers. Offsets must be programmed as 48-bit (25.23 format) numbers.
•
Slopes k0, k1, and k2 define whether compression or expansion is to be performed within a given region,
and the degree of compression or expansion to be applied. Slopes are programmed as 28-bit (5.23 format)
numbers.
3.7.1
DRC Implementation
Figure 3−21 shows the three elements comprising the DRC: (1) an rms estimator, (2) a compression/expansion
coefficient computation engine, and (3) an attack/decay controller.
•
RMS estimator—This DRC element derives an estimate of the rms value of the audio data stream into the
DRC. For the DRC block shared by CH1 and CH2, two estimates are computed—an estimate of the CH1
audio data stream into the DRC, and an estimate of the CH2 audio data stream into the DRC. The outputs
of the two estimators are then compared, sample-by-sample, and the larger valued sample is forwarded
to the compression/expansion coefficient computation engine.
Two programmable parameters, ae and (1 – ae), set the effective time window over which the rms estimate
is made. For the DRC block shared by CH1 and CH2, the programmable parameters apply to both rms
estimators. The time window over which the rms estimation is computed can be determined by:
t
window
+
*1
F ȏn(1 * ae)
S
•
Compression/expansion coefficient computation—This DRC element converts the output of the rms
estimator to a logarithmic number, determines the region that the input resides, and then computes and
outputs the appropriate coefficient to the attack/decay element. Seven programmable parameters—T1, T2,
O1, O2, k0, k1, and k2—define the three compression/expansion regions implemented by this element.
•
Attack/decay control—This DRC element controls the transition time of changes in the coefficient computed
in the compression/expansion coefficient computation element. Four programmable parameters define the
operation of this element. Parameters ad and 1 − ad set the decay or release time constant to be used for
volume boost (expansion). Parameters aa and 1 − aa set the attack time constant to be used for volume
cuts. The transition time constants can be determined by:
ta +
*1
F ȏn(1 * aa)
S
*1
t +
d
F ȏn(1 * ad)
S
3−31
3−32
Audio Input
CH2
Audio Input
CH1 or CH3
5.23 Format
RMS
Voltage
Estimator
32
Comparator
T2-LSBits
25.23
Format
1−ae
ae
O1
{
T1
O2
T2
{
CH1/CH2 = 0xB3
CH3 = 0xB8
25.23
Format
K2
K1
K0
CH1/CH2 = 0xB5
CH3 = 0xBA
td
ta ≈ −1/[FS x ln(1−aa)]
td ≈ −1/[FS x ln(1−ad)]
Cut
Attack / Decay Control
ta
l
Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxxs
b
l
Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxxs
b
l
Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxxs
b
l
Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxxs
b
Volume
5.23 Format
m
S Slave Addr Ack Sub-Addr Ack 0000 s xxx
b
m
0000 s xxx
b
m
0000 s xxx
b
m
0000 s xxx
b
5.23 Format
Figure 3−21. DRC Block Diagram
tWindow ≈ −1/[FS x ln(1−ae)] Where FS = Audio Sample Frequency
CH1/CH2 = 0xB4
CH3 = 0xB9
m
S Slave Addr Ack Sub-Addr Ack 00000000 Ack 00000000 Ack s xxxxxxx Ack xxxxxxxx
b
l
xxxxxxxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxxs
b
m
00000000 Ack 00000000 Ack s xxxxxxx Ack xxxxxxxx
b
l
xxxxxxxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxxs
b
NOTE: Compression / Expansion / Compression Displayed
K0
K1
Compression / Expansion
Coefficient Computation
K2
ae and (1−ae) Set Time Window Over Which RMS Value is Computed
Applies to DRC Servicing CH1/CH2 Only
RMS
Voltage
Estimator
32
Ack
T2-MSBits
T1-LSBits
Ack
Ack
T1-MSBits
Ack
m
l
S Slave Addr Ack Sub-Addr Ack 0000 s xxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxxs Ack
b
b
m
l
0000 s xxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxxs Ack
b
b
m
l
0000 s xxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxxs Ack
b
b
m
S Slave Addr Ack Sub-Addr Ack 00000000 Ack 00000000 Ack s xxxxxxx Ack xxxxxxxx Ack
b
l
xxxxxxxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxxs Ack
b
CH1/CH2 = 0xB1
CH3 = 0xB6
xxxxxxxx Ack
00000000 Ack
xxxxxxxx Ack
S Slave Addr Ack Sub-Addr Ack 00000000 Ack
m
00000000 Ack s xxxxxxx Ack xxxxxxxx
b
l
xxxxxxxx Ack xxxxxxxx Ack xxxxxxxs
b
m
00000000 Ack s xxxxxxx Ack xxxxxxxx
b
l
xxxxxxxx Ack xxxxxxxx Ack xxxxxxxs
b
CH1/CH2 = 0xB2
CH3 = 0xB7
1−aa
ad
1−ad
Ack
Ack
Ack
5.23 Format
28
aa
DRC-Derived
Gain Coefficien
O2-LSBits
O2-MSBits
O1-LSBits
O1-MSBits
Ack
Ack
Ack
Ack
Ack
3.7.2
Compression/Expansion Coefficient Computation Engine Parameters
There are seven programmable parameters assigned to each DRC block: two threshold parameters - T1 and T2, two
offset parameters - O1 and O2, and three slope parameters - k0, k1, and k2. The threshold parameters establish the
three regions of the DRC transfer curve, the offsets anchor the transfer curve by establishing known gain settings
at the threshold levels, and the slope parameters define whether a given region is a compression or an expansion
region.
The audio input stream into the DRC must pass through DRC-dedicated programmable input mixers. These mixers
are provided to scale the 32-bit input into the DRC to account for the positioning of the audio data in the 48-bit DAP
word and the net gain or attenuation in signal level between the SAP input and the DRC. The selection of threshold
values must take the gain (attenuation) of these mixers into account. The DRC implementation examples that follow
illustrate the effect these mixers have on establishing the threshold settings.
T2 establishes the boundary between the high-volume region and the mid-volume region. T1 establishes the
boundary between the mid-volume region and the low-volume region. Both thresholds are set in logarithmic space,
and which region is active for any given rms estimator output sample is determined by the logarithmic value of the
sample.
Threshold T2 serves as the fulcrum or pivot point in the DRC transfer function. O2 defines the boost (> 0 dB) or cut
(< 0 dB) implemented by the DRC-derived gain coefficient for an rms input level of T2. If O2 = 0 dB, the value of the
derived gain coefficient is 1.0 (0x00, 80, 00, 00 in 5.23 format). k2 is the slope of the DRC transfer function for rms
input levels above T2 and k1 is the slope of the DRC transfer function for rms input levels below T2 (and above T1).
The labeling of T2 as the fulcrum stems from the fact that there cannot be a discontinuity in the transfer function at
T2. The user can, however, set the DRC parameters to realize a discontinuity in the transfer function at the boundary
defined by T1. If no discontinuity is desired at T1, the value for the offset term O1 must obey the following equation.
O1
No Discontinuity
+ |T1 * T2|
k1 ) O2 For ( |T1| w |T2| )
T1 and T2 are the threshold settings in dB, k1 is the slope for region 1, and O2 is the offset in dB at T2. If the user
chooses to select a value of O1 that does not obey the above equation, a discontinuity at T1 is realized.
Going down in volume from T2, the slope k1 remains in effect until the input level T1 is reached. If, at this input level,
the offset of the transfer function curve from the 1:1 transfer curve does not equal O1, there is a discontinuity at this
input level as the transfer function is snapped to the offset called for by O1. If no discontinuity is wanted, O1 and/or
k1 must be adjusted so that the value of the transfer curve at the input level T1 is offset from the 1:1 transfer curve
by the value O1. The examples that follow illustrate both continuous and discontinuous transfer curves at T1.
Going down in volume from T1, starting at the offset level O1, the slope k0 defines the compression/expansion activity
in the lower region of the DRC transfer curve.
3.7.2.1 Threshold Parameter Computation
For thresholds,
TdB = −6.0206TINPUT = −6.0206TSUB_ADDRESS_ENTRY
If, for example, it is desired to set T1 = -64 dB, then the subaddressaddress entry required to set T1 to -64 dB is:
T1
SUB_ADDRESS_ENTRY
+
*64 + 10.63
*6.0206
From Figure 3−21, it can be seen that T1 is entered as a 48-bit number in 25.23 format. Therefore:
T1 = 10.63 = 0_1010.1010_0001_0100_0111_1010_111
= 0x00000550A3D7 in 25.23 format
3−33
3.7.2.2 Offset Parameter Computation
The offsets set the boost or cut applied by the DRC-derived gain coefficient at the threshold point. An equivalent
statement is that offsets represent the departure of the actual transfer function from a 1:1 transfer at the threshold
point. Offsets are 25.23 formatted 48-bit logarithmic numbers. They are computed by the following equation.
O
INPUT
+
O
) 24.0824 dB
DESIRED
6.0206
Gains or boosts are represented as negative numbers; cuts or attenuation are represented as positive numbers. For
example, to achieve a boost of 21 dB at threshold T1, the I2C coefficient value entered for O1 must be:
O1
INPUT
+ –21 dB ) 24.0824 dB + 0.51197555
6.0206
+ 0.1000_0011_0001_1101_0100
+ 0x00000041886A in 25.23 format
More examples of offset computations are included in the following examples.
3.7.2.3 Slope Parameter Computation
In developing the equations used to determine the subaddress input value required to realize a given compression
or expansion within a given region of the DRC, the following convention is adopted.
DRC Transfer = Input Increase : Output Increase
If the DRC realizes an output increase of n dB for every dB increase in the rms value of the audio into the DRC, a
1:n expansion is being performed. If the DRC realizes a 1 dB increase in output level for every n dB increase in the
rms value of the audio into the DRC, a n:1 compression is being performed.
For 1:n expansion, the slope k can be found by:
k=n−1
For n:1 compression, the slope k can be found by: k + 1
n –1
In both expansion (1:n) and compression (n:1), n is implied to be greater than 1. Thus, for expansion:
k = n −1 means k > 0 for n > 1. Likewise, for compression, k + 1
n –1 means −1 < k < 0 for n > 1. Thus, it appears that
k must always lie in the range k > −1.
The DRC imposes no such restriction and k can be programmed to values as negative as −15.999. To determine what
results when such values of k are entered, it is first helpful to note that the compression and expansion equations
for k are actually the same equation. For example, a 1:2 expansion is also a 0.5:1 compression.
0.5 Compression å k + 1 –1 + 1
0.5
1 : 2 Expansion å k + 2–1 + 1
As can be seen, the same value for k is obtained either way. The ability to choose values of k less than −1 allows the
DRC to implement negative slope transfer curves within a given region. Negative slope transfer curves are usually
not associated with compression and expansion operations, but the definition of these operations can be expanded
to include negative slope transfer functions. For example, if k = −4
1
Compression Equation : k + *4 + 1
n *1 å n + – 3 å *0.3333 : 1 compression
Expansion Equation : k + *4 + n–1 å n + –3 å 1 : *3 expansion
With k = −4, the output decreases 3 dB for every 1 dB increase in the rms value of the audio into the DRC. As the
input increases in volume, the output decreases in volume.
3−34
3.7.3
DRC Compression/Expansion Implementation Examples
The following four examples illustrate the steps that must be taken to calculate the DRC compression/expansion
coefficients for a specified DRC transfer function. The first example is an expansion/compression/expansion
implementation without discontinuities in the transfer function and represents a typical application. This first example
also illustrates one of the three modes of DRC saturation—32-bit dynamic range limitation saturation. The second
example is a compression/expansion/compression implementation. There is no discontinuity at T1 and 32-bit
dynamic range saturation occurs at low volume levels into the DRC. Example 2 also illustrates another form of DRC
saturation—maximum gain saturation. Example 3 illustrates the concept of infinite compression. Also, in Example 3,
32-bit dynamic range saturation occurs at low volume levels and the third form of DRC saturation is
illustrated—minimum gain saturation. Example 4 illustrates the ability of the DRC to realize a negative slope transfer
function. This example also illustrates two of the three forms of saturation—32-bit dynamic range saturation at low
volume levels and minimum gain saturation.
CAUTION:
The examples presented all exhibit some form of DRC saturation. This is not intended
to imply that all (or most) DRC transfer implementation exhibit some form of saturation.
Most practical implementations do not exhibit saturation. The examples are chosen to
explain by example the three types of saturation that can be encountered. But the
phenomenon of saturation can also be used to advantage in that it effectively provides
a means to implement more than three zones or regions of operation. If saturation is
intended, the regions exhibiting the transfer characteristic set by k0, k1, and k2 provide
three regions and the regions exhibiting saturation provide the additional regions of
operation.
3.7.3.1 Example 1—Expansion/Compression/Expansion Transfer Function With 32-Bit
Dynamic Range Saturation
For this example, the following transfer characteristics are chosen.
•
•
•
•
•
Threshold point 2: T2 = −26 dB, O2 = 30 dB
Threshold point 1: T1 = −101 dB, O1 = −7.5 dB
Region 0 slope: k0 = 0.05 ≥ 1:1.05 Expansion
Region 1 slope: k1 = −0.5 ≥ 2:1 Compression
Region 2 slope: k2 = 0.1 ≥ 1:1.1 Expansion
The thresholds T1 and T2 are typically referenced, by the user, to the 0-dB signal level into the TAS3103. But to
determine the equivalent threshold point at the DRC input, it is necessary to take into account the processing gain
(or loss) between the TAS3103 SAP input and the DRC. Assume, as an example, the processing gain structure shown
in Figure 3−22. Inputting the data below the 8-bit headroom in the 48-bit DAP word and then routing only the upper
32 bits of the 48-bit word into the DRC, results in a 48-dB (8 bits x 6 dB/bit = 48 dB) attenuation of the signal level
into the DRC. Channel processing gain and use of the dedicated mixer into the DRC can revise this apparent 48-dB
attenuation in signal level into the DRC. In Figure 3−22, the 24 mixer gain into the DRC, coupled with a net channel
gain of 0 dB, changes the net 48-dB attenuation of the signal level into the DRC to a net attenuation of 24 dB.
For slopes:
Region 0 = 1:1.05 Expansion
≥ k0 = 1.05 − 1 = 0.05
= 00000.0000_1100_1100_1100_1100_110
= 0x0066666 in 5.23 format
Region 1 = 2:1 Compression
≥ k1 = 1/2 − 1 = -0.5
= 11111.1000_0000_0000_0000_0000_000
= 0xFC00000 in 5.23 format
Region 2 = 1:1.1 Expansion
≥ k2 = 1.1 − 1 = 0.1
= 00000.0001_1001_1001_1001_1001_100
= 0x00CCCCC in 5.23 format
3−35
0-dB Gain
Channel 1
Processing
48-Bit
DAP
Word
ÎÎÎÎ
ÎÎÎÎ
ÎÎÎÎ
Headroom
1
2
8
2 0
4
B
B
3
i
2 B i
t
i t
B t
i
t
SAP
Input
Port
48-Bit
DAP
Word
24
47
Headroom
40
39
ÎÎÎÎ
ÎÎÎÎ
47
44
43
1
2
8
2 0
4
B
B
3
i
2 B i
t
i t
B t
i
t
1
6
B
i
t
1
6
B
i
t
DRC
16
ÎÎÎÎ
ÎÎÎÎ
ÎÎÎÎ
Resolution
Resolution
8
7
0
ÎÎÎÎ
ÎÎÎÎ
ÎÎÎÎ
8
7
24
0
Channel 2
Processing
0-dB Gain
Figure 3−22. DRC Input Word Structure for 0-dB Channel Processing Gain
The resulting DRC transfer function for the above parameters is shown in Figure 3−23. The threshold T2 is set at the
DRC rms input level of −50 dB, which corresponds to a −26-dB rms signal level at the SAP input. The
DRC-compensated output at T2 is cut 30 dB with respect to the 1:1 transfer function (O2 = 30 dB). The threshold T1
is set at the DRC rms input level of −125 dB, which corresponds to a −101-dB rms signal level at the SAP input. The
DRC-compensated output at T1 is boosted by 7.5 dB with respect to the 1:1 transfer function (O1 = −7.5 dB).
For thresholds, TdB = -6.0206TINPUT = -6.0206TSUB_ADDRESS_ENTRY.
Therefore,
T2 = −26 dB −24 dB = −50 dB ≥ T2INPUT
= −50/−6.0206 = 8.30482
= 01000.0100_1110_0000_1000_1010_111
= 0x000004270457 in 25.23 format
T1 = −101 dB −24 dB = −125 dB ≥ T1INPUT = −125/−6.0206 = 20.76205
= 010100.1100_0011_0001_0101_1011_010
= 0x00000A618ADA in 25/23 format
For offsets, O
INPUT
Therefore, O2
INPUT
ƪ
ƫ
+
1
O ) 24.0824 .
6.0206 dB
+
1 [30 ) 24.0824] + 8.982892
6.0206
+ 01000.1111_1011_1001_1110_1100_111
+ 0x0000047DCF67 in 25.23 format
3−36
O1
INPUT
+
1 [*7.5 ) 24.0824] + 2.754277
6.0206
+ 010.1100_0001_0001_1000_0100_110
+ 0x000001608C26 in 25.23 format
For input levels above the T2 threshold, the transfer function exhibits a 1:1.1 expansion. For input levels below T2,
the transfer function exhibits a 2:1 compression. Also, by definition, it is seen that there is no discontinuity in the
transfer function at T2. When the 2:1 compression curve in region 1 intersects the T1 threshold level, the output level
is 7.5 dB above the 1:1 transfer, an offset value identical to O1. Thus, there is no discontinuity at T1. For input levels
below T1, the transfer function exhibits a 1:1.05 expansion.
DRC rms input levels below −192 dB fall below the 32-bit precision of the DRC input (32 bits x −6 dB/bit = −192 dB).
This means that for levels below −192 dB, the DRC sees a constant input level of 0, and thus the computed DRC gain
coefficient remains fixed at the value computed when the input was at −192 dB. The transfer function then has a 1:1
slope below the –192-dB input level and is offset from the 1:1 transfer curve by the offset present at the –192-dB input
level.
The change from a 1:1.05 expansion to a 1:1 transfer below −192 dB is the result of 32-bit dynamic range saturation
at the DRC input. This type of saturation always occurs at a DRC input level of −192 dB. However, the input level at
which this type of saturation occurs depends on the channel gain. For this example, the saturation occurs at an input
level of −168 dB (−192-dB DRC input + 48 dB 8-bit headroom –24-dB mixer gain into DRC).
3.7.3.2 Example 2—Compression/Expansion/Compression Transfer Function With Maximum Gain
Saturation and 32-Bit Dynamic Range Saturation
The transfer function parameters for this example are given in Table 3−5. In setting the threshold levels it is assumed
that the net processing gain between the SAP input and the DRC is 0 dB. This is the same as Example 1 except that
the gain of the mixer into the DRC is set to 1 instead of 24. Because of the 8-bit headroom in the 48-bit DAP word,
the upper eight bits of the 32-bit DRC input word are zero, resulting in 0-dB signal levels at the SAP input being seen
as –48-dB signal levels at the DRC.
Figure 3−23 shows the DRC transfer function resulting from the parameters given in Table 3−5. At threshold level
T2 (−70 dB), O2 specifies a boost of 30 dB. But the signed 5.23 formatted gain coefficient only provides a 24-dB boost
capability (5 integer bits = Sxxxx ≥ 24 × 6 dB/octave = 24 dB). Internally, the DRC operates in 48-bit space and thus
computes a 30-dB boost. But the 5.23 formatted gain coefficient saturates or clips at 24 dB. The transfer curve thus
resides 24 dB above the 1:1 transfer curve at T2.
3−37
+30
+20
+10
0
−10
−20
−30
−40
1:1 Transfer Function
−50
Implemented Transfer Function
−60
Slope change points
DRC − Compensated Output (dB)
O2 = 30 dB
−70
k2 = 1 : 1.1
−80
−90
−100
O1 = −7.5 dB
−110
k1 = 2 : 1
−120
−130
−140
−150
−160
k0 = 1 : 1.05
−170
−180
−190
−200
−210
k = 1:1
(32-Bit Dynamic Range Saturation)
−220
−220 −210 −200 −190 −180 −170 −160 −150 −140 −130 −120 −110 −100 −90
−80
−70
T1
DRC INPUT (dB)
Figure 3−23. DRC Transfer Curve—Example 1
3−38
−60
−50
T2
−40
−30
−20
−10
0
The transfer curve remains a constant 24 dB above the 1:1 transfer curve for input levels above and below T2 until
the computed DRC gain coefficient falls within the dynamic range of a 5.23 format number. For input levels above
T2, k2 implements a 5:1 compression, At an input level 7.5 dB above T2 (−62.5 dB), the DRC transfer curve has risen
7.5/5 = 1.5 dB. The boost at this point is 30 dB - (7.5 dB - 1.5 dB) = 24 dB. The DRC has come out of gain saturation.
For input levels above −62.5 dB, the transfer curve exhibits 5:1compression.
Table 3−5. DRC Example 2 Parameters
I2C COEFFICIENT VALUE
DRC
PARAMETER
REQUIRED (SPECIFIED) VALUE
(NET GAINSAP Input-DRC = 0 dB)
T2
−22 dBInput ≥ −70 dBDRC
T1
−102 dBInput ≥ −150 dBDRC
O2
−30 dB
(−30 + 24.0824)/6.0206 = −0.982892
= 0xFFFFFF82309825.23 Format
O1
50 dB
(50 + 24.0824)/6.0206 = 12.304820
= 0x00000627045825.23 Format
k2
5:1 Compression
k1
1:2 Expansion
k0
2:1 Compression
−70/−6.0206= 11.626748
= 0x000005D0394825.23 Format
−150/−6.0206 = 24.91446
= 0x00000C750D0925.23 Format
(1/5) − 1 = −0.8 = 0xF99999A5.23 Format
2 − 1 = 1 = 0x08000005.23 Format
(1/2) − 1 = −0.5 = 0xFC000005.23 Format
For input levels below T2, k1 implements a 1:2 expansion. With a 1:2 expansion in effect, the transfer curve has
dropped 12 dB at an input level 6 dB below T2. The boost at this level is 30 dB − (12 dB − 6 dB) = 24 dB. The DRC
gain coefficient has again come out of saturation. For input levels below −76 dB and above −150 dB, the transfer curve
exhibits a 1:2 expansion.
At T1 (-150 dB), the transfer curve is 50 dB below the 1:1 transfer curve. Since O1 = 50 dB, there is no discontinuity
in the transfer function. For inputs below −150 dB, k0 implements a 2:1 compression. The change from a 2:1
compression to a 1:1 transfer at −192 dB is due to 32-bit dynamic range saturation at the DRC input.
3.7.3.3 Example 3—1:1 Transfer/Infinite Compression With Minimum Gain Saturation, 32-Bit
Dynamic Range Saturation, and Equal Threshold Settings (T1=T2)
The DRC transfer function parameters for this example are given in Table 3−6. In addition to illustrating minimum gain
saturation, this example also illustrates the operation of the DRC when T1 and T2 are set equal.
3−39
k = 1:1
(Gain Saturation)
k2 = 5:1
− 40
−24 dB
− 50
O2 = − 30 dB
− 60
−24 dB
−70
1:1 Transfer Function
− 80
Ideal Transfer Function (Unlimited Resolution)
− 90
− 100
Implemented Transfer Function
k1 = 1 : 2
Slope change points
− 110
DRC − Compensated Output (dB)
− 120
− 130
− 140
− 150
− 160
− 170
O1 = 50 dB
− 180
− 190
− 200
− 210
− 220
k0 = 2 : 1
− 230
− 240
− 250
− 260
− 192 dB
− 270
− 280
− 62.5 dB
− 76 dB
k = 1:1
(32-Bit Dynamic
Range Saturation)
− 290
−290 −280 −270 −260 −250 −240 −230 −220 −210 −200 −190 −180 −170 −160 −150 −140 −130 −120 −110 −100 −90 −80 −70 −60 −50
T1
DRC INPUT (dB)
Figure 3−24. DRC Transfer Curve—Example 2
3−40
T2
When T1 and T2 are set equal, the following questions arise:
•
If O1 ≠ O2, what roles do O1 and O2 have?
•
Which slope parameter, k0 or k1, has control of the transfer function for input levels below the common
threshold point?
•
Does k2 control the transfer function for inputs above the common threshold point?
This example addresses and answers those questions.
Table 3−6. DRC Example 3 Parameters
I2C COEFFICIENT VALUE
DRC
PARAMETER
REQUIRED (SPECIFIED) VALUE
(NET GAINSAP Input-DRC = 0 dB)
T1 and T2
−148.7 dBInput ≥ −172.7 dBDRC
O2
−20 dB
(−20 + 24.0824)/6.0206 = 0.678072
= 0x00000056CB0F25.23 Format
O1
10 dB
(10 + 24.0824)/6.0206 = 5.660964
= 0x000002D49A7825.23 Format
k2
∞:1 Compression
k1
1:1 Transfer
k0
2:1 Compression
−172.7/−6.0206= 28.684849
= 0x00000E57A91F25.23 Format
(1/∞) − 1 = −1 = 0xF8000005.23 Format
(1/1) − 1 = 0 = 0x00000005.23 Format
(1/2) − 1 = −0.5 = 0xFC000005.23 Format
For this example it is assumed that a net processing gain of 24 (24 dB) is realized from the SAP input and the DRC
(which is identical to the net processing gain assumed for Example 1). The 24 gain results in reducing the 8-bit
headroom in the 48-bit DAP word to a headroom of four bits. The 32-bit data into the DRC then resides in bits 27:0,
which means that the data level into the DRC is down 24 dB with respect to the input level at the SAP. Data input into
the TAS3103 (SAP) at a level of −148.7 dB is seen as a −148.7 dB − 24 dB = −172.7-dB signal at the DRC. T1 and
T2 must be set to −172.7 dB to realize a common threshold point at an incoming signal level of −148.7 dB.
Figure 3−25 shows the transfer function resulting from entering the I2C coefficient values given in Table 3−6. At the
T1/T2 threshold, a discontinuity of 30 dB is observed. For inputs above the threshold, the transfer curve is horizontal
(infinite compression), and the horizontal line starts 20 dB above the 1:1 transfer curve at the threshold point. Thus,
for cases when T1 = T2, O2 governs the offset with regard to the starting point of the transfer curve above the common
threshold point and k2 determines the slope of the transfer curve. For inputs below the common threshold point, the
transfer curve exhibits a 2:1 compression and starts 10 dB below the 1:1 transfer curve. Thus, O1 sets the offset at
the threshold point for the transfer curve at and below the common threshold point and k0 determines the slope of
this curve. Slope parameter k1 plays no role when T1 = T2. The value of 0 (1:1 transfer) used in this example for k1
could be changed to any value and the resulting transfer function would not be altered. The change from a 2:1
compression to a 1:1 transfer at −192 dB is due to 32-bit dynamic range saturation at the DRC input.
3−41
0
−10
−20
−30
−40
DRC − Compensated Output (dB)
1:1 Transfer Function
−50
Implemented Transfer Function
−60
Slope change points
−70
−80
−90
−100
−110
−120
k = 1:1
(Minimum Gain Saturation)
−130
−140
−150
O2 = −20 dB
k2 = ∞:1
−160
30 dB
−170
−180
−190
−200
k = 1:1
(32-Bit Dynamic
Range Saturation)
O1 = 10 dB
k0 = 2:1
−210
−14.7 dB
−192 dB
−220
−220 −210 −200 −190 −180 −170 −160 −150 −140 −130 −120 −110 −100 −90
−80
−70
T1/T2
DRC INPUT (dB)
Figure 3−25. DRC Transfer Curve—Example 3
3−42
−60
−50
−40
−30
−20
−10
0
The horizontal slope of the transfer curve above the common threshold point does not remain horizontal indefinitely.
At a point 158 dB above the common threshold point (−14.7-dB DRC input level), the transfer function has gone from
a boost of 20 dB to a cut of 138 dB. A cut of 138 dB is the maximum cut possible for a 5.23 format gain coefficient
(2−23 ≥ 23 octaves × 6 dB/octave = 138 dB). Thus, at a DRC input level of −14.7 dB, minimum gain saturation has
been reached. For inputs above this saturation point, the DRC-derived gain coefficient remains constant at the
minimum gain value (2−23), and the transfer function exhibits a 1:1 transfer slope.
3.7.3.4 Example 4—Expansion/Cut/Expansion With Gain Saturation and 32-Bit Dynamic Range
Saturation
The three previous examples restricted the slope factor k to lie in the range k ≥ −1. This example illustrates the transfer
characteristic obtained using a value of k less than −1. For this example it is assumed that the net processing gain
into the DRC is 0 dB. This means that the 8-bit headroom in the 48-bit DAP processing word structure does not contain
data. Since the DRC receives the upper 32 bits of this 48-bit word, data at the DRC is down 48 dB (8 bits x 6 dB/ bit
= 48 dB) with respect to the signal level at the SAP input (TAS3103 input). The transfer function parameters for this
example are given in Table 3−7. Figure 3−26 shows the transfer function resulting from entering the I2C coefficient
values given in Table 3−7.
At the threshold point T2 (−70 dB), the transfer function is 100 dB below the 1:1 transfer slope (O2 = 100 dB). For
input levels above T2, the transfer function exhibits a 1:1.4 expansion. For input levels below T2, the transfer function
exhibits a negative slope; for every dB the input decreases, the output increases by 1 dB. At an input level 62 dB below
T2 (−132 dB), the transfer curve has risen 62 dB, for a net boost of 124 dB. The transfer curve at this input level is
24 dB above the 1:1 transfer curve. This boost value puts the DRC-derived gain coefficient into gain saturation. For
input levels below −132 dB, the gain coefficient remains constant at maximum gain and the transfer function exhibits
a 1:1 transfer slope, parallel to the 1:1 transfer curve but 24 dB above it.
At T1 (-150 dB), the transfer curve snaps back to the 1:1 transfer curve since O1 = 0 dB. The DRC gain coefficient
is no longer in gain saturation and for inputs below −150 dB, the transfer function exhibits a 1:1.5 expansion. The
change from a 1:1.5 expansion to a 1:1 transfer below −192 dB is the result of 32-bit dynamic range saturation.
3−43
+30
+20
+10
0
−10
−20
−30
−40
DRC − Compensated Output (dB)
1:1 Transfer Function
−50
Implemented Transfer Function
−60
Slope change points
−70
−80
−90
k = 1:1
Gain Saturation)
−100
−110
O2 = 100 dB
−120
k1 = 1:−1
−130
O1 = 0 dB
k2 = 1:1.4
−140
−150
−160
−170
−180
k = 1:1
(32-Bit Dynamic
Range Saturation)
−190
−200
k0 = 1:1.5
−210
−220
−132 dB
−192 dB
−220 −210 −200 −190 −180 −170 −160 −150 −140 −130 −120 −110 −100 −90
−80
−70
T2
T1
DRC INPUT (dB)
Figure 3−26. DRC Transfer Curve—Example 4
3−44
−60
−50
−40
−30
−20
−10
0
Table 3−7. DRC Example 4 Parameters
DRC
PARAMETER
REQUIRED (SPECIFIED) VALUE
(NET GAINSAP Input-DRC = 0 dB)
T2
−22 dBInput ≥ −70 dBDRC
T1
−102 dBInput ≥ −150 dBDRC
O2
100 dB
O1
0 dB
k2
1:1.4 Expansion
k1
1:−1 Transfer
k0
1:1.5 Expansion
I2C COEFFICIENT VALUE
−70/−6.0206= 11.626748
= 0x000005D0394825.23 Format
−150/−6.0206 = 24.91446
= 0x00000C750D0925.23 Format
(100 + 24.0824)/6.0206 = 20.609640
= 0x00000A4E08B025.23 Format
(0 + 24.0824)/6.0206
= 4.000000
= 0x00000200000025.23 Format
1.4 − 1 = 0.4 = 0X03333335.23 Format
(1/−1) − 1 = −1 −1 = −2 = 0XF0000005.23 Format
1.5 − 1 = 0.5 = 0x04000005.23 Format
3.8 Spectrum Analyzer/VU Meter
The TAS3103 contains an I2C bus programmable function block that can serve as either a spectrum analyzer or a
volume unit (VU) meter. Figure 3−26 shows the structure of this function block and lists the I2C subaddress of the
parameters that control it.
The block consists of 10 biquad filters, each followed by an rms estimator and a logarithmic converter. Two nodes
provide input to the block, with each node servicing five of the 10 biquad filters. Audio from input node s can either
come exclusively from channel 1, channel 2, channel 3, or from a gain-weighted combination of these channels. Audio
from input node t can also come exclusively from either channel 1, channel 2, or channel 3, or from a gain-weighted
combination of these channels. The spectrum analyzer then can be used to divide the audio frequency band into ten
frequency bins to examine the spectrum of the audio data stream on channel 1, channel 2, channel 3, or any
combination of these channels. The spectrum analyzer can also be used to divide the audio frequency band into five
frequency bins to examine the spectral content of two of the channels independently.
The VU meter is a special case of the spectrum analyzer that uses only the outputs from biquad 5 and biquad 6.
Typically, for the VU meter, one channel would be routed to biquad 5 (node s) and a different channel would be routed
to biquad 6 (node t). Each biquad filter would be assigned a band pass transfer function that encompasses most of
the audio band, or the filter could be configured as a pass-through device to see the full spectral band. The two outputs
then would be a measure of the energy on the two channels. Other options for the VU meter are also available. For
example, by properly setting the coefficients on biquad 5 and biquad 6, the concurrent measurement of bass and
treble volume levels on a single channel could be made.
Mixer and summation elements preceding the two input nodes s and t provide a means of adjusting the spectrum
analyzer and VU meter outputs relative to the incoming audio data stream. The spectrum analyzer and VU meter
outputs are unsigned 5.3 format base 2 logarithmic numbers. The integer part of the number designates the most
significant bit (in the 48-bit digital audio processor − DAP − word) occupied by the magnitude of the rms estimate of
the biquad filter output. A value of 31 means the magnitude of the rms estimate occupies bit 47 of the 48-bit DAP word
(bit 48 is the sign bit, and using the absolute value of the biquad filter output in determining the rms estimate makes
this bit always 0 in value). A value of 30 means the magnitude of the rms estimate occupies bit 46 and this pattern
continues with a value of 1 signifying the magnitude of the rms estimate occupies bit 17. A value of 0 signifies that
the magnitude of the rms estimate is below bit 17. The fractional digits in the 5.3 formatted number are simply the
three bits below the most significant data bit. If the rms estimate lies below bit 16 of the 48-bit DAP word, the spectrum
analyzer/VU meter output is 0.0. Figure 3−27 gives examples of logarithmic outputs for different 48-bit rms estimate
values.
3−45
3−46
Channel 3
Channel 2
Channel 1
GMix: q to r
GMix: p to r
GMix: o to r
Node r
Node q
Node p
GMix: p to t
RMS Voltage
Estimator
BiQuad 2
Ack b2
Ack b1
Ack b0
Ack a2
Ack a1
VU Meter
Output 1
(BiQuad 5)
VU Meter
Output 1
(BiQuad 6)
asa and (1−asa) Set Time Window Over Which RMS Value Is Computed
tWindow ≈ −1/[FS x ln(1−asa)] Where FS = Audio Sample Frequency
Figure 3−27. Spectrum Analyzer/VU Meter Block Diagram
Sub-Addresses 0xBC (BiQuad 1) − 0xC5 (BiQuad 10)
BiQuad Filter Coefficients Sub-Address Format
Ack xxxxxxxx Ack xxxxxxxx Ack
Ack xxxxxxxx Ack xxxxxxxx Ack
Ack xxxxxxxx Ack xxxxxxxx Ack
Ack xxxxxxxx Ack xxxxxxxx Ack
Ack xxxxxxxx Ack xxxxxxxx Ack
Sub-Address 0xBB
RMS Voltage Estimator Coefficients
Sub-Address Format
xxxxx.xxx Ack
S Slave Addr Ack Sub-Addr Ack xxxxx.xxx Ack
Sub-Address 0xFE
VU Meter Outputs
Sub-Address Format
I2C Bus
xxxxx.xxx Ack BiQuad 10
xxxxx.xxx Ack BiQuad 9
xxxxx.xxx Ack BiQuad 8
xxxxx.xxx Ack BiQuad 7
xxxxx.xxx Ack BiQuad 6
xxxxx.xxx Ack BiQuad 5
xxxxx.xxx Ack BiQuad 4
xxxxx.xxx Ack BiQuad 3
xxxxx.xxx Ack BiQuad 2
S Slave Addr Ack Sub-Addr Ack xxxxx.xxx Ack BiQuad 1
Spectrum Analyzer Outputs Sub-Address Format
Sub-Address 0xFD
m
l
S Slave Addr Ack Sub-Addr Ack 0000 s xxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxs Ack asa
b
b
m
l
0000 s xxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxs Ack 1−asa
b
b
Log
Log
RMS Voltage
Estimator
BiQuad 9
RMS Voltage
Estimator
Log
RMS Voltage
Estimator
BiQuad 8
BiQuad 10
Log
RMS Voltage
Estimator
BiQuad 7
Node
t
Log
Log
BiQuad 6
RMS Voltage
Estimator
RMS Voltage
Estimator
Log
RMS Voltage
Estimator
BiQuad 4
BiQuad 5
Log
BiQuad 3
Log
Log
RMS Voltage
Estimator
l
xxxxxxs
b
l
xxxxxxs
b
l
xxxxxxs
b
l
xxxxxxs
b
l
xxxxxxs
b
RMS Voltage
Estimator
BiQuad 1
Spectrum Analyzer / VU Meter
t
Σ
Σ
Node
s
m
S Slave Addr Ack Sub-Addr Ack 0000 s xxx
b
m
0000 s xxx
b
m
0000 s xxx
b
m
0000 s xxx
b
m
0000 s xxx
b
Σ
GMix: r to s & t
GMix: o to s
Node o
m
l
S Slave Addr Ack Sub-Addr Ack 0000 s xxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxs Ack
b
b
Mixer Gain Coefficient
Sub-Address Format
GMix: o to r
→Sub-Address 0x84
GMix: o to s
→Sub-Address 0x85
GMix: p to r
→Sub-Address 0x86
GMix: p to t
→Sub-Address 0x87
GMix: q to r
→Sub-Address 0x88
GMix: r to s & t →Sub-Address 0x89
Sub-Address Decoder
Spectrum Analyzer
and
VU Meter Output
48-Bit RMS Estimate
31
0
48
0 1 1 0 0 1 0
11111.100
28
0
48
0 0 0 0 1 1 0 1 0 1
11100.101
1
48
0 0 0
18
0 1 0 1 0 1 1 0
0
00001.010
0
48
0 0 0
18
0 0 1 1 1 0 1 0
0
48
0 0 0
18
0 0 0 1 1 0 1 0 1
0
00000.110
00000.000
Figure 3−28. Logarithmic Number Conversions—Spectrum Analyzer/VU Meter
The time window over which the rms estimate is conducted is programmable via the I2C bus (subaddress 0xBB). The
time window for a given set of coefficients is approximately:
t
Window
[*
1
F ȏn(1 * asa)
S
Where FS is the audio sample rate and asa is a 5.23 format number. The variable asa (and 1 – asa) must be kept
within the range of greater than zero and less than one. The time constant programmed applies to all 10 rms estimate
blocks.
CAUTION: The spectrum analyzer and VU meter functions are only accessible in the I2C
slave mode.
3.9 Dither
The TAS3103 provides a dither block for adding triangular or quadratic (sum of two uncorrelated triangular
distributions) distributed noise to the processed audio data stream prior to routing to the output serial audio port
(SAP). Each of the three monaural channels in the TAS3103 has its own dedicated dither data stream that is
statistically independent from the dither data streams used by the other two monaural channels. The statistical
distribution of the dither data stream, triangular or quadratic, is selectable, but the selection made applies to the dither
data streams for all three monaural channels. Each monaural channel is also assigned a mixer for adjusting the level
at which the dither data stream is inserted into the audio data stream.
Figure 3−29 is a detailed block diagram of the dither block. Five subaddresses are used to fully configure the dither
blocks and the associated channel mixers. In the I2C master mode, these dither parameters are set by the EEPROM
content and cannot be subsequently changed.
3−47
3.9.1
Dither Seeds
The dither circuit consists of two linear feedback shift registers—LFSR1 and LFSR2. The dither seed subaddress
(0xC7) consists of a byte-wide seed for LFSR1 (bits 7:0) and a byte-wide seed for LFSR2 (bits 15:8). The seeds serve
to define the starting point of each LFSR sequence, but not the feedback structure itself. Each linear feedback shift
register (LFSR) is a 26-bit structure that runs off the digital audio processor (DAP) clock. For a maximum DAP clock
frequency of 135.168 MHz [12.288 MHz (MCLKI) x 11 (PLL multiplier)], the 26-bit LFSR has a cycle time of 496.5 ms
(226/135.168 MHz). LFSR1 and LFSR2 use the same feedback structure but different taps for outputting. As long as
the seeds for the two LFSRs are not ±13 counts apart (in which case the two different sets of output taps would
correlate), any set of seed values for LFSR1 and LFSR2 is suitable.
When two or more TAS3103s are powered by the same supply, a concern as to whether or not there is correlation
between the dither data streams on different chips arises. At power turn on, a TAS3103 does not begin the dither
process until reset is deactivated. If the TAS3103s are reset by the internal power good signals from the internal
regulators, the chip-to-chip variance of the logic voltage threshold point at which the power good signal is deactivated
assures, with a high probability, that the dither data streams between chips are uncorrelated. However, when an
external logic-driven reset is applied to the TAS3103s, the probability of correlation between the dither data streams
on different chips after the rest is removed significantly increases. For this reason, the dither seeds have been made
programmable via the I2C bus.
3−48
0x7F
S
Slave Addr
m
Ack Sub-Addr Ack 0000 s xxx Ack
b
Slave Addr
m
Ack Sub-Addr Ack 0000 s xxx Ack
b
Slave Addr
m
Ack Sub-Addr Ack 0000 s xxx Ack
b
xxxxxxxx
Ack
xxxxxxxx
l
Ack xxxxxxx s Ack
b
xxxxxxxx
Ack
xxxxxxxx
l
Ack xxxxxxx s Ack
b
xxxxxxxx
Ack
xxxxxxxx
l
Ack xxxxxxx s Ack
b
xxxxxxxx
Ack
xxxxxxxx
xxxxxxxx
Ack
xxxxxxxx
0x80
S
0x81
S
Dither Block
0xC6
S
Slave Addr
m
Ack Sub-Addr Ack 0000 s xxx Ack
b
m
0000 s xxx Ack
b
l
Ack xxxxxxx s Ack
b
l
Ack xxxxxxx s Ack
b
Distribution 1 Mix
Distribution 2 Mix
LFSR1 Mix and LFSR2 Mix Are 5.23 Format Coefficients
Linear Feedback Shift Register Block
p
0.5
0.25
− W 0 +W
Output
Sampler
Dither 1
St 1
Σ
St2
L
O
G
I
C
St3
Dither 2
Σ
St4
St5
Dither 3
LFSR1
LFSR2
St6
Σ
Seed Build Logic
Condensed
LFSR2
Seed
Condensed
LFSR1
Seed
0xC7
S
Slave Addr
Ack Sub-Addr Ack
00000000
Ack
00000000
m
l
l
m
Ack s xxxxxx s Ack s xxxxxx s Ack
b
b
b
b
Dither Seed
NOTE: W = 16.0 => 0x000008000000 in 25.23 Format
Figure 3−29. Dither Data Block Diagram
3−49
When updating multiple TAS3103s with dither seeds, timing should be taken into account. The recommended seed
update process is to load all TAS3103s with their seed values in less time then the minimum LFSR cycle time of
496.5 ms, and use the same set of seeds for all TAS3103s. Each TAS3103 immediately begins running, starting at
the state set by the new seed, upon receiving the new seed. The sequential delivery, in time, of the seeds to the
multiple TAS3103s assures the TAS3103s do not all start with their new seeds at the same time, and the completion
of the process in less than 496.5 ms assures that previously programmed TAS3103s are not repeating the cycle at
the same time another TAS3103 is being programmed with the same seed set—causing correlation between the
dither data streams of the two TAS3103s.
CAUTION: The state of the digital audio processor may prevent the loading of a new I2C
—commanded dither seed value. Anytime a new seed is loaded into the TAS3103 via an
I2C write transaction to address 0xC7, it must be followed by an I2C read transaction to
address 0xC7 to verify that the new seed value was accepted. If the new seed was not
accepted, the write-read sequence must be repeated.
3.9.2
Dither Mix Options
In Figure 3−29 it is seen that the two LFSRs are logically combined to produce a triangular probability distribution.
This distribution is then sampled at six different points in time, by the DAP processing clock, to create six statistically
independent data streams. Sampling the LSFR outputs at different points in time within an audio sample period
(1/LRCLK) assures the six dither data streams are uncorrelated. The six uncorrelated dither data streams are then
routed through mixers. Each pair of mixer outputs is then applied to a summation block. It is noted in Figure 3−29
that each of the three mixers pairs has the same set of coefficients (set by I2C subaddress 0xC6). If the coefficients
for both mixers are set to 1.0, a quadratic distribution is obtained. If either coefficient is set to 0.0, a triangular
distribution is obtained. If both coefficients are set to 0.0, dither is disabled.
3.9.3
Dither Gain Mixers
Figure 3−29 shows the peak magnitude of the triangular distribution to be ±16 in the 48-bit DAP word (25.23 format).
The peak magnitude of the quadratic dither data is twice this, or ±32. Figure 3−30 shows the position of this peak
magnitude value in reference to the DAP 48-bit data word. Table 3−8 lists the mixer gains required to position the
dither data stream at the LSB of the output data word for different data sample word sizes.
Table 3−8. Mixer Gain Setting for LSB Dither Data Insertion
MIXER GAIN COEFFICIENT
DISTRIBUTION
3−50
Triangular
32-BIT SAMPLE
2−19
24-BIT SAMPLE
2−11
20-BIT SAMPLE
2−7
18-BIT SAMPLE
2−5
16-BIT SAMPLE
2−3
Quadratic
2−20
2−12
2−8
2−6
2−4
Quadratic Distribution
48-Bit DAP
Data Word
47
0.375
ρ
8-Bit
Headroom
0.25
40
39
32
Bit
Sample
0.0625
−2W −W 0
W 2W
24
Bit
Sample
20
Bit
Sample
18
Bit
Sample
16
Bit
Sample
16-Bit
Output SAP
Word Size
Output
28
27
Triangular Distribution
32-Bit
Output SAP
Word Size
24
22
20
16
ρ
0.5
0.25
8
−W
0
W
Output
0
8-Bit
Resolution
Band
Figure 3−30. Dither Data Magnitude (Gain = 1.0)
3.9.4
Dither Statistics
Figure 3−31 presents plots of the autocorrelation and channel-to-channel correlation properties of the dither data
stream when configured as triangular distributed noise. Figure 3−31(a) is the circular autocorrelation of 16K samples
of dither data collected from the TAS3103. The audio signal level was set to zero, and the dither data stream was
inserted at the LSB of the output word. The autocorrelation contains a single line of value 8000, at the point of
correlation, and random noise terms of approximately ±150 counts in value. The value 8000 agrees with the selection
of the triangular distribution—50% of the 16K dither output samples are of value ±1. Figure 3−31(b) is the circular
correlation of 16K samples of dither data from CH1 and 16K samples of dither data from CH2. There are no points
of correlation in this plot, verifying that the two data streams are uncorrelated. The random noise terms are again
approximately ±150 counts in value
3−51
(a) Auto-Correlation Plot − CH1
(b) Correlation Plot − CH2 and CH3
Figure 3−31. Triangular Dither Statistics
Figure 3−32 presents plots of the autocorrelation and channel-to-channel correlation properties of the dither data
stream when configured as quadratic distributed noise. Figure 3−32(a) is the circular autocorrelation of 16K samples
of dither data collected from the TAS3103. The audio signal level was set to zero and the dither data stream was
inserted at the LSB+1 level of the output word. The autocorrelation contains a single line of value 16,000 at the point
of correlation and random noise terms of approximately ±300 counts in value. The value 16,000 agrees with the
3−52
selection of the quadratic distribution—50% of the 16K dither output samples are of value ±1 (0.5 x 12 x 16,000 = 8000)
and 12.5% of the 16K dither output samples are ±2 (0.125 x 22 x 16,000= 8000). Figure 3−32(b) is the circular
correlation of 16K samples of dither data from CH2 and 16K samples of dither data from CH3. There are no points
of correlation is this plot, verifying that the two data streams are uncorrelated. The random noise terms are
approximately ±300 counts in value, as the dither data pattern was inserted at the LSB+1 bit of the output word instead
of the LSB bit, as was the case for triangular dither.
(a) Auto-Correlation Plot − CH1
(b) Correlation Plot − CH2 and CH3
Figure 3−32. Triangular Dither Statistics
3−53
3.10 Output Crossbar Mixers
The TAS3103 has three serial output ports—SDOUT1, SDOUT2 and SDOUT3. Each serial output port is assigned
two processing nodes within the TAS3103. One of the two nodes sources the left stereo data sample and the other
node sources the right stereo data sample. Figure 3−33 shows the assignment of these internal nodes to the serial
output ports. Two cases are shown in Figure 3−33—discrete mode and TDM mode. The discrete mode connections
are straightforward, but the TDM connections are considerably more involved in order to support the different
one-chip and two-chip TDM modes (see the Input and Output Serial Port (SAP) section for more discussion on the
TDM modes).
The purpose of the output crossbar is to give each of the three monaural channels in the TAS3103 access to any of
the six internal processing nodes (U, V, W, X, Y, and Z) that supply data to the three serial output ports. This flexibility
in the routing of the monaural channel outputs to the serial output ports, coupled with the flexibility in the routing of
the serial input ports to the monaural channels, fully decouples the input data from the output data. A given process
flow and output data topology can be obtained from any ordering of data into the TAS3103.
Figure 3−34 shows the output crossbar mixer topology. Each monaural channel feeds six mixers. The six mixers, in
turn, feed the six output nodes U, V, W, X, Y, and Z. A given monaural channel can thus be connected to either the
left or right side of SDOUT1, SDOUT2 and SDOUT3.
The mixers, although capable of performing boost (gain) and cut (attenuation) on the outgoing audio, are typically
used to facilitate on / off switching (a 5.23 format coefficient value of 0x0800000 turns the mixer on and a coefficient
value of 0x0000000 turns the mixer off). The audio data streams at the input to these mixers include dither, and any
boost or cut in the audio at this point affects the dither levels as well.
Node r in Figure 3−34 provides a means of outputting a post-processed sum of the audio on channel 1 and channel
2. This capability could be used to generate a center audio component from L and R components being processed
on channels 1 and 2. This would allow channel 3 to be a sub-woofer channel. Node r could also be used to create
a subwoofer channel, assuming an active subwoofer with filtering capability is receiving the subwoofer output. This
option would then free channel 3 for center channel processing.
3−54
SDOUT1
Internal
Processing
Nodes
L
R
LRCLK
Time
U
L
V
R
Internal
Processing
Nodes
SDOUT1
U
Node U
SDOUT2
Internal
Processing
Nodes
L
R
Nodes U & V
V
Nodes V & W
LRCLK
Time
W
L
X
R
Nodes V, W & X
W
Nodes V, W, X & Y
Nodes V, W, Y & Z
X
Nodes W, X, Y & Z
Node X
SDOUT3
Internal
Processing
Nodes
L
R
LRCLK
Time
Y
L
Z
R
Y
(a) Discrete Mode − For I2S Format, Polarity of
LRCLK Opposite That Shown
Z
(b) TDM Mode
Figure 3−33. Processing Node to Serial Output Port Topology
3−55
Mixer Gain Coefficient
Sub-Address Format
m
l
S Slave Addr Ack Sub-Addr Ack 0000 s xxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxs Ack
b
b
z to U =>
0x8F
z to X =>
0x8C
z to V =>
0x8E
z to Y =>
0x8B
z to W =>
0x8D
z to Z =>
0x8A
U
o
z
Delay
Monaural CH 1
Dither
Mixer Gain Coefficient
Sub-Address Format
0x84
m
l
S Slave Addr Ack Sub-Addr Ack 0000 s xxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxs Ack
b
b
y to U =>
0x95
y to X =>
0x92
y to V =>
0x94
y to Y =>
0x91
y to W =>
0x93
y to Z =>
0x90
V
0x86
p
Monaural CH 2
Delay
y
W
Dither
Mixer Gain Coefficient
Sub-Address Format
m
l
S Slave Addr Ack Sub-Addr Ack 0000 s xxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxs Ack
b
b
x to U =>
0x9B
x to X =>
0x98
x to V =>
0x9A
x to Y =>
0x97
x to W =>
0x99
x to Z =>
0x96
0x82
X
0x83
q
Delay
Monaural CH 3
x
Dither
Y
Mixer Gain Coefficient
Sub-Address Format
m
l
S Slave Addr Ack Sub-Addr Ack 0000 s xxx Ack xxxxxxxx Ack xxxxxxxx Ack xxxxxxs Ack
b
b
r to U =>
0xA1
r to X =>
0x9E
r to V =>
0xA0
r to Y =>
0x9D
r to W =>
0x9F
r to Z =>
0x9C
Z
r
Figure 3−34. Output Crossbar Mixer Topology
3−56
4 Electrical Specifications
4.1 Absolute Maximum Ratings Over Operating Temperature Ranges (unless
otherwise noted)†
Supply voltage range:
VDDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.5 to 3.8 V
A_VDDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.5 to 3.8 V
Input voltage range, VI: 3.3-V LVCMOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.5 V to VDDS + 0.5 V
1.8-V LVCMOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.5 V to AVDD(1) + 0.5 V
Output voltage range, VO: 3.3-V LVCMOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.5 V to VDDS + 0.5 V
1.8-V LVCMOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.5 V to DVDD(2) + 0.5 V
1.8-V LVCMOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.5 V to AVDD(3) + 0.5 V
Input clamp current, IIK (VI < 0 or VI > DVDD) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ±20 mA
Output clamp current, IOK (VO < 0 or VO > DVDD) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ±20 mA
Operating free-air temperature (Commercial) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0°C to 70°C
Operating free-air temperature (Industrial) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −40°C to 85°C
Storage temperature range, Tstg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −65°C to 150°C
Lead temperature 1,6 mm (1/16 inch) from case for 10 seconds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260°C
† Stresses beyond those listed under “absolute maximum ratings” may cause permanent damage to the device. These are stress ratings only, and
functional operation of the device at these or any other conditions beyond those indicated under “recommended operating conditions” is not
implied. Exposure to absolute-maximum-rated conditions for extended periods may affect device reliability.
NOTES: 1. AVDD is a 1.8-V supply derived from a regulator in the TAS3103 chip. Pin XTALI is the only TAS3103 input that is referenced to this
1.8 V logic supply. The absolute maximum rating listed is for reference; only a crystal should be connected to XTALI.
2. DVDD is a 1.8-V supply derived from regulators internal to the TAS3103 chip. DVDD is routed to pin 29 (DVDD_BYPASS_CAP)
to provide access to external filter capacitors, but should not be used to source power to external devices.
3. Pin XTALO is the only TAS3103 output that is derived from the internal 1.8-V logic supply AVDD. The absolute maximum rating listed
is for reference; only a crystal should be connected to XTALO. AVDD is also routed to pin 6 (AVDD_BYPASS_CAP) to provide access
to external filter capacitors, but should not be used to source power to external devices.
DISSIPATION RATING TABLE (High-k Board, 105°C Junction)
PACKAGE
TA ≤ 25°C
POWER RATING
DBT
1.094 W
13.68 mW/°C
0.478 mW
IDBT
1.094 W
13.68 mW/°C
0.478 mW
OPERATING FACTOR
ABOVE TA = 25°C
TA = 85°C
POWER RATING
4.2 Recommended Operating Conditions
Digital supply voltage, VDDS
Analog supply voltage, A_VDDS
High-level input voltage, VIH
Low-level input voltage, VIL
Input voltage, VI
Output voltage, VO
MAX
3.3
3.6
V
3
3.3
3.6
V
0.7 VDDS
VDDS
1.8-V LVCMOS (XTALI)
0.7 AVDD
AVDD
3.3-V LVCMOS
0
0.3 VDDS
1.8-V LVCMOS (XTALI)
0
0.3 AVDD
3.3-V LVCMOS
0
VDDS
1.8-V LVCMOS (XTALI)
0
AVDD
3.3-V LVCMOS
0.8 VDDS
VDDS
1.8-V LVCMOS (XTALO)
0.8 AVDD
AVDD
Industrial
Commercial
Operating junction temperature range, TJ
NOM
3
3.3-V LVCMOS
Commercial
Operating ambient air temperature range, TA
MIN
Industrial
UNITS
V
V
V
V
0
70
°C
−40
85
°C
0
105
°C
−40
105
°C
4−1
4.3 Electrical Characteristics Over Recommended Operating Conditions (Unless
Otherwise Noted)
PARAMETER
VOH
High-level output
voltage
VOL
Low-level output voltage
3.3-V LVCMOS
TEST CONDITIONS
1.8-V LVCMOS (XTALO)
IOH = −4 mA
IOH = −0.55 mA
3.3-V LVCMOS
IOL = 4 mA
1.8-V LVCMOS (XTALO)
IOL = 0.75 mA
IOZ
High-impedance output
current
Low-level input
current(4)
3.3-V LVCMOS
IIL
High-level input
current(5)
3.3-V LVCMOS
IIH
1.8-V LVCMOS (XTALI)
1.8-V LVCMOS (XTALI)
0.22
VDDS
VI = VIH
DSP clock = 135 MHz,
±20
LRCLK = 24 kHz
Power down enabled
44
LRCLK = 48 kHz
DSP clock = 33.75 MHz,
LRCLK = 24 kHz
µA
A
µA
A
mA
25
2.2
DSP clock = 67.5 MHz,
µA
75
No LRCLK, SCLK.
MCLKI/XTALI =
12.288 MHz
LRCLK = 96 kHz
Analog supply current
±20
3.5
DSP clock = 135 MHz,
MCLKI / XTALI = 12.228 MHz
±20
LRCLK = 48 kHz,
MCLKI/XTALI = 12.288
MHz
No LRCLK, SCLK, or
MCLKI/XTALI
V
0.22 AVDD
±20
LRCLK = 48 kHz
UNITS
V
0.8 AVDD
VI = VIL
VI = VIL
VI = VIH
DSP clock = 33.75 MHz,
Digital supply current
MAX
±20
DSP clock = 67.5 MHz,
MCLKI / XTALI = 12.228 MHz
IA_DVDD
TYP
3.3-V LVCMOS
LRCLK = 96 kHz
IDVDD
MIN
0.8 VDDS
mA
2
2.9
2.7
mA
2.4
LRCLK = 48 kHz,
Power down enabled
MCLKI/XTALI = 12.288
MHz
No LRCLK, SCLK, or
MCLKI/XTALI
1.5
mA
1.5
NOTES: 4. Value given is for those input pins that connect to an internal pullup resistor as well as an input buffer. For inputs that have a pulldown
resistor or no resistor, IIL is ±1 µA.
5. Value given is for those input pins that connect to an internal pulldown resistor as well as an input buffer. For inputs that have a pullup
resistor or no resistor, IIH is ±1 µA.
4−2
4.4 TAS3100 Timing Characteristics
4.4.1
Master Clock Signals Over Recommended Operating Conditions (Unless Otherwise
Noted)
PARAMETER
TEST CONDITIONS
MIN
TYP
MAX
UNITS
f(XTALI)
Frequency, XTALI (1/tc(1))
2.8
20
MHz
f(MCLKI)
tw(MCLKI)
Frequency, MCLKI (1/tc(2))
2.8
25
MHz
Pulse duration, MCLKI high
See Note 6
HMCLKI − 25
HMCLKI
HMCLKI + 25
MCLKI jitter
2.8
ns
±5
ns
25
MHz
f(MCLKO)
tr(MCLKO)
Frequency, MCLKO (1/tc(3))
Rise time, MCLKO
CL = 30 pF
9.5
ns
tf(MCLKO)
tw(MCLKO)
Fall time, MCLKO
CL = 30 pF
9.5
ns
Pulse duration, MCLKO high
See Note 9
HMCLKO
ns
80
ps
XTALI master clock source
MCLKI jitter
td(MI−MO)
MTALI master clock source
See Note 10
Delay time, MCLKI
rising edge to
MCLKO rising edge
MCKLO = MCLKI
See Note 7
17
ns
MCLKO < MCLKI
See Note 7 and Note 8
17
ns
NOTES: 6.
7.
8.
9.
HMCLKI = 1 / 2MCLKI
Only applies when MCLKI is selected as master source clock.
Also applies to MCLKO falling edge when MCLKO = MCLKI/2 or MCLKI/4
HMCLKO = 1 / 2MCLKO. MCLKO has the same duty cycle as MCLKI when MCLKO = MCLKI. When MCLKO = 0.5 MCLKI or 0.25
MCLKI, the duty cycle of MCLKO is typically 50%.
10. When MCLKO is derived from MCLKI, MCLKO jitter = MCLKI jitter
XTALI
tc(1)
tw(MCLKI)
MCLKI
tc(2)
tw(MCLKO)
td(MI-MO)
tf(MCLKO)
tr(MCLKO)
MCLKO
tc(3)
Figure 4−1. Master Clock Signals Timing Waveforms
4−3
4.4.2
Control Signals Over Recommended Operating Conditions (Unless Otherwise Noted)
PARAMETER
TEST CONDITIONS
MIN
10
TYP
MAX
UNITS
tw1(L)
Pulse duration, RST low
ns
tpd1
tpd2
Propagation delay, PWRDN high to power down state asserted
See Note 11
µs
Propagation delay, PWRDN low to power down state deasserted
See Note 12
µs
NOTES: 11. The maximum worst case value for tpd1 is given by
t
pd1_worst_case
80
+ 4096 ) GPIOFSCOUNT )
LRCLK
Microprocessor_Clock
12. tpd2 is determined by the time it takes the internal digital PLL to reach a locked condition, which, in turn, is governed by the MCLKI/
XTALI frequency and the PLL output frequency. For a 135-MHz PLL output and an MCLKI value of 24.576 MHz, tpd2 is typically 25 µs.
For an 11.264-MHz PLL output clock and a 1.024-MHz MCLKI/XTALI master clock, tpd2 is typically 360 µs.
tw1(L)
RST
tpd2
PWRDN
tpd1
Figure 4−2. Control Signals Timing Waveforms
4−4
4.4.3
Serial Audio Port Slave Mode Signals Over Recommended Operating Conditions (unless
otherwise noted)
TEST
CONDITIONS
PARAMETER
MIN
fLRCLK
tw(SCLKIN)
Frequency, LRCLK (FS)
Pulse duration, SCLKIN high
See Note 14
0.25 HSCLKIN
fSCLKIN
tcyc
Frequency, SCLKIN
See Note 13
Cycle time, SCLKIN
See Note 13
32FS
40
tpd1
TYP
MAX
HSCLKIN
0.75 HSCLKIN
8
96
25
Propagation delay, SCLKIN falling edge to
SDOUT
tsu1
th1
Setup time, LRCLK to SCLKIN rising edge
tsu2
Setup time, SDIN to SCLKIN rising edge
th2
Hold time, SDIN from SCLKIN rising edge
tpd2
Propagation delay,
SCLKIN falling edge
to SCLKOUT2 falling
edge
Hold time, LRCLK from SCLKIN rising edge
UNITS
kHz
ns
MHz
1/32FS
ns
15.1
ns
6.6
ns
0
ns
1.15
ns
2.3
ns
SCLKOUT2 = SCLKIN
12.4
ns
SCLKOUT2 < SCLKIN
12.5
ns
NOTES: 13. Typical duty cycle is 50/50.
14. HSCLKIN = 1/2fSCLKIN
tw(SCLKIN)
tcyc
SCLKIN
th1
tsu1
LRCLK
(Input)
tpd1
SDOUT1
SDOUT2
SDOUT3
th2
tsu2
SDIN1
SDIN2
SDIN3
SDIN4
tpd2
SCLKOUT2
Figure 4−3. Serial Audio Port Slave Mode Timing Waveforms
4−5
4.4.4
Serial Audio Port Master Mode Signals Over Recommended Operating Conditions
(unless otherwise noted)
PARAMETER
TEST CONDITIONS
MIN
TYP
8
MAX
f(LRCLK)
tr(LRCLK)
Frequency LRCLK
Rise time, LRCLK
CL = 30 pF
11.4
ns
tf(LRCLK)
f(SCLKOUT)
Fall time, LRCLK
CL = 30 pF
11.2
ns
Frequency (1/tcyc), SCLKOUT1/SCLKOUT2
See Note 13
25
MHz
tr(SCLKOUT)
tf(SCLKOUT)
Rise time, SCLKOUT1/SCLKOUT2
CL = 30 pF
9.5
ns
Fall time, SCLKOUT1/SCLKOUT2
CL = 30 pF
9.8
ns
tpd1(SCLKOUT1)
tpd1(SCLKOUT2)
Propagation delay, SCLKOUT1 falling edge to LRCLK edge
4.1
ns
Propagation delay, SCLKOUT2 falling edge to LRCLK edge
4.3
ns
tpd2
tsu
Propagation delay, SCLKOUT2 falling edge to SDOUT
3.4
ns
th
tsk
Hold time, SDIN from SCLKOUT1 rising edge
Setup time, SDIN to SCLKOUT1 rising edge
96
UNITS
32FS
18.4
ns
23
ns
Skew time, SCLKOUT1 to SCLKOUT2
0.8
3
NOTE 13: Typical duty cycle is 50/50.
tr(SCLKOUT)
tf(SCLKOUT)
SCLKOUT2
tr(SCLKOUT)
tsk
tf(SCLKOUT)
SCLKOUT1
tpd1(SCLKOUT2)
tpd1(SCLKOUT1)
LRCLK
(Output)
tf(LRCLK), tr(LRCLK)
SDOUT1
tpd2
SDOUT2
SDOUT3
th
SDIN1
tsu
SDIN2
SDIN3
SDIN4
Figure 4−4. TAS3100 Serial Audio Port Master Mode Timing Waveforms
4−6
kHz
ns
4.4.5
I2C Slave Mode Interface Signals Over Recommended Operating Conditions (unless
otherwise noted)
STANDARD MODE
TEST
CONDITIONS
PARAMETER
MIN
TYP
0
FAST MODE
MAX
MIN
100
0
TYP
MAX
fSCL
Frequency, SCL
tW(H)
Minimum pulse duration,
SCL high
See Note 15
0.43
0.12
µs
tW(L)
Minimum pulse duration,
SCL low
See Note 15
1.3
0.9
µs
tr
tf
Rise time, SDAOutput
CL = 330 pF
48.3
48.3
ns
Fall time, SDAOutput
CL = 330 pF
35.9
35.9
ns
tsu1
Setup time, SDAInput to SCL See Note 15
th1
Hold
time
tbuf
Minimum bus free time
between STOP and START
condition
See Note 15
0.82
0.38
µs
tsu2
Minimum setup time, SCL to
START condition
See Note 15
0.7
0.4
µs
th2
Minimum hold time, START
condition to SCL
See Note 15
1.5
0.05
µs
tsu3
Minimum setup time, SCL to
STOP condition
See Note 15
0.9
0.4
µs
0
SCL to SDAInput
See Note 15
0
SCL to SDAOutput
See Note 16
Toversamp
400
UNITS
0
ns
0
2Toversamp
kHz
ns
Toversamp
2Toversamp
µs
NOTES: 15. The maximum and/or minimum values for the TAS3103 I2C port parameters meets those characteristics of the SDA and SCL bus
lines listed below, as specified in the I2C bus devices bus specification.
16. Toversamp is the period of the oversample clock provided to the I2C master salve controller. This clock is dependent on the
microprocessor clock setting and the value set for the variable N in subaddress 0xFB. See Figure 2−14.
SPECIFIED I2C BUS CHARACTERISTICS
PARAMETER
Pulse duration high, tw(H)
SCL
Pulse duration low, tw(L)
Frequency fSCL
SDAInput
START/STOP conditions
STANDARD MODE
FAST MODE
4 µs MIN
0.6 µs MIN
4.7 µs MIN
1.3 µs MIN
100 kHz
400 kHz
Setup time tsu1, SDAInput to SCL
250 ns MIN
100 ns MIN
Hold Time th1, SCL to SDAInput
Bus free time between STOP and START condition tbuf
0 ns MIN
4.7 µs MIN
0 ns MIN
1.3 µs MIN
Setup time tsu2, SCL to START condition
4.7 µs MIN
0.6 µs MIN
Hold time th2, START condition to SCL
4 µs MIN
0.6 µs MIN
Setup time tsu3, SCL to STOP condition
4 µs MIN
0.6 µs MIN
4−7
tw(L)
tw(H)
tr
tf
SCL
tsu1
th1
SDA
Figure 4−5. I2C SCL and SDA Timing Waveforms
SCL
tsu2
tsu3
th2
SDA
Figure 4−6. I2C Start and Stop Conditions Timing Waveforms
4−8
t(buf)
A−1
1
1
Mix A to b
Mix A to c
Mix A to d
Mix A to e
Mix A to f
Mix B to a
Mix B to b
Mix B to c
Mix B to d
Mix B to e
Mix B to f
Mix C to a
Mix C to b
Mix C to c
Mix C to d
Mix C to e
Mix C to f
Mix D to a
Mix D to b
Mix D to c
Mix D to d
Mix D to e
Mix D to f
Mix E to a
Mix E to b
Mix E to c
Mix E to d
Mix E to e
Mix E to f
Mix F to a
0x02
0x03
0x04
0x05
0x06
0x07
0x08
0x09
0x0A
0x0B
0x0C
0x0D
0x0E
0x0F
0x10
0x11
0x12
0x13
0x14
0x15
0x16
0x17
0x18
0x19
0x1A
0x1B
0x1C
0x1D
0x1E
0x1F
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Mix A to a
1
Starting I2C check word
NUMBER OF
4-BYTE WORDS
0x01
REGISTER NAME
0x00
SUBADDRESS
(0xSS)
A.1 I2C Subaddress Table
u(31:28)F_a(27:24), F_a(23:16), F_a(15:8), F_a(7:0)
u(31:28)E_f(27:24), E_f(23:16), E_f(15:8), E_f(7:0)
u(31:28)E_e(27:24), E_e(23:16), E_e(15:8), E_e(7:0)
u(31:28)E_d(27:24), E_d(23:16), E_d(15:8), E_d(7:0)
u(31:28)E_c(27:24), E_c(23:16), E_c(15:8), E_c(7:0)
u(31:28)E_b(27:24), E_b(23:16), E_b(15:8), E_b(7:0)
u(31:28)E_a(27:24), E_a(23:16), E_a(15:8), E_a(7:0)
u(31:28)D_f(27:24), D_f(23:16), D_f(15:8), D_f(7:0)
u(31:28)D_e(27:24), D_e(23:16), D_e(15:8), D_e(7:0)
u(31:28)D_d(27:24), D_d(23:16), D_d(15:8), D_d(7:0)
u(31:28)D_c(27:24), D_c(23:16), D_c(15:8), D_c(7:0)
u(31:28)D_b(27:24), D_b(23:16), D_b(15:8), D_b(7:0)
u(31:28)D_a(27:24), D_a(23:16), D_a(15:8), D_a(7:0)
u(31:28)C_f(27:24), C_f(23:16), C_f(15:8), C_f(7:0)
u(31:28)C_e(27:24), C_e(23:16), C_e(15:8), C_e(7:0)
u(31:28)C_d(27:24), C_d(23:16), C_d(15:8), C_d(7:0)
u(31:28)C_c(27:24), C_c(23:16), C_c(15:8), C_c(7:0)
u(31:28)C_b(27:24), C_b(23:16), C_b(15:8), C_b(7:0)
u(31:28)C_a(27:24), C_a(23:16), C_a(15:8), C_a(7:0)
u(31:28)B_f(27:24), B_f(23:16), B_f(15:8), B_f(7:0)
u(31:28)B_e(27:24), B_e(23:16), B_e(15:8), B_e(7:0)
u(31:28)B_d(27:24), B_d(23:16), B_d(15:8), B_d(7:0)
u(31:28)B_c(27:24), B_c(23:16), B_c(15:8), B_c(7:0)
u(31:28)B_b(27:24), B_b(23:16), B_b(15:8), B_b(7:0)
u(31:28)B_a(27:24), B_a(23:16), B_a(15:8), B_a(7:0)
u(31:28)A_f(27:24), A_f(23:16), A_f(15:8), A_f(7:0)
u(31:28)A_e(27:24), A_e(23:16), A_e(15:8), A_e(7:0)
u(31:28)A_d(27:24), A_d(23:16), A_d(15:8), A_d(7:0)
u(31:28)A_c(27:24), A_c(23:16), A_c(15:8), A_c(7:0)
u(31:28)A_b(27:24), A_b(23:16), A_b(15:8), A_b(7:0)
u(31:28)A_a(27:24), A_a(23:16), A_a(15:8), A_a(7:0)
SCW(31:24), SCW(23:16), SCW(15:8), SCW(7:0)
CONTENTS
(u Indicates Unused Bits)
Appendix A
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x40, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x40, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x81, 0x42, 0x24, 0x18
INITIALIZATION VALUE
A−2
1
1
Mix F to d
Mix F to e
Mix F to f
Mix a to c
Mix b to c
Mix a to g
Mix b to h
Mix a to d via BQ and Rev/D
Mix a to e via BQ and Rev/D
Mix b to d via BQ and Rev/D
Mix b to e via BQ and Rev/D
Mix g to d via BQ
Mix g to e via BQ
Mix h to d via BQ
Mix h to e via BQ
Mix c to d via BQ
Mix c to e via BQ
Mix f to g and h
a_de path, biquad 1
a_de path, biquad 2
a_de path, biquad 3
0x22
0x23
0x24
0x25
0x26
0x27
0x28
0x29
0x2A
0x2B
0x2C
0x2D
0x2E
0x2F
0x30
0x31
0x32
0x33
0x34
0x35
0x36
5
5
5
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Mix F to c
1
Mix F to b
NUMBER OF
4-BYTE WORDS
0x21
REGISTER NAME
0x20
SUBADDRESS
(0xSS)
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
INITIALIZATION VALUE
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
u(31:28)f_gh(27:24), f_gh(23:16), f_gh(15:8), f_gh(7:0)
u(31:28)c_e(27:24), c_e(23:16), c_e(15:8), c_e(7:0)
u(31:28)c_d(27:24), c_d(23:16), c_d(15:8), c_d(7:0)
u(31:28)h_e(27:24), h_e(23:16), h_e(15:8), h_e(7:0)
u(31:28)h_d(27:24), h_d(23:16), h_d(15:8), h_d(7:0)
u(31:28)g_e(27:24), g_e(23:16), g_e(15:8), g_e(7:0)
u(31:28)g_d(27:24), g_d(23:16), g_d(15:8), g_d(7:0)
u(31:28)b_e(27:24), b_e(23:16), b_e(15:8), b_e(7:0)
u(31:28)b_d(27:24), b_d(23:16), b_d(15:8), b_d(7:0)
u(31:28)a_e(27:24), a_e(23:16), a_e(15:8), a_e(7:0)
u(31:28)a_d(27:24), a_d(23:16), a_d(15:8), a_d(7:0)
u(31:28)b_h(27:24), b_h(23:16), b_h(15:8), b_h(7:0)
u(31:28)a_g(27:24), a_g(23:16), a_g(15:8), a_g(7:0)
u(31:28)b_c(27:24), b_c(23:16), b_c(15:8), b_c(7:0)
u(31:28)a_c(27:24), a_c(23:16), a_c(15:8), a_c(7:0)
u(31:28)F_f(27:24), F_f(23:16), F_f(15:8), F_f(7:0)
u(31:28)F_e(27:24), F_e(23:16), F_e(15:8), F_e(7:0)
u(31:28)F_d(27:24), F_d(23:16), F_d(15:8), F_d(7:0)
u(31:28)F_c(27:24), F_c(23:16), F_c(15:8), F_c(7:0)
u(31:28)F_b(27:24), F_b(23:16), F_b(15:8), F_b(7:0)
CONTENTS
(u Indicates Unused Bits)
A−3
a_de path, biquad 4
b_de path, biquad 1
b_de path, biquad 2
b_de path, biquad 3
b_de path, biquad 4
g_de path, biquad 1
g_de path, biquad 2
0x38
0x39
0x3A
0x3B
0x3C
0x3D
REGISTER NAME
0x37
SUBADDRESS
(0xSS)
5
5
5
5
5
5
5
NUMBER OF
4-BYTE WORDS
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
INITIALIZATION VALUE
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
CONTENTS
(u Indicates Unused Bits)
A−4
g_de path, biquad 3
g_de path, biquad 4
h_de path, biquad 1
h_de path, biquad 2
h_de path, biquad 3
h_de path, biquad 4
c_de path, biquad 1
0x3F
0x40
0x41
0x42
0x43
0x44
REGISTER NAME
0x3E
SUBADDRESS
(0xSS)
5
5
5
5
5
5
5
NUMBER OF
4-BYTE WORDS
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
INITIALIZATION VALUE
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
CONTENTS
(u Indicates Unused Bits)
A−5
c_de path, biquad 2
c_de path, biquad 3
c_de path, biquad 4
f_CH3 path, biquad 1
f_CH3 path, biquad 2
f_CH3 path, biquad 3
f_CH3 path, biquad 4
0x46
0x47
0x48
0x49
0x4A
0x4B
REGISTER NAME
0x45
SUBADDRESS
(0xSS)
5
5
5
5
5
5
5
NUMBER OF
4-BYTE WORDS
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
INITIALIZATION VALUE
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
CONTENTS
(u Indicates Unused Bits)
A−6
CH1 biquad 1
CH1 biquad 2
CH1 biquad 3
CH1 biquad 4
CH1 biquad 5
CH1 biquad 6
0x50
0x51
0x52
0x53
0x54
f_CH3 path, Reverb Gain Rg1
f_CH3 path, Reverb Gain Rg0
b_de path, Reverb Gain Rg1
b_de path, Reverb Gain Rg0
0x4F
0x4E
0x4D
a_de path, Reverb Gain Rg0
0x4C
a_de path, Reverb Gain Rg1
REGISTER NAME
SUBADDRESS
(0xSS)
5
5
5
5
5
5
2
2
2
NUMBER OF
4-BYTE WORDS
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
INITIALIZATION VALUE
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
u(31:28)Rg1(27:24), Rg1(23:16), Rg1(15:8), Rg1(7:0)
u(31:28)Rg0(27:24), Rg0(23:16), Rg0(15:8), Rg0(7:0)
u(31:28)Rg1(27:24), Rg1(23:16), Rg1(15:8), Rg1(7:0)
u(31:28)Rg0(27:24), Rg0(23:16), Rg0(15:8), Rg0(7:0)
u(31:28)Rg1(27:24), Rg1(23:16), Rg1(15:8), Rg1(7:0)
u(31:28)Rg0(27:24), Rg0(23:16), Rg0(15:8), Rg0(7:0)
CONTENTS
(u Indicates Unused Bits)
A−7
CH1 biquad 7
CH1 biquad 8
CH1 biquad 9
CH1 biquad 10
CH1 biquad 11
CH1 biquad 12
CH2 biquad 1
0x56
0x57
0x58
0x59
0x5A
0x5B
REGISTER NAME
0x55
SUBADDRESS
(0xSS)
5
5
5
5
5
5
5
NUMBER OF
4-BYTE WORDS
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
INITIALIZATION VALUE
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
CONTENTS
(u Indicates Unused Bits)
A−8
CH2 biquad 2
CH2 biquad 3
CH2 biquad 4
CH2 biquad 5
CH2 biquad 6
CH2 biquad 7
CH2 biquad 8
0x5D
0x5E
0x5F
0x60
0x61
0x62
REGISTER NAME
0x5C
SUBADDRESS
(0xSS)
5
5
5
5
5
5
5
NUMBER OF
4-BYTE WORDS
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
INITIALIZATION VALUE
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
CONTENTS
(u Indicates Unused Bits)
A−9
CH2 biquad 9
CH2 biquad 10
CH2 biquad 11
CH2 biquad 12
CH3 biquad 1
CH3 biquad 2
CH3 biquad 3
0x64
0x65
0x66
0x67
0x68
0x69
REGISTER NAME
0x63
SUBADDRESS
(0xSS)
5
5
5
5
5
5
5
NUMBER OF
4-BYTE WORDS
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
INITIALIZATION VALUE
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
CONTENTS
(u Indicates Unused Bits)
A−10
CH3 biquad 4
CH3 biquad 5
CH3 biquad 6
CH3 biquad 7
CH3 biquad 8
CH3 biquad 9
CH3 biquad 10
0x6B
0x6C
0x6D
0x6E
0x6F
0x70
REGISTER NAME
0x6A
SUBADDRESS
(0xSS)
5
5
5
5
5
5
5
NUMBER OF
4-BYTE WORDS
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
INITIALIZATION VALUE
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
CONTENTS
(u Indicates Unused Bits)
A−11
Bass and treble bypass 1
0x73
Mix w to m
Mix j to i
Mix l to k
Mix n to m
Mix j to o via DRC mult
0x78
0x79
0x7A
0x7B
0x7C
Mix dither1 to o
Mix dither2 to p
Mix dither3 to q
Mix delay3 to o
Mix delay3 to p
Mix o to r
Mix o to s
Mix p to r
0x80
0x81
0x82
0x83
0x84
0x85
0x86
DRC bypass 3
Mix n to q via DRC mult
0x7F
0x7E
DRC bypass 2
Mix l to p via DRC mult
DRC bypass 1
Mix v to k
0x77
0x7D
Mix u to i
Bass and treble inline 3
Bass and treble bypass 3
0x76
0x75
Bass and treble inline 2
Bass and treble bypass 2
Bass and treble inline 1
CH3 biquad 12
0x72
0x74
CH3 biquad 11
REGISTER NAME
0x71
SUBADDRESS
(0xSS)
1
1
1
1
1
1
1
1
2
2
2
1
1
1
1
1
1
2
2
2
5
5
NUMBER OF
4-BYTE WORDS
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)p_r(27:24), p_r(23:16), p_r(15:8), p_r(7:0)
u(31:28)o_s(27:24), o_s(23:16), o_s(15:8), o_s(7:0)
u(31:28)o_r(27:24), o_r(23:16), o_r(15:8), o_r(7:0)
u(31:28)Dth3_p(27:24), Dth3_p(23:16), Dth3_p(15:8), Dth3_p(7:0)
u(31:28)Dth3_o(27:24), Dth3_o(23:16), Dth3_o(15:8), Dth3_o(7:0)
u(31:28)Dth3_q(27:24), Dth3_q(23:16), Dth3_q(15:8), Dth3_q(7:0)
u(31:28)Dth2_p(27:24), Dth2_p(23:16), Dth2_p(15:8), Dth2_p(7:0)
u(31:28)DRCby3(27:24), DRCby3(23:16), DRCby3(15:8), DRCby3(7:0)
u(31:28)Dth1_o(27:24), Dth1_o(23:16), Dth1_o(15:8), Dth1_o(7:0)
u(31:28)DRCby2(27:24), DRCby2(23:16), DRCby2(15:8), DRCby2(7:0)
u(31:28)n_q(27:24), n_q(23:16), n_q(15:8), n_q(7:0)
u(31:28)DRCby1(27:24), DRCby1(23:16), DRCby1(15:8), DRCby1(7:0)
u(31:28)l_p(27:24), l_p(23:16), l_p(15:8), l_p(7:0)
u(31:28)j_o(27:24), j_o(23:16), j_o(15:8), j_o(7:0)
u(31:28)n_m(27:24), n_m(23:16), n_m(15:8), n_m(7:0)
u(31:28)l_k(27:24), l_k(23:16), l_k(15:8), l_k(7:0)
u(31:28)j_i(27:24), j_i(23:16), j_i(15:8), j_i(7:0)
u(31:28)w_m(27:24), w_m(23:16), w_m(15:8), w_m(7:0)
u(31:28)v_k(27:24), v_k(23:16), v_k(15:8), v_k(7:0)
u(31:28)u_i(27:24), u_i(23:16), u_i(15:8), u_i(7:0)
u(31:28)BT3(27:24), BT3(23:16), BT3(15:8), BT3(7:0)
u(31:28)BTby3(27:24), BTby3(23:16), BTby3(15:8), BTby3(7:0)
u(31:28)BT2(27:24), BT2(23:16), BT2(15:8), BT2(7:0)
u(31:28)BTby2(27:24), BTby2(23:16), BTby2(15:8), BTby2(7:0)
u(31:28)BT1(27:24), BT1(23:16), BT1(15:8), BT1(7:0)
0x00, 0x40, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x40, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
u(31:28)BTby1(27:24), BTby1(23:16), BTby1(15:8), BTby1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
INITIALIZATION VALUE
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
CONTENTS
(u Indicates Unused Bits)
A−12
1
1
1
1
Mix r to s and t
Mix z to Z
Mix z to Y
Mix z to X
Mix z to W
Mix z to V
Mix z to U
Mix y to Z
Mix y to Y
Mix y to X
Mix y to W
Mix y to V
Mix y to U
Mix x to Z
Mix x to Y
Mix x to X
Mix x to W
Mix x to V
Mix x to U
Mix r to Z
Mix r to Y
Mix r to X
Mix r to W
Mix r to V
Mix r to U
CH1 loudness log2 G
CH1 loudness log2 O
CH1 loudness G
CH1 loudness O
0x89
0x8A
0x8B
0x8C
0x8D
0x8E
0x8F
0x90
0x91
0x92
0x93
0x94
0x95
0x96
0x97
0x98
0x99
0x9A
0x9B
0x9C
0x9D
0x9E
0x9F
0xA0
0xA1
0xA2
0xA3
0xA4
0xA5
2
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Mix q to r
1
Mix p to t
NUMBER OF
4-BYTE WORDS
0x88
REGISTER NAME
0x87
SUBADDRESS
(0xSS)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:24), u(23:16), O47:40(15:8), O39:32(7:0)
O31:24(31:24), O23:16(23:16), O15:8(15:8), O7:0(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
LO31:24(31:24), LO23:16(23:16), LO15:8(15:8), LO7:0(7:0)
u(31:28)G(27:24), G(23:16), G(15:8), G(7:0)
0x00, 0x40, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
INITIALIZATION VALUE
u(31:24), u(23:16), LO47:40(15:8), LO39:32(7:0)
u(31:28)LG(27:24), LG(23:16), LG(15:8), LG(7:0)
u(31:28)r_U(27:24), r_U(23:16), r_U(15:8), r_U(7:0)
u(31:28)r_V(27:24), r_V(23:16), r_V(15:8), r_V(7:0)
u(31:28)r_W(27:24), r_W(23:16), r_W(15:8), r_W(7:0)
u(31:28)r_X(27:24), r_X(23:16), r_X(15:8), r_X(7:0)
u(31:28)r_Y(27:24), r_Y(23:16), r_Y(15:8), r_Y(7:0)
u(31:28)r_Z(27:24), r_Z(23:16), r_Z(15:8), r_Z(7:0)
u(31:28)x_U(27:24), x_U(23:16), x_U(15:8), x_U(7:0)
u(31:28)x_V(27:24), x_V(23:16), x_V(15:8), x_V(7:0)
u(31:28)x_W(27:24), x_W(23:16), x_W(15:8), x_W(7:0)
u(31:28)x_X(27:24), x_X(23:16), x_X(15:8), x_X(7:0)
u(31:28)x_Y(27:24), x_Y(23:16), x_Y(15:8), x_Y(7:0)
u(31:28)x_Z(27:24), x_Z(23:16), x_Z(15:8), x_Z(7:0)
u(31:28)y_U(27:24), y_U(23:16), y_U(15:8), y_U(7:0)
u(31:28)y_V(27:24), y_V(23:16), y_V(15:8), y_V(7:0)
u(31:28)y_W(27:24),y_W(23:16), y_W(15:8), y_W(7:0)
u(31:28)y_X(27:24), y_X(23:16), y_X(15:8), y_X(7:0)
u(31:28)y_Y(27:24), y_Y(23:16), y_Y(15:8), y_Y(7:0)
u(31:28)y_Z(27:24), y_Z(23:16), y_Z(15:8), y_Z(7:0)
u(31:28)z_U(27:24), z_U(23:16), z_U(15:8), z_U(7:0)
u(31:28)z_V(27:24), z_V(23:16), z_V(15:8), z_V(7:0)
u(31:28)z_W(27:24), z_W(23:16), z_W(15:8), z_W(7:0)
u(31:28)z_X(27:24), z_X(23:16), z_X(15:8), z_X(7:0)
u(31:28)z_Y(27:24), z_Y(23:16), z_Y(15:8), z_Y(7:0)
u(31:28)z_Z(27:24), z_Z(23:16), z_Z(15:8), z_Z(7:0)
u(31:28)r_st(27:24), r_st(23:16), r_st(15:8), r_st(7:0)
u(31:28)q_r(27:24), q_r(23:16), q_r(15:8), q_r(7:0)
u(31:28)p_t(27:24), p_t(23:16), p_t(15:8), p_t(7:0)
CONTENTS
(u Indicates Unused Bits)
A−13
CH2 loudness O
CH2 loudness biquad
CH3 loudness log2 G
CH3 loudness log2 O
CH3 loudness G
CH3 loudness O
CH3 loudness biquad
CH1/2 DRCE ae
0xAB
0xAC
0xAD
0xAE
0xAF
0xB0
0xB1
0xB3
0xB2
1
CH2 loudness G
0xA9
0xAA
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
LO31:24(31:24), LO23:16(23:16), LO15:8(15:8), LO7:0(7:0)
u(31:28)G(27:24), G(23:16), G(15:8), G(7:0)
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
LO31:24(31:24), LO23:16(23:16), LO15:8(15:8), LO7:0(7:0)
u(31:28)G(27:24), G(23:16), G(15:8), G(7:0)
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)k2’(27:24), k2’(23:16), k2’(15:8), k2’(7:0)
CH1/2 k2’
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x01
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x01
T231:24(31:24), T223:16(23:16), T215:8(15:8), T27:0(7:0)
u(31:28)k0’(27:24), k0’(23:16), k0’(15:8), k0’(7:0)
0x00, 0x00, 0x00, 0x00
T131:24(31:24), T123:16(23:16), T115:8(15:8), T17:0(7:0)
u(31:24), u(23:16), T247:40(15:8), T239:32(7:0)
0x00, 0x00, 0x00, 0x00
u(31:24), u(23:16), T147:40(15:8), T139:32(7:0)
u(31:28)1-ae(27:24), 1-ae(23:16), 1-ae(15:8), 1-ae(7:0)
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
u(31:28)ae(27:24), ae(23:16), ae(15:8), ae(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
O31:24(31:24), O23:16(23:16), O15:8(15:8), O7:0(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:24), u(23:16), O47:40(15:8), O39:32(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:24), u(23:16), LO47:40(15:8), LO39:32(7:0)
0x00, 0x40, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)LG(27:24), LG(23:16), LG(15:8), LG(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
O31:24(31:24), O23:16(23:16), O15:8(15:8), O7:0(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:24), u(23:16), O47:40(15:8), O39:32(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:24), u(23:16), LO47:40(15:8), LO39:32(7:0)
u(31:28)k1’(27:24), k1’(23:16), k1’(15:8), k1’(7:0)
3
4
2
5
2
1
2
1
5
2
0x00, 0x80, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x40, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)LG(27:24), LG(23:16), LG(15:8), LG(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
INITIALIZATION VALUE
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
CONTENTS
(u Indicates Unused Bits)
CH1/2 k1’
CH1/2 k0’
CH1/2 DRCE T2
CH1/2 DRCE T1
CH1/2 DRCE 1-ae
2
CH2 loudness log2 O
0xA8
1
CH2 loudness log2 G
0xA7
5
NUMBER OF
4-BYTE WORDS
CH1 loudness biquad
REGISTER NAME
0xA6
SUBADDRESS
(0xSS)
A−14
0xBC
0xBB
0xBA
0xB9
0xB8
0xB7
0xB6
0xB5
0xB4
SUBADDRESS
(0xSS)
Spectrum analyzer BQ1
Spectrum analyzer 1-asa
5
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x01, 0xFF, 0xFF, 0xFF
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
u(31:28)1-asa(27:24), 1-asa(23:16), 1-asa(15:8), 1-asa(7:0)
u(31:28)asa(27:24), asa(23:16), asa(15:8), asa(7:0)
u(31:28)1-ad(27:24), 1-ad(23:16), 1-ad(15:8), 1-ad(7:0)
2
CH3 DRCE 1-ad
Spectrum analyzer asa
u(31:28)ad(27:24), ad(23:16), ad(15:8), ad(7:0)
CH3 DRCE ad
O231:24(31:24), O223:16(23:16), O215:8(15:8), O27:0(7:0)
u(31:28)aa(27:24), aa(23:16), aa(15:8), aa(7:0)
u(31:28)1-aa(27:24), 1-aa(23:16), 1-aa(15:8), 1-aa(7:0)
4
0x01, 0xFF, 0xFF, 0xFF
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
O131:24(31:24), O123:16(23:16), O115:8(15:8), O17:0(7:0)
u(31:24), u(23:16), O247:40(15:8), O239:32(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:24), u(23:16), O147:40(15:8), O139:32(7:0)
CH3 DRCE 1-aa
CH3 DRCE aa
CH3 DRCE O2
4
u(31:28)k2’(27:24), k2’(23:16), k2’(15:8), k2’(7:0)
CH3 DRCE O1
u(31:28)k1’(27:24), k1’(23:16), k1’(15:8), k1’(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x01
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x01
T131:24(31:24), T123:16(23:16), T115:8(15:8), T17:0(7:0)
u(31:24), u(23:16), T247:40(15:8), T239:32(7:0)
T231:24(31:24), T223:16(23:16), T215:8(15:8), T27:0(7:0)
u(31:28)k0’(27:24), k0’(23:16), k0’(15:8), k0’(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x01, 0xFF, 0xFF, 0xFF
u(31:24), u(23:16), T147:40(15:8), T139:32(7:0)
CH3 k2’
3
4
u(31:28)1-ae(27:24), 1-ae(23:16), 1-ae(15:8), 1-ae(7:0)
CH3 k1’
CH3 k0’
CH3 DRCE T2
CH3 DRCE T1
CH3 DRCE 1-ae
u(31:28)ae(27:24), ae(23:16), ae(15:8), ae(7:0)
u(31:28)1-ad(27:24), 1-ad(23:16), 1-ad(15:8), 1-ad(7:0)
2
CH1/2 DRCE 1-ad
CH3 DRCE ae
u(31:28)ad(27:24), ad(23:16), ad(15:8), ad(7:0)
CH1/2 DRCE ad
O231:24(31:24), O223:16(23:16), O215:8(15:8), O27:0(7:0)
u(31:28)aa(27:24), aa(23:16), aa(15:8), aa(7:0)
0x01, 0xFF, 0xFF, 0xFF
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
O131:24(31:24), O123:16(23:16), O115:8(15:8), O17:0(7:0)
u(31:24), u(23:16), O247:40(15:8), O239:32(7:0)
INITIALIZATION VALUE
u(31:24), u(23:16), O147:40(15:8), O139:32(7:0)
CONTENTS
(u Indicates Unused Bits)
u(31:28)1-aa(27:24), 1-aa(23:16), 1-aa(15:8), 1-aa(7:0)
4
4
NUMBER OF
4-BYTE WORDS
CH1/2 DRCE 1-aa
CH1/2 DRCE aa
CH1/2 DRCE O2
CH1/2 DRCE O1
REGISTER NAME
A−15
Spectrum analyzer BQ2
Spectrum analyzer BQ3
Spectrum analyzer BQ4
Spectrum analyzer BQ5
Spectrum analyzer BQ6
Spectrum analyzer BQ7
Spectrum analyzer BQ8
0xBE
0xBF
0xC0
0xC1
0xC2
0xC3
REGISTER NAME
0xBD
SUBADDRESS
(0xSS)
5
5
5
5
5
5
5
NUMBER OF
4-BYTE WORDS
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x80, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
INITIALIZATION VALUE
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
CONTENTS
(u Indicates Unused Bits)
A−16
Spectrum analyzer BQ10
Dither LFSR1 mix
0xC5
0xC6
1
1
Factory test
Factory test
Mix G to g
Mix G to f
Mix G to Y
Mix H to h
Mix H to f
Mix H to Z
Mix d to aa
Mix e to aa
Reserved
Reserved
Reserved
Reserved
Reserved
Reserved
Reserved
Reserved
Reserved
Reserved
Reserved
Reserved
0xC9
0xCA
0xCB
0xCC
0xCD
0xCE
0xCF
0xD0
0xD1
0xD2
0xD3
0xD4
0xD5
0xD6
0xD7
0xD8
0xD9
0xDA
0xDB
0xDC
0xDD
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
1
Dither seed
0xC8
1
2
5
5
NUMBER OF
4-BYTE WORDS
0xC7
Dither LFSR2 mix
Spectrum analyzer BQ9
REGISTER NAME
0xC4
SUBADDRESS
(0xSS)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:28)e_aa(27:24), e_aa(23:16), e_aa(15:8), e_aa(7:0)
u(31:28)d_aa(27:24), d_aa(23:16), d_aa(15:8), d_aa(7:0)
u(31:28)H_Z(27:24), H_Z(23:16), H_Z(15:8), H_Z(7:0)
u(31:28)H_f(27:24), H_f(23:16), H_f(15:8), H_f(7:0)
u(31:28)H_h(27:24), H_h(23:16), H_h(15:8), H_h(7:0)
u(31:28)G_Y(27:24), G_Y(23:16), G_Y(15:8), G_Y(7:0)
u(31:28)G_f(27:24), G_f(23:16), G_f(15:8), G_f(7:0)
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
N/A
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:28)G_g(27:24), G_g(23:16), G_g(15:8), G_g(7:0)
N/A
N/A
0x00, 0x00, 0x22, 0x49
0x00, 0x80, 0x00, 0x00
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), LFSR2_SEED(15:8), LFSR1_SEED(7:0)
u(31:28)LFSR2(27:24), LFSR2(23:16), LFSR2(15:8), LFSR2(7:0)
0x00, 0x80, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
u(31:28)LFSR1(27:24), LFSR1(23:16), LFSR1(15:8), LFSR1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
u(31:28)b2(27:24), b2(23:16), b2(15:8), b2(7:0)
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)b1(27:24), b1(23:16), b1(15:8), b1(7:0)
0x00, 0x00, 0x00, 0x00
0x00, 0x80, 0x00, 0x00
u(31:28)b0(27:24), b0 (23:16), b0(15:8), b0(7:0)
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
0x00, 0x00, 0x00, 0x00
u(31:28)a2(27:24), a2(23:16), a2(15:8), a2(7:0)
INITIALIZATION VALUE
u(31:28)a1(27:24), a1(23:16), a1(15:8), a1(7:0)
CONTENTS
(u Indicates Unused Bits)
A−17
Reserved
Reserved
Reserved
Reserved
Reserved
Reserved
Reserved
Reserved
Reserved
Reserved
Reserved
Reserved
Watchdog timer enable
Factory test
Factory test
GPIO port I/O value
GPIO parameters
Master mute/unmute
Vol, T and B slew rates
CH1 volume (5.23 precision)
CH2 volume (5.23 precision)
CH3 volume (5.23 precision)
Bass filter set (1-5)
Bass filter index
Treble filter set (1-5)
Treble filter index
I2S command word
Delay/reverb times−CH1
0xE0
0xE1
0xE2
0xE3
0xE4
0xE5
0xE6
0xE7
0xE8
0xE9
0xEA
0xEB
0xEC
0xED
0xEE
0xEF
0xF0
0xF1
0xF2
0xF3
0xF4
0xF5
0xF6
0xF7
0xF8
0xFA
u(31:28)D3(27:24), D3(23:16), u(15:12)R3(11:8), R3(7:0)
1
1
Delay/reverb times−CH3
I2C M and N
Ending I2C check word
0xFB
0xFC
ECW(31:24), ECW(23:16), ECW(15:8), ECW(7:0)
u(31:24), u(23:16), u(15:8), u(7)M(6:3)N(2:0)
u(31:28)D2(27:24), D2(23:16), u(15:12)R2(11:8), R2(7:0)
u(31:28)D1(27:24), D1(23:16), u(15:12)R1(11:8), R1(7:0)
MLRCLK(31:24), SCLK(23:16), DWFMT(15:8), IOM(7:0)
1
3
u(31:24), CH3Tf(23:16), CH2Tf(15:8), CH1Tf(7:0)
u(31:24), u(23:19)CH3Ts(18:16), u(15:11) CH2Ts(10:8), u(7:3)CH1Ts(2:0),
u(31:24), CH3Bf(23:16), CH2Bf(15:8), CH1Bf(7:0)
u(31:24), u(23:19)CH3Bs(18:16), u(15:11) CH2Bs(10:8),
u(7:3)CH1Bs(2:0),
u(31:28)Vol3(27:24), Vol3(23:16), Vol3(15:8), Vol3(7:0)
u(31:28)Vol2(27:24), Vol2(23:16), Vol2(15:8), Vol2(7:0)
u(31:28)Vol1(27:24), Vol1(23:16), Vol1(15:8), Vol1(7:0)
u(31:24), u(23:16), u(15:9)VSC(8), TBLC(7:0)
(31:24), u(23:16), u(15:8), u(7:3)CH3M_U(2)CH2M_U(1)CH1M_U(0)
u(31:24), u(23:20)GPIODIR(19:16), GPIOFSCOUNT(15:8),
GPIO_samp_int(7:0)
0x81, 0x42, 0x24, 0x18
0x00, 0x00, 0x00, 0x41
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x01, 0x01, 0x09, 0x11
0x00, 0x72, 0x72, 0x72
0x00, 0x03, 0x03, 0x03
0x00, 0x72, 0x72, 0x72
0x00, 0x03, 0x03, 0x03
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x00
0x00, 0x00, 0x00, 0x40
0x00, 0x00, 0x00, 0x00
0x00, 0x0F, 0x6E, 0x6D
0x00, 0x00, 0x00, 0x0X(1)
N/A
u(31:24), u(23:16), u(15:8), u(7:4)GPIO_in_out(3:0)
N/A
u(31:24), u(23:16), u(15:8), u(7:0)
N/A
0x00, 0x00, 0x00, 0x01
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
INITIALIZATION VALUE
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:1)R1(0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
u(31:24), u(23:16), u(15:8), u(7:0)
CONTENTS
(u Indicates Unused Bits)
1
1
1
1
1
1
1
1
1
1
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
NUMBER OF
4-BYTE WORDS
Delay/reverb times−CH2
0xF9
Reserved
0xDF
REGISTER NAME
0xDE
SUBADDRESS
(0xSS)
A−18
SA7(7:0)
SA8(7:0)
SA9(7:0)
SA10(7:0)
Spectrum analyzer output 7
Spectrum analyzer output 8
Spectrum analyzer output 9
Spectrum analyzer output 10
Flag register
VU meter output 2 (SA6)
0.25
SA6(7:0)
Spectrum analyzer output 6
u(7:1)VolBusy(0)
SA6(7:0)
SA5(7:0)
SA5(7:0)
Spectrum analyzer output 5
0.5
SA4(7:0)
Spectrum analyzer output 4
VU meter output 1 (SA5)
SA3(7:0)
Spectrum analyzer output 3
SA1(7:0)
SA2(7:0)
2.5
NUMBER OF
4-BYTE WORDS
Spectrum analyzer output 2
Spectrum analyzer output 1
REGISTER NAME
CONTENTS
(u Indicates Unused Bits)
N/A
(Always data dependent)
(Always data dependent)
(Always data dependent)
(Always data dependent)
(Always data dependent)
(Always data dependent)
(Always data dependent)
(Always data dependent)
(Always data dependent)
(Always data dependent)
(Always data dependent)
(Always data dependent)
INITIALIZATION VALUE
NOTE 1: GPIO ports are initialized to be read ports. The initial input values read then are dependent on what is connected to the GPIO pins. If a given GPIO pin is left unconnected,
the internal pullup results in a logic 1 being read.
0xFF
0xFE
0xFD
SUBADDRESS
(0xSS)
A−19
H
G
F
E
D
C
B
A
c
b
a
f
h
g
4
BQ
4
BQ
4
BQ
4
BQ
4
BQ
4
BQ
Rev
Del
Rev
Del
Rev
Del
A.2 TAS3103 Firmware Block Diagram
e
d
12
Bq
12
Bq
aa
12
Bq
BTby2
BT2
BT1
BT3
BTby3
Bass &
Treble
Bass &
Ch3 Treble
Ch2
Ch1
Bass &
Treble
BTby1
w
v
u
Soft
Vol
m
Soft
Vol
k
i
Soft
Vol
j
Loud−
ness
Dynamic
Range
Control
Loud−
ness
n
l
Dynamic
Range
Control
Loud−
ness
r
DRCby3
q
DRCby2
DRCby1
p
o
Delay1
Delay3
Delay2
Dither 3
32-Bit
Clip
32-Bit
Clip
32−48
xpndr
32−48
xpndr
r
x
y
32−48 z
xpndr
Sp Analyzer
+ VU Meter
Dither 2
t
s
Dither 1
32-Bit
Clip
ZZ
Y
Y
X
W
V
V
U
U
A−20
Appendix B
Mechanical Information
DBT (R-PDSO-G**)
PLASTIC SMALL-OUTLINE PACKAGE
30 PINS SHOWN
0,50
0,27
0,17
30
16
0,08 M
0,15 NOM
4,50
4,30
6,60
6,20
Gage Plane
0,25
1
15
0°−ā 8°
0,75
0,50
A
Seating Plane
0,15
0,05
1,20 MAX
PINS **
0,10
20
24
28
30
38
44
50
A MAX
5,10
6,60
7,90
7,90
9,80
11,10
12,60
A MIN
4.90
6,40
7,70
7,70
9,60
10,90
12,40
DIM
4073252/E 02/02
NOTES: A.
B.
C.
D.
All linear dimensions are in millimeters.
This drawing is subject to change without notice.
Body dimensions do not include mold flash or protrusion.
Falls within JEDEC MO-153
B−1
B−2