HMP8115 NTSC/PAL Video Decoder April 1998 Features Description • (M) NTSC and (B, D, G, H, I, M, N, NC) PAL Operation - Optional Auto Detect of Video Standard - ITU-R BT.601 (CCIR601) and Square Pixel Operation The HMP8115 is a high quality NTSC and PAL decoder with internal A/D converters. It is compatible with NTSC M, PAL B, D, G, H, I, M, N, and combination N (NC) video standards. • Digital Output Formats - VMI Compatible - 8-Bit, 16-Bit 4:2:2 YCbCr - 15-Bit (5,5,5), 16-Bit (5,6,5) RGB - Linear or Gamma-Corrected - 8-Bit BT.656 Both composite and S-video (Y/C) input formats are supported. A 2-line comb filter plus a user-selectable chrominance trap filter provide high quality Y/C separation. User adjustments include brightness, contrast, saturation, hue, and sharpness. Data during the vertical blanking interval (VBI), such as closed captioning, widescreen signalling and teletext, may be captured and output as BT.656 ancillary data. Closed captioning and widescreen signalling information may also be read out via the I2C interface. • Analog Input Formats - Three Analog Composite Inputs - Analog Y/C (S-Video) Input • “Sliced” VBI Data Capture Capabilities - Closed Captioning - Widescreen Signalling (WSS) - BT.653 System B, C and D Teletext - NABTS (North American Broadcast Teletext) - WST (World System Teletext) Ordering Information • 2-Line (1H) Comb Filter Y/C Separator NOTES: PART NUMBER HMP8115CN HMPVIDEVAL/ISA TEMP. RANGE (oC) 0 to 70 PACKAGE PKG. NO. 80 Ld PQFP Q80.14x20 Evaluation Board: ISA Frame Grabber • Fast I2C Interface 1. PQFP is also known as QFP and MQFP. • Two 8-Bit ADCs 2. Evaluation Board and Reference Design descriptions are in the Applications section. Applications • Multimedia PCs • Video Conferencing • Video Compression Systems • Video Security Systems • LCD Projectors and Overhead Panels • Related Products - NTSC/PAL Encoders: HMP815x, HMP817x - NTSC/PAL Decoders: HMP8112A • Related Literature - AN9644: Composite Video Separation Techniques - AN9716: Widescreen Signalling - AN9717: YCbCr to RGB Considerations - AN9728: BT.656 Video Interface for ICs - AN9738: VMI Video Interface for ICs CAUTION: These devices are sensitive to electrostatic discharge. Users should follow proper IC Handling Procedures. http://www.intersil.com or 407-727-9207 | Copyright © Intersil Corporation 1999 1 File Number 4283.5 HMP8115 Table of Contents PAGE Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 External Video Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ANALOG VIDEO INPUTS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ANTI-ALIASING FILTERS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . S-VIDEO CHROMA GAIN CONTROL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 5 5 5 Digitization of Video . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A/D CONVERSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AGC AND DC RESTORATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . INPUT SIGNAL DETECTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VERTICAL SYNC AND FIELD DETECTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Y/C SEPARATION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . INPUT SAMPLE RATE CONVERTER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . COMB FILTER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CHROMA DEMODULATION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . OUTPUT SAMPLE RATE CONVERTER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CLK2 INPUT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 5 5 6 6 6 6 6 6 6 7 Digital Processing of Video . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . UV TO CbCr CONVERSION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DIGITAL COLOR GAIN CONTROL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . COLOR KILLER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Y PROCESSING. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CbCr PROCESSING. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . YCbCr OUTPUT FORMAT PROCESSING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RGB OUTPUT FORMAT PROCESSING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . BUILT-IN VIDEO GENERATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 7 7 7 7 8 8 8 9 Pixel Port Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 HSYNC AND VSYNC TIMING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FIELD TIMING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . BLANK AND DVALID TIMING. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PIXEL OUTPUT PORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-BIT YCbCr OUTPUT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-BIT YCbCr, 15-BIT RGB, OR 16-RGB OUTPUT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-BIT BT.656 OUTPUT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 9 10 11 11 12 14 Advanced Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CLOSED CAPTIONING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WIDESCREEN SIGNALLING (WSS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . BT.656 ANCILLARY DATA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . BT.656 CLOSED CAPTIONING AND WIDE SCREEN SIGNALLING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TELETEXT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . REAL TIME CONTROL INTERFACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 15 16 17 17 18 20 Host Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 HMP8115 Control Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Pinout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Pin Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Applications Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PCB LAYOUT CONSIDERATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . EVALUATION BOARD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RELATED APPLICATION NOTES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 38 38 38 Electrical Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 2 3 C CLAMP LOGIC AND GAIN CONTROL AGC AND CLAMP LOGIC INPUT MUX EXTERNAL ANTIALIASING FILTER C_CAP GAIN_CTRL LAGC_CAP L_CAP NTSC-PAL3/Y NTSC-PAL2 NTSC-PAL1 + + - 8-BIT ADC CLAMP RESET INTREQ MICROPROCESSOR INTERFACE AND CONTROL 8-BIT ADC DIGITAL COMPARATOR SYNC LEVEL BLACK LEVEL WHITE PEAK LEVEL DIGITAL COMPARATORS YIN EXTERNAL ANTIALIASING FILTER YOUT Functional Block Diagram SDA SCL INPUT SAMPLE RATE CONVERTER COLOR TRAP FIELD VSYNC LINE LOCK PLL VSYNC DETECT LOCKED HSYNC DETECT COLOR ADJUST USER ADJUST CHROMA PLL COLOR DEMODULATION Y/C SEPARATION VBI DETECTION AND DECODING LOGIC VBI STATUS BITS OUTPUT TIMING AND FIFO OUTPUT SAMPLE RATE CONVERTER P[15:0] DVALID HSYNC BLANK VBIVALID RGB LOGIC USER ADJUST. HMP8115 Functional Block Diagram (Continued) CLK (24.54, 27.0 OR 29.5MHz) 4FSC CLOCK HUE ADJUST CHROMA PLL LOOP FILTER CHROMA PLL NCO FIELD VSYNC GENLOCK LOSS VSYNC DETECT CHROMA PHASE DETECTOR HSYNC DETECT LINE LOCKED PLL LOOP FILTER CLK TO 4FSC RATIO AGC ADJUST LINE LOCKED NCO SATURATION ADJUST CR[7:0] C C,CVBS DATA 4 INPUT SAMPLE RATE CONVERTER Y,CVBS Y DATA C DATA LINE DELAY COMB FILTER CHROMA DEMODULATOR U,V U, V TO CbCr COLOR SPACE CONVERTER AND COLOR KILLER LP FILTER CbCr ENABLE Y DATA HORIZONTAL M Y DATA AND VERTICAL Y DATA U SHARPNESS X ADJUST SHARPNESS ADJUST OUTPUT SAMPLE RATE CONVERTER CHROMA TRAP Y ENABLE VBI DETECTION AND DECODING LOGIC LOCKED HMP8115 M U X C,CVBS DATA HSYNC M U X SYNC STRIPPER, BRIGHTNESS, AND CONTRAST ADJUST STANDARD SELECT RGB LOGIC LP FILTER MUX MUX ENABLE P[15:0] HSYNC,VSYNC, BLANK, FIELD, DVALID, VBIVALID OUTPUT TIMING AND FIFO HMP8115 Introduction inside the decoder as determined by bits 7 and 6 of the COLOR PROCESSING register 06H. In addition to the internal AGC, the designer can also apply some gain to the chroma before it reaches the internal AGC logic. This gain is controlled by pin 28. The voltage at this pin determines the gain of the chroma before it gets digitized by the chroma A/D with a typical gain performance as shown in Figure 2. The HMP8115 is designed to decode baseband composite or S-video NTSC and PAL signals, and convert them to either digital YCbCr or RGB data. In addition to performing the basic decoding operations, the HMP8115 includes hardware to decode different types of VBI data and to generate digital video patterns for a blue screen, black screen and full screen color bars. 7 The digital PLLs are designed to synchronize to all NTSC and PAL standards. A chroma PLL is used to maintain chroma lock for demodulation of the color information; a linelocked PLL is used to maintain vertical spatial alignment. The PLLs are designed to maintain lock even in the event of VCR headswitches and multipath noise. 6 TEMPERATURE = 25oC VCC = 5V LINEAR GAIN 5 The HMP8115 contains two 8-bit A/D converters and an I2C interface for programming internal registers. 4 3 2 External Video Processing 1 Before a video signal can be digitized the decoder has some external processing considerations that need to be addressed. This section discusses those external aspects of the HMP8115. 0 1.6 2.2 2.4 2.6 2.8 3.0 3.2 3.4 FIGURE 2. CHROMINANCE AMPLIFIER GAIN The HMP8115 supports either three composite or two composite and one S-video input. Digitization of Video Three analog video inputs (NTSC/PAL 1-3) are used to select which one of three composite video sources are to be decoded. To support S-video applications, the Y channel drives the NTSC/PAL 3 analog input, and the C channel drives the C analog input. Prior to A/D conversion, the video signal is DC restored and gained to generate known video levels into the digital processing logic. This process is addressed in the “AGC and DC Restoration” section. After digitization, sample rate converters and a comb filter are used to perform color separation and demodulation. The analog inputs must be AC-coupled to the video signals, as shown in the Applications section. A/D CONVERSION Video data is sampled at the CLK2 frequency then processed by the input sample rate converter. The output levels of the ADC after AGC processing are: ANTI-ALIASING FILTERS An external anti-alias filter is required to achieve optimum performance and prevent high frequency components from being aliased back into the video image. (M) NTSC (M, N) PAL For the NTSC/PAL 1-3 inputs, a single filter is connected between the YOUT and YIN pins. For the C input, the antialiasing filter should be connected before the C input. A recommended filter is shown in Figure 1. R1 L1 332 8.2µH C1 33pF 2.0 CONTROL VOLTAGE ON THE GAIN_CTRL PIN ANALOG VIDEO INPUTS YOUT 1.8 C2 82pF white black blank sync YIN 224 76 64 0 (B, D, G, H, I, NC) PAL 213 64 64 0 AGC AND DC RESTORATION R2 4.02K S-VIDEO CHROMA GAIN CONTROL The AGC amplifier attenuates or amplifies the analog video signal to ensure that the sync tip level generates code 0. The difference from the ideal sync tip level of 0 is used to control the amount of attenuation or gain of the analog video signal. The capacitor on the LAGC_CAP pin is used to store the voltage which sets the gain level of the input video amplifier. The Chroma portion of S-video is AC coupled through an anti-aliasing filter as shown in the applications section. Unlike the composite/luma inputs, the Automatic Gain Control (AGC) for the chroma portion of S-video is done digitally DC restoration positions the video signal such that the DC level of the back porch generates an average code 64. The back porch is sampled to determine the average value. The capacitor on the L_CAP pin is used to store the voltage FIGURE 1. RECOMMENDED ANTI-ALIASING FILTER 5 HMP8115 Y/C SEPARATION which sets the DC offset level into the input video amplifier. During the S-video mode of operation the capacitor on the C_CAP pin performs the same function as the L_CAP capacitor, except the chroma video amplifier’s DC offset is set so that the chroma A/D generates a code of 128 during the back porch. The internal timing windows for AGC and DC Restoration are show in Figure 3. A composite video signal has the luma (Y) and chroma (C) information mixed in the same video signal. The Y/C separation process is responsible for separating the composite video signal into these two components. The HMP8115 utilizes a comb filter to minimize the artifacts that are associated with the Y/C separation process. INPUT SAMPLE RATE CONVERTER The input sample rate converter is used to convert video data sampled at the CLK2 rate to a virtual 4xfSC sample rate for comb filtering and color demodulation. An interpolating filter is used to generate the 4xfSC samples as illustrated in Figure 4. VIDEO INPUT INCOMING VIDEO SAMPLES DC RESTORE TIME AGC RESAMPLED VIDEO FIGURE 3. AGC AND DC RESTORE INTERNAL TIMING TIME 4xfSC INPUT SIGNAL DETECTION It is assumed there is no video input if a horizontal sync is not detected for 16 consecutive lines. When no video has been detected, nominal video timing is generated for the previously detected or programmed standard. A maskable interrupt is included to flag when no video has been detected (bit 6 of the INTERRUPT MASK register 0FH) allowing for blue/black/color bar output modes to be enabled if desired. The vertical sync interrupt can be used in determining when a video signal is present at the currently selected video mux input. Bit 0 of register 0FH is used to enable vertical sync interrupts. FIGURE 4. SAMPLE RATE CONVERSION COMB FILTER A 2-line comb filter, using a single line delay, is used to perform part of the Y/C separation process. During S-video operation, the Y signal bypasses the comb filter; the C signal is processed by the comb filter since it is an integral part of the chroma demodulator. During PAL operation, the chroma trap filter should also be enabled for improved performance. Since a single line store is used, the chroma will normally have a half-line vertical offset from the luma data. This may be eliminated, vertically aligning the chroma and luma samples, at the expense of vertical resolution of the luma. Bit 0 of the OUTPUT FORMAT register 02H controls this option. VERTICAL SYNC AND FIELD DETECTION The vertical sync and field detect circuit uses a low time counter to detect the vertical sync sequence in the video data stream. The low time counter accumulates the low time encountered during any sync pulse, including serration and equalization pulses. When the low time count exceeds the vertical sync detect threshold, VSYNC is asserted immediately. FIELD is asserted at the same time that VSYNC is asserted. FIELD is asserted low for odd fields and high for even fields. Field is determined from the location in the video line where VSYNC is detected. If VSYNC is detected in the first half of the line, the field is odd. If VSYNC is detected in the second half of a line, the field is even. CHROMA DEMODULATION The output of the comb filter is further processed using a patented frequency domain transform to complete the Y/C separation and demodulate the chromanance. Demodulation is done at a virtual 4xfSC sample rate using the interpolated data samples to generate U and V data. The demodulation process decimates by 2 the U/V sample rate. OUTPUT SAMPLE RATE CONVERTER In the case of lost vertical sync or excessive noise that would prevent the detection of vertical sync, the FIELD output will continue to toggle. Lost vertical sync is declared if after 337 lines, a vertical sync period was not detected for 1 or 3 (selectable) successive fields as specified by bit 2 of the GENLOCK CONTROL register 04H. When this occurs, the PLLs are initialized to the acquisition state. The output sample rate converter converts the Y, U and V data from a virtual 4xfSC sample rate to the desired output sample rate (i.e., 13.5MHz). It also vertically aligns the samples based on the horizontal sync information embedded in the digital video data stream. The output sample rate is determined by the selected video standard and whether 6 HMP8115 with the two previous lines having a color burst to limit lineto-line variations. A gain of 0.5x to 4x is used for Cb and Cr. square or rectangular pixels are output. The output format is 4:2:2 for all modes except the RGB modes which use a 4:4:4 output format. If “fixed gain control” is selected, the amplitude of the color difference signals (CbCr) is multiplied by a constant, regardless of variations in the color burst amplitude. The constant gain value is specified by the COLOR GAIN register 1CH. A gain of 0.5x to 4x is used for Cb and Cr. Limiting the gain to 4x limits the amount of amplified noise. CLK2 INPUT Note that the color subcarrier is derived from CLK2. Any jitter on CLK2 will be transferred to the color subcarrier, resulting in color changes. Thus, CLK2 should be derived from a stable clock source, such as a crystal. The use of a PLL to generate CLK2 is not recommended. CLK2 must have a 50ppm accuracy and at least a 60/40% duty cycle to ensure proper operation. If “freeze automatic gain control” is selected, the amplitude of the color difference signals (CbCr) is multiplied by a constant. This constant is the value the AGC circuitry generated when the “freeze automatic gain” command was selected. The CLK2 clock rate must be one of the following frequencies: COLOR KILLER 24.54MHz 27.00MHz 29.50MHz The frequency of CLK2 must be 2x the desired output sample rate. The values in Table 1 below indicate the CLK2 clock rate based on the video standard and pixel mode. The output sample rate for the given video standard and pixel mode is half the CLK2 clock rate. If “enable color killer” is selected, the color output is turned off when the running average of the color burst amplitude is below approximately 25% of nominal for four consecutive fields. When the running average of the color burst amplitude is above approximately 25% of nominal for four consecutive fields, the color output is turned on. The color output is also turned off when excessive phase error of the chroma PLL is present. TABLE 1. VIDEO STANDARD CLOCK RATE SELECTION SUMMARY If “force color off” is selected, color information is never present on the outputs. If “force color on” is selected, color information is present on the outputs regardless of the color burst amplitude or chroma PLL phase error. ALLOWABLE CLK2 FREQUENCIES (MHz) RECTANGULAR PIXEL MODE SQUARE PIXEL MODE (M) NTSC 27.00 24.54 (B, D, G, H, I, N) PAL 27.00 29.50 (M) PAL 27.00 24.54 (NC) PAL 27.00 29.50 VIDEO FORMAT Y PROCESSING The black level is subtracted from the luminance data to remove sync and any blanking pedestal information. Negative values of Y are supported at this point to allow proper decoding of “below black” luminance levels. Scaling is done to position black at 8-bit code 0 and white at 8-bit code 219. Digital Processing of Video Once the luma and chroma have been separated the HMP8115 then performs programmable modifications (i.e. contrast, coring, color space conversions, color AGC, etc.) to the decoded video signal. The baseband U and V signals are scaled and offset to generate a nominal range of 16-240 for both the Cb and Cr data. A chroma trap filter may be used to remove any residual color subcarrier from the luminance data. The center frequency of the chroma trap is automatically determined from the video standard being decoded. The chroma trap should be disabled during S-video operation to maintain maximum luminance bandwidth. Alternately, a 3MHz lowpass filter may be used to remove high-frequency Y data. This may make a noisy image more pleasing to the user, although softer. DIGITAL COLOR GAIN CONTROL Coring of the high-frequency Y data may be done to reduce low-level high frequency noise. UV TO CbCr CONVERSION There are four types of color gain control modes available: no gain control, automatic gain control, fixed gain control, and freeze automatic gain control. Coring of the Y data may also be done to reduce low-level noise around black. This forces Y data with the following values to a value of 0: If “no gain control” is selected, the amplitude of the color difference signals (CbCr) is not modified, regardless of variations in the color burst amplitude. Thus, a gain of 1x is always used for Cb and Cr. coring = 1: ± 1 coring = 2: ± 1, ± 2 coring = 3: ± 1, ± 2. ± 3 If “automatic gain control” is selected, the amplitude of the color difference signals (CbCr) is compensated for variations in the color burst amplitude. The burst amplitude is averaged High-frequency components of the luminance signal may be “peaked” to control the sharpness of the image. Maximum gain may be selected to occur at either 2.6MHz or the color 7 HMP8115 The 15-bit R′G′B′ data may be converted to 15-bit linear RGB, using the following equations. Although the PAL specifications specify a gamma of 2.8, a gamma of 2.2 is normally used. The HMP8115 allows the selection of the gamma to be either 2.2 or 2.8, independent of the video standard. subcarrier frequency. This may be used to make the displayed image more pleasing to the user. It should not be used if the output video will be compressed, as the circuit introduces high-frequency components that will reduce the compression ratio. for gamma = 2.2: The brightness control adds or subtracts a user-specified DC offset to the Y data. The contrast control multiplies the Y data by a user-specified amount. These may be used to make the displayed image more pleasing to the user. for R′G′B′ < 0.0812*31 R = (31)((R′/31)/4.5) G = (31)((G′/31)/4.5) B = (31)((B′/31)/4.5) Finally, a value of 16 is added to generate a nominal range of 16 (black) to 235 (white). for R′G′B′ >= 0.0812*31 CbCr PROCESSING R = (31)(((R′/31) + 0.099)/1.099)2.2 G = (31)(((G′/31) + 0.099)/1.099)2.2 B = (31)(((B′/31) + 0.099)/1.099)2.2 The CbCr data is lowpass filtered to either 0.85 or 1.5MHz. Coring of the CbCr data may be done to reduce low-level noise around zero. This forces CbCr data with the following values to a value of 128. for gamma = 2.8: coring = 1: 127, 129 coring = 2: 126, 127, 129, 130 coring = 3: 125, 126, 127, 129, 130, 131 R = (31)(R′/31)2.8 G = (31)(G′/31)2.8 B = (31)(B′/31)2.8 The saturation control multiplies the CbCr data by a userspecified amount. This may be used to make the displayed image more pleasing to the user. The CbCr data may also be optionally multiplied by the contrast value to avoid color shifts when changing contrast. 16-Bit R′G′B′ The following YCbCr to R′G′B′ equations are used to maintain the proper black and white levels: R′ = 0.142(Y - 16) + 0.194(Cr - 128) G′ = 0.288(Y - 16) - 0.201(Cr - 128) - 0.097(Cb - 128) B′ = 0.142(Y - 16) + 0.245(Cb - 128) The hue control provides a user-specified phase offset to the color subcarrier during decoding. This may be used to correct slight hue errors due to transmission. The resulting 16-bit R′G′B′ data has a range of 0 to 31 for R′ and B′, and a range of 0 to 63 for G′. Values less than 0 are made 0; R′ and B′ values greater than 31 are made 31, G′ values greater than 63 are made 63. YCbCr OUTPUT FORMAT PROCESSING Y has a nominal range of 16 to 235. Cb and Cr have a nominal range of 16 to 240, with 128 corresponding to zero. Values less than 1 are made 1 and values greater than 254 are made 254. The 16-bit R′G′B′ data may be converted to 16-bit linear RGB, using the following equations. Although the PAL specifications specify a gamma of 2.8, a gamma of 2.2 is normally used. The HMP8115 allows the selection of the gamma to be either 2.2 or 2.8, independent of the video standard. While BLANK is asserted, Y is forced to have a value of 16, with Cb and Cr forced to have a value of 128, unless VBI data is present. for gamma = 2.2: RGB OUTPUT FORMAT PROCESSING for R′B′ < 0.0812*31, G′ < 0.0812*63 The 4:2:2 YCbCr data is converted to 4:4:4 YCbCr data and then converted to either 15-bit or 16-bit gamma-corrected RGB (R′G′B′) data. While BLANK is asserted, RGB data is forced to a value of 0. R = (31)((R′/31)/4.5) G = (63)((G′/63)/4.5) B = (31)((B′/31)/4.5) for R′B′ >= 0.0812*31, G′ >= 0.0812*63 15-Bit R′G′B′ R = (31)(((R′/31) + 0.099)/1.099)2.2 G = (63)(((G′/63) + 0.099)/1.099)2.2 B = (31)(((B′/31) + 0.099)/1.099)2.2 The following YCbCr to R′G′B′ equations are used to maintain the proper black and white levels: R′ = 0.142(Y - 16) + 0.194(Cr - 128) G′ = 0.142(Y - 16) - 0.099(Cr - 128) - 0.048(Cb - 128) B′ = 0.142(Y - 16) + 0.245(Cb - 128) for gamma = 2.8: R = (31)(R′/31)2.8 G = (63)(G′/63)2.8 B = (31)(B′/31)2.8 The resulting 15-bit R′G′B′ data has a range of 0 to 31. Values less than 0 are made 0 and values greater than 31 are made 31. 8 HMP8115 BUILT-IN VIDEO GENERATION a fixed latency due to internal pipeline processing. The pulse width of the HSYNC is defined by the END HSYNC register 36H, where the trailing edge of HSYNC has a programmable delay of 0-510 CLK2 cycles from the leading edge. When the blue screen, black screen or color bar output is selected, a full-screen of blue, black or 75% colorbar output is generated using the currently selected output format. The type of screen to be generated is determined by bits 2 and 1 of the OUTPUT FORMAT register 02H. When built-in video generation is not desired, the bits need to be set for normal operation to pass decoded video. The leading edge of VSYNC is asserted approximately half way through the first serration pulse of each field. For an odd field, the trailing edge of VSYNC is 5±1 CLK2 cycles after the trailing edge of the HSYNC that follows the last equalization pulse. Refer to Figures 5 and 7. For an even field, the trailing edge of VSYNC is 5±1 CLK2 cycles leading the leading edge of the HSYNC that follows the last equalization pulse. Refer to Figures 6 and 8. If a video source is input, it will be used to provide the video timing. If an input video source is not detected, internallygenerated video timing will be used. Pixel Port Timing FIELD TIMING The the timing and format of the output data and control signals is presented in the following sections. When field information can be determined from the input video source, the FIELD output pin reflects the video source field state. When field information cannot be determined from the input video source, the FIELD output pin alternates its state at the beginning of each field. FIELD changes state 5±1 CLK2 cycles before the leading edge of VSYNC. HSYNC AND VSYNC TIMING The HSYNC and VSYNC output timing is VMI v1.4 compatible. Figures 5-8 illustrate the video timing. The leading edge of HSYNC is synchronous to the video input signal and has NTSC(M) LINE # 524 525 1 2 3 4 5 6 7 8 9 10 PAL(M) LINE # 521 522 523 524 525 1 2 3 4 5 6 7 VIDEO INPUT HSYNC VSYNC FIELD ‘ODD’ FIELD ‘EVEN’ FIELD NOTE: 3. The trailing edge of VSYNC is 5±1 clocks after the trailing edge of HSYNC to be VMI compatible and to indicate a transition to an odd field. FIGURE 5. NTSC(M) AND PAL(M) HSYNC, VSYNC AND FIELD TIMING DURING AN EVEN TO ODD FIELD TRANSITION NTSC(M) LINE # PAL(M) LINE # 262 263 264 265 266 267 268 269 270 271 272 273 259 260 261 262 263 264 265 266 267 268 269 270 VIDEO INPUT HSYNC VSYNC FIELD ‘ODD’ FIELD ‘EVEN’ FIELD NOTE: 4. The trailing edge of VSYNC is 5±1 clocks after the leading edge of HSYNC to be VMI compatible and to indicate a transition to an even field. FIGURE 6. NTSC(M) AND PAL(M) HSYNC, VSYNC AND FIELD TIMING DURING AN ODD TO EVEN FIELD TRANSITION 9 HMP8115 LINE # 621 622 623 624 625 1 2 3 4 5 6 7 VIDEO INPUT HSYNC VSYNC FIELD ‘ODD’ FIELD ‘EVEN’ FIELD NOTE: 5. The trailing edge of VSYNC is 5±1 clocks after the trailing edge of HSYNC is to be VMI compatible and to indicate a transition to an odd field. FIGURE 7. PAL(B,D,G,H,I,N,NC) HSYNC, VSYNC AND FIELD TIMING DURING AN EVEN TO ODD FIELD TRANSITION LINE # 309 310 311 312 313 314 315 316 317 318 319 320 VIDEO INPUT HSYNC VSYNC FIELD ‘ODD’ FIELD ‘EVEN’ FIELD NOTE: 6. The trailing edge of VSYNC is 5±1 clocks after the leading edge of HSYNC to be VMI compatible and to indicate a transition to an even field. FIGURE 8. PAL(B,D,G,H,I,N,NC) HSYNC, VSYNC AND FIELD TIMING DURING AN ODD TO EVEN FIELD TRANSITION BLANK AND DVALID TIMING During active scan lines BLANK is negated when the horizontal pixel count matches the value in the END H_BLANK register 32H. A count of 00H corresponds to the 50% point of the leading edge of the sync tip after leaving the part. BLANK is asserted when the horizontal pixel count matches the value in the START H_BLANK register 31H/30H. Note that horizontally, BLANK is programmable with two pixel resolution. DVALID is asserted when P15-P0 contain valid data. The timing and behavior of DVALID is dependent on the output video format and the programmed values for bit 4 (DVLD_DCYC) and bit 5 (DVLD_LTC) of the GENLOCK CONTROL register 04H. Refer to the specific output video format sections that follow for the specific behavior for DVALID. START V_BLANK register 34H/33H and END V_BLANK register 35H determine which scan lines are blanked for each field. During inactive scan lines, BLANK is asserted during the entire scan line. Half-line blanking of the output video cannot be done. Reference Figure 9 for active video timing and use Table 2 for typical blanking programming values. BLANK is used to determine if the HMP8115 is generating active video data. BLANK should be used in conjunction with DVALID to capture digital data from the decoder. BLANK, DVALID and the video data are output after the internal pipeline latency and synchronous with the rising edge of CLK2. TABLE 2. TYPICAL VALUES FOR HBLANK AND VBLANK REGISTERS VIDEO STANDARD (MSB/LSB) ACTIVE PIXELS/ LINE TOTAL PIXELS/ LINE LAST PIXEL COUNT START H_BLANK (31H/30H) END H_BLANK (32H) START V_BLANK (34H/33H) END V_BLANK (35H) 720 720 858 864 857 (0359H) 863 (035FH) 842 (034AH) 852 (0354H) 122 (7AH) 132 (84H) 259 (0103H) 310 (0136H) 19 (13H) 22 (16H) 640 768 780 944 779 (030BH) 943 (03AFH) 758 (02F6H) 922 (039AH) 118 (76H) 154 (9AH) 259 (0103H) 310 (0136H) 19 (13H) 22 (16H) RECTANGULAR PIXELS NTSC (M), PAL (M) PAL (B, D, G, H, I,N, NC) SQUARE PIXELS NTSC (M), PAL (M) PAL (B, D, G, H, I,N, NC) 10 HMP8115 NTSC M PAL B, D, G, H, I, N, NC LINES 1 - 22 NOT ACTIVE LINES 1 - 22 NOT ACTIVE ODD FIELD SYNC AND BACK PORCH 240 ACTIVE LINES PER FIELD (LINES 23-262) 288 ACTIVE LINES PER FIELD (LINES 23 - 310) VERTICAL BLANKING 480 ACTIVE LINES/FRAME (NTSC, PAL M) EVEN FIELD LINES 263 - 284 NOT ACTIVE 240 ACTIVE LINES PER FIELD (LINES 285 - 524) LINE 525 NOT ACTIVE LINES 311 - 335 NOT ACTIVE FRONT PORCH 576 ACTIVE LINES/FRAME (PAL) 288 ACTIVE LINES PER FIELD (LINES 336 - 623) NUMBER OF PIXELS RECTANGULAR (SQUARE) NTSC LINES 624-625 NOT ACTIVE PAL TOTAL PIXELS 858 (780) 864 (944) TOTAL PIXELS ACTIVE PIXELS 720 (640) 720 (768) ACTIVE PIXELS NOTE: 7. The line numbering for PAL (M) followings NTSC (M) line count minus 3 per the video standards. FIGURE 9. TYPICAL ACTIVE VIDEO REGIONS TABLE 3. PIXEL OUTPUT FORMATS PIN NAME 8-BIT, 4:2:2, YCbCr 16-BIT, 4:2:2, YCbCr 15-BIT, RGB, (5,5,5) 16-BIT, RGB, (5,6,5) BT.656 P0 P1 P2 P3 P4 P5 P6 P7 0 0 0 0 0 0 0 0 Cb0, Cr0 Cb1, Cr1 Cb2, Cr2 Cb3, Cr3 Cb4, Cr4 Cb5, Cr5 Cb6, Cr6 Cb7, Cr7 B0 B1 B2 B3 B4 G0 G1 G2 B0 B1 B2 B3 B4 G0 G1 G2 0 0 0 0 0 0 0 0 P8 P9 P10 P11 P12 P13 P14 P15 Y0, Cb0, Cr0 Y1, Cb1, Cr1 Y2, Cb2, Cr2 Y3, Cb3, Cr3 Y4, Cb4, Cr4 Y5, Cb5, Cr5 Y6, Cb6, Cr6 Y7, Cb7, Cr7 Y0 Y1 Y2 Y3 Y4 Y5 Y6 Y7 G3 G4 R0 R1 R2 R3 R4 0 G3 G4 G5 R0 R1 R2 R3 R4 YCbCr Data, Ancillary Data, SAV and EAV Sequences PIXEL OUTPUT PORT If DLVD_LTC=1, DVALID has the same internal timing as the first mode, but is ANDed with the CLK2 signal, and the result is output onto the DVALID pin. This results in a gated CLK2 signal being output during the active video time on active scan lines. Refer to Figure 11. Pixel data is output via the P0-P15 pins. Refer to Table 3 for the output pin definition as a function of the output mode. 8-BIT YCbCr OUTPUT If 8-bit YCbCr data is generated, it is output following each rising edge of CLK2. The YCbCr data is multiplexed as [Cb Y Cr Y′ Cb Y Cr Y′...], with the first active data each scan line containing Cb data. The pixel output timing is shown in Figures 10 and 11. The DVALID output pin may be configured to operate in one of two ways. The configuration is determined by the DVLD_LTC bit (bit 4) of the GENLOCK CONTROL register 04H. If DVLD_LTC=0, the DVALID output is continuously asserted during the entire active video time on active scan lines if CLK2 is exactly 2x the desired output sample rate. DVALID being asserted indicates valid pixel data is present on the P15-P8 pixel outputs. DVALID is never asserted during the blanking intervals. Refer to Figure 10. BLANK, HSYNC, VSYNC, DVALID, VBIVALID, and FIELD are output following the rising edge of CLK2. When BLANK is asserted and VBIVALID is deasserted, the YCbCr outputs have a value of 16 for Y and 128 for Cb and Cr. 11 HMP8115 CLK DVALID BLANK Cb0 P[15-8] NOTE: Y0 Cr0 Y1 Cb2 Y2 Cr2 Y3 Cb4 Y4 tDVLD 8. Y0 is the first active luminance pixel data of a line. Cb0 and Cr0 are first active chrominance pixel data in a line. Cb and Cr will alternate every cycle due to the 4:2:2 subsampling. Pixel data is not output during the blanking period, but the values on the ports are forced to blanking levels. FIGURE 10. OUTPUT TIMING FOR 8-BIT YCbCr MODE (DVLD_LTC = 0) CLK DVALID BLANK P[15-8] NOTES: Cb0 Y0 Cr0 Y1 Cb2 Y2 Cr2 Y3 Cb4 Y4 tDVLD 9. Y0 is the first active luminance pixel data of a line. Cb0 and Cr0 are first active chrominance pixel data in a line. Cb and Cr will alternate every cycle due to the 4:2:2 subsampling. Pixel data is not output during the blanking period, but the values on the ports are forced to blanking levels. 10. When DVLD_LTC is set to 1, the polarity of DVALID needs to be set to active low, otherwise DVALID will stay low during active video and be gated with the clock only during the blanking interval. FIGURE 11. OUTPUT TIMING FOR 8-BIT YCbCr MODE (DVLD_LTC = 1) 16-BIT YCbCr, 15-BIT RGB, OR 16-RGB OUTPUT which DVALID did not have a 50% duty cycle timing and other timing variations. The timing diagrams for this mode can be found in Figures 14 and 15. In these output modes, DVALID may be configured to operate in one of four modes as controlled by the DVLD_LTC and DVLD_DCYC bits of the GENLOCK CONTROL register (04H). Bit 4 is the DVLD_LTC bit and bit 5 is the DVLD_DCYC bit. If DVLD_LTC=1 and DVLD_DCYC=0, DVALID is present the entire line time on all scan lines. DVALID may occasionally be negated for two consecutive CLK2 cycles just prior to active video. In this mode DVALID is guaranteed have a 50% duty cycle only during the active video times. The timing for this mode differs from the timing shown in Figures 12 and 13 only in that DVALID will also be asserted during the blanking portion of the video line time as described above. If DVLD_LTC=0 and DVLD_DCYC=0, DVALID is present only during the active video time on active scan lines. Thus, DVALID being asserted indicates valid pixel data is present on the P0-P15 pixel outputs. DVALID is never asserted during the blanking intervals. In this mode DVALID will have a 50% duty cycle only during the active video times. The timing diagrams for this mode can be found in Figures 12 and 13. If DVLD_LTC=1 and DVLD_DCYC=1, DVALID is present during the entire line time on all scan lines. DVALID is asserted during the blanking intervals as needed to ensure a constant number of total samples per line. The timing for this mode differs from the timing shown in Figures 14 and 15 only in that DVALID will also be asserted during the blanking portion of the video line time as described above. If DVLD_LTC=0 and DVLD_DCYC=1, DVALID behaves the same as the first mode, with the exception that DVALID does not have a 50% duty cycle. This mode is intended for backward compatibility with HMP8112(A) timing dependencies in 12 HMP8115 BLANK, HSYNC, VSYNC, DVALID, VBIVALID, and FIELD are output following the rising edge of CLK2. When BLANK is asserted and VBIVALID is deasserted, the YCbCr outputs have a value of 16 for Y and 128 for Cb and Cr; the RGB outputs have a value of 0. If 16-bit YCbCr, 15-bit RGB data, or 16-bit RGB data is generated, it is output following the rising edge of CLK2 while DVALID is asserted. Either linear or gamma-corrected RGB data may be output. The pixel output timing is shown in Figures 12 to 15. CLK DVALID BLANK P15-P8 Y0 Y1 Y2 Y3 Y4 P7-P0 Cb0 Cr0 Cb2 Cr2 Cb4 tDVLD NOTES: 11. Y0 is the first active luminance pixel data of a line. Cb0 and Cr0 are first active chrominance pixel data in a line. Cb and Cr will alternate every cycle due to the 4:2:2 subsampling. 12. BLANK is asserted per Figure 9. FIGURE 12. OUTPUT TIMING FOR 16-BIT YCbCr MODE (DVLD_LTC = 0, DVLD_DCYC = 0) CLK DVALID P15-P11 [P14-P10] R0 R1 R2 R3 R4 P10-P5 [P9-P5] G0 G1 G2 G3 G4 P4-P0 B0 B1 B2 B3 B4 tDVLD NOTE: 13. BLANK is asserted per Figure 9. FIGURE 13. OUTPUT TIMING FOR 16-BIT [15-BIT] RGB MODE (DVLD_LTC = 0, DVLD_DCYC = 0) 13 HMP8115 CLK DVALID P15-P8 P7-P0 Y0 Y1 Y2 Y3 Y4 Cb0 Cr0 Cb2 Cr2 Cb4 tDVLD NOTES: 14. Y0 is the first active luminance pixel of a line. Cb0 and Cr0 are first active chrominance pixels in a line. Cb and Cr will alternate every cycle due to the 4:2:2 subsampling. 15. BLANK is asserted per Figure 9. 16. DVALID is asserted for every valid pixel during both active and blanking regions. DVALID is not a 50% duty cycle synchronous output and will appear to jitter as the Output Sample Rate converter adjusts the output timing for various data rates and clock frequency inputs. FIGURE 14. OUTPUT TIMING FOR 16-BIT YCbCr MODE (DVLD_LTC = 0, DVLD_DCYC = 1) CLK DVALID BLANK P15-P11 [P14-P10] R0 R1 R2 R3 R4 P10-P5 [P9-P5] G0 G0 G2 G2 G4 P4-P0 B0 B1 B2 B3 B4 NOTES: tDVLD 17. BLANK is asserted per Figure 9. 18. DAVLID is asserted for every valid pixel during both active and blanking regions. DVALID is not a 50% duty cycle synchronous output and will appear to jitter as the Output Sample Rate converter adjusts the output timing for various data rates and clock frequency inputs. FIGURE 15. OUTPUT TIMING FOR 16-BIT [15-BIT] RGB MODE (DVLD_LTC = 0, DVLD_DCYC = 1) 8-BIT BT.656 OUTPUT If BT.656 data is generated, it is output following each rising edge of CLK2. The BT.656 EAV and SAV formats are shown in Table 4 and the pixel output timing is shown in Figure 16. The EAV and SAV timing is determined by the programmed horizontal and vertical blank timing During the blanking intervals, the YCbCr outputs have a value of 16 for Y and 128 for Cb and Cr, unless ancillary data is present. Due to the use of digital PLLs and source video timing the total # of samples per line may not equal exactly 1716 (NTSC) or 1728 (PAL). The active video portion of the BT.656 data stream is always exactly 1440 continuous samples. Any line-to-line timing difference from nominal # of samples per line, plus or minus, is accommodated in the horizontal blanking interval. BLANK, HSYNC, VSYNC, DVALID, VBIVALID, and FIELD are output following the rising edge of CLK2. For proper operation, CLK2 must be exactly 2x the desired output sample rate. The DVALID output is continuously asserted during the entire active video time. 14 HMP8115 . CLK DVALID BLANK P[15-8] FF 00 00 Status Cb0 Y0 Cr0 Y1 Cb2 Y2 tDVLD NOTES: 19. Y0 is the first active luminance pixel data of a line. Cb0 and Cr0 are first active chrominance pixel data in a line. Cb and Cr will alternate every cycle due to the 4:2:2 subsampling. Pixel data is not output during the blanking period. 20. Notice that DVALID is not asserted during the preamble and that BLANK is still asserted. 21. See Table 4 for Status bit definitions. FIGURE 16. OUTPUT TIMING FOR 8-BIT BT.656 MODE TABLE 4. BT.656 EAV AND SAV SEQUENCES PIXEL INPUT P15 P14 P13 P12 P11 P10 P9 P8 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 F V H P3 P2 P1 P0 Preamble Status Word NOTES: 22. P3 = V xor H; P2 = F xor H; P1 = F xor V; P0 = F xor V xor H 23. F: “0” = field 1; “1” = field 2 24. V: “1” during vertical blanking 25. H: “0” at SAV (start of active video); “1” at EAV (end of active video) Advanced Features Detection of Closed Captioning In addition to digitizing an analog video signal the HMP8115 has hardware to process different types of Vertical Blanking Interval (VBI) data as described in the following sections. The closed caption decoder monitors the appropriate scan lines looking for the clock run-in and start bits used by captioning. If found, it locks to the clock run-in, the caption data is sampled and loaded into shift registers, and the data is then transferred to the caption data registers. “SLICED” VBI DATA CAPTURE If the clock run-in and start bits are not found, it is assumed the scan line contains video data unless other VBI information is detected, such as teletext. The HMP8115 implements “sliced” data capture of select types of VBI data. The VBI decoders incorporate detection hysteresis to prevent them from rapidly turning on and off due to noise and transmission errors. In order to handle realworld signals, the VBI decoders also compensate for DC offsets and amplitude variations. Once the clock run-in and start bits are found on the appropriate scan line for four consecutive odd fields, the Closed Captioning odd field Detect status bit is set to “1”. It is reset to “0” when the clock run-in and start bits are not found on the appropriate scan lines for four consecutive odd fields. CLOSED CAPTIONING During closed captioning capture, the scan lines containing captioning information are monitored. If closed captioning is enabled and captioning data is present, the caption data is loaded into the caption data registers. Once the clock run-in and start bits are found on the appropriate scan line for four consecutive even fields, the Closed Captioning even field Detect status bit is set to “1”. It is reset to “0” when the clock run-in and start bits are not found on the appropriate scan lines for four consecutive even fields. 15 HMP8115 Reading the Caption Data Detection of WSS The caption data registers may be accessed in two ways: via the I2C interface or as BT.656 ancillary data. The WSS decoder monitors the appropriate scan lines looking for the run-in and start codes used by WSS. If found, it locks to the run-in code, the WSS data is sampled and loaded into shift registers, and the data is then transferred to the WSS data registers. Captioning Disabled on Both Lines In this case, any caption data present is ignored. The Caption odd field Read status bit and the Caption even field Read status bit are always a “0”. If the run-in and start codes are not found, it is assumed the scan line contains video data unless other VBI information is detected, such as teletext. Odd Field Captioning Once the run-in and start codes are found on the appropriate scan line for four consecutive odd fields, the WSS Line 20 Detect status bit is set to “1”. It is reset to “0” when the run-in and start codes are not found on the appropriate scan lines for four consecutive odd fields. In this case, any caption data present on line 284 (or line 281 or 335 in the PAL modes) is ignored. Caption data present on line 21 (or line 18 or 22 in the PAL modes) is captured into a shift register then transferred to CLOSED CAPTION_ODD_A register 20H and CLOSED CAPTION_ODD_B register 21H. Once the run-in and start codes are found on the appropriate scan line for four consecutive even fields, the WSS Line 283 Detect status bit is set to “1”. It is reset to “0” when the clock run-in and start bits are not found on the appropriate scan lines for four consecutive even fields. The Caption even field Read status bit is always a “0”. The Caption odd field Read status bit is set to “1” after data has been transferred from the shift register to the CLOSED CAPTION_ODD_A and CLOSED CAPTION_ODD_B registers. It is set to “0” after the data has been read out. Reading the WSS Data The WSS data registers may be accessed in two ways: via the I2C interface or as BT.656 ancillary data. Even Field Captioning In this case, any caption data present on line 21 (or line 18 or 22 in the PAL modes) is ignored. Caption data present on line 284 (or line 281 or 335 in the PAL modes) is captured into a shift register then transferred to CLOSED CAPTION_EVEN_A register 22H and CLOSED CAPTION_EVEN_B register 23H. WSS Disabled on Both Lines In this case, any WSS data present is ignored. The WSS odd field Read status bit and the WSS even field Read status bit are always a “0”. The Caption odd field Read status bit is always a “0”. The Caption even field Read status bit is set to “1” after data has been transferred from the shift register to the CLOSED CAPTION_EVEN_A and CLOSED CAPTION_EVEN_B registers. It is set to “0” after the data has been read out. Odd Field WSS In this case, any WSS data present on line 283 (or line 280 or 336 in the PAL modes) is ignored. WSS data present on line 20 (or line 17 or 23 in the PAL modes) is captured into a shift register then transferred to the WSS_ODD_A and WSS_ODD_B data registers. Odd and Even Field Captioning Caption data present on line 21 (or line 18 or 22 in the PAL modes) is captured into a shift register then transferred to the CLOSED CAPTION_ODD_A and CLOSED CAPTION_ODD_B registers. Caption data present on line 284 (or line 281 or 335 in the PAL modes) is captured into a shift register then transferred to the CLOSED CAPTION_EVEN_A and CLOSED CAPTION_EVEN_B registers. The WSS even field Read status bit is always a “0”. The WSS odd field Read status bit is set to “1” after data has been transferred from the shift register to the WSS_ODD_A and WSS_ODD_B registers. It is set to “0” after the data has been read out. The Caption odd field Read status bit is set to “1” after data has been transferred from the shift register to the CLOSED CAPTION_ODD_A and CLOSED CAPTION_ODD_B registers. It is set to “0” after the data has been read out. Even Field WSS In this case, any WSS data present on line 20 (or line 17 or 23 in the PAL modes) is ignored. WSS data present on line 283 (or line 280 or 336 in the PAL modes) is captured into a shift register then transferred to the WSS_EVEN_A and WSS_EVEN_B data registers. The Caption even field Read status bit is set to “1” after data has been transferred from the shift register to the CLOSED CAPTION_EVEN_A and CLOSED CAPTION_EVEN_B registers. It is set to “0” after the data has been read out. The WSS odd field Read status bit is always a “0”. The WSS even field Read status bit is set to “1” after data has been transferred from the shift register to the WSS_EVEN_A and WSS_EVEN_B registers. It is set to “0” after the data has been read out. WIDESCREEN SIGNALLING (WSS) During WSS capture (ITU-R BT.1119 and EIAJ CPX-1204), the scan lines containing WSS information are monitored. If WSS is enabled and WSS data is present, the WSS data is loaded into the WSS data registers. 16 HMP8115 Real-Time Control Interface (RTCI) information. Teletext and RTCI data is only available as BT.656 ancillary data. Odd and Even WSS WSS data present on line 20 (or line 17 or 23 in the PAL modes) is captured into a shift register then transferred to the WSS_ODD_A and WSS_ODD_B registers. WSS data present on line 283 (or line 280 or 336 in the PAL modes) is captured into a shift register then transferred to the WSS_EVEN_A and WSS_EVEN_B registers. VBIVALID OUTPUT TIMING The VBIVALID output is asserted when outputting closed captioning, widescreen signalling, teletext or RTCI data as BT.656 ancillary data. It is asserted during the entire BT.656 ancillary data packet time, including the preamble. The WSS odd field Read status bit is set to “1” after data has been transferred from the shift register to the WSS_ODD_A and WSS_ODD_B registers. It is set to “0” after the data has been read out. BT.656 CLOSED CAPTIONING AND WIDE SCREEN SIGNALLING Table 5 illustrates the format when outputting the caption data registers as BT.656 ancillary data. The ancillary data is present during the horizontal blanking interval after the line containing the captioning information. The WSS even field Read status bit is set to “1” after data has been transferred from the shift register to the WSS_EVEN_A and WSS_EVEN_B registers. It is set to “0” after the data has been read out. Table 6 illustrates the format when outputting the WSS data registers as BT.656 ancillary data. The ancillary data is present during the horizontal blanking interval after the line containing the WSS information. BT.656 ANCILLARY DATA Through the BT.656 interface the HMP8115 can generate non-active video data which contains CC, WSS, teletext or CLK VBIVALID P[15-8] 00 FF FF DATA ID BLK # # BYTES/4 BYTE #1 BYTE #2 BYTE #3 BYTE #4 tDVLD NOTES: 26. BT.656 VBI ancillary starts with a 00H, FFH and FFH sequence which is opposite to the SAV/EAV sequence of FFH, 00H and 00H. 27. During active VBI data intervals, DVALID is deasserted and BLANK is asserted. FIGURE 17. OUTPUT TIMING FOR BT.656 VBI DATA TRANSFERS (CC, WSS, TELETEXT, RTCI) TABLE 5. READING THE CLOSED CAPTION DATA AS BT.656 ANCILLARY DATA PIXEL OUTPUT Preamble P15 P14 P13 P12 P11 P10 P9 P8 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Data ID P14 ep 1 1 0 0 0 0 = odd field data 1 = even field data Data Block Number P14 ep 0 0 0 0 0 1 Data Word Count P14 ep 0 0 0 0 0 1 Caption Data P14 ep 0 0 bit 15 bit 14 bit 13 bit 12 P14 ep 0 0 bit 11 bit 10 bit 9 bit 8 P14 ep 0 0 bit 7 bit 6 bit 5 bit 4 P14 ep 0 0 bit 3 bit 2 bit 1 bit 0 P14 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 CRC NOTES: 28. ep = even parity for P8-P13. 29. CRC = Sum of P8-P14 of Data ID through last user data word. Preset to all zeros, carry is ignored. 17 HMP8115 TABLE 6. OUTPUTTING THE SLICED WSS DATA AS BT.656 ANCILLARY DATA PIXEL OUTPUT P15 P14 P13 P12 P11 P10 P9 P8 Preamble 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Data ID P14 ep 1 1 0 0 1 0 =odd field data 1 =even field data Data Block Number P14 ep 0 0 0 0 0 1 Data Word Count P14 ep 0 0 0 0 1 0 WSS Data P14 ep 0 0 0 0 bit 13 bit 12 P14 ep 0 0 bit 11 bit 10 bit 9 bit 8 P14 ep 0 0 bit 7 bit 6 bit 5 bit 4 P14 ep 0 0 bit 3 bit 2 bit 1 bit 0 P14 ep 0 0 0 0 bit 5 bit 4 P14 ep 0 0 bit 3 bit 2 bit 1 bit 0 P14 ep 0 0 0 0 0 0 P14 ep 0 0 0 0 0 0 P14 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 WSS CRC Data CRC NOTES: 30. ep = even parity for P8-P13. 31. WSS CRC data = “00 0000” during PAL operation. 32. CRC = Sum of P8-P14 of Data ID through last user data word. Preset to all zeros, carry is ignored. TELETEXT If a teletext clock run-in is found before line 23 or line 289 for NTSC and (M) PAL, or line 336 for (B, D, G, H, I, N, NC) PAL, the VBI Teletext Detect status bit is immediately set to “1”. If not found by these lines, the status bit is immediately reset to “0”. The HMP8115 supports ITU-R BT.653 625-line and 525-line teletext system B, C and D capture. NABTS (North American Broadcast Teletext Specification) is the same as BT.653 525line system C, which is also used to transmit Intel Intercast™ information. WST (World System Teletext) is the same as BT.653 system B. Figure 18 shows the basic structure of a video signal that contains teletext data. Accessing the Teletext Data The teletext data must be output as BT.656 ancillary data. The I2C interface does not have the bandwidth to output teletext information when needed. The scan lines containing teletext information are monitored. If teletext is enabled and teletext data is present, the teletext data is output as BT.656 ancillary data. Table 7 illustrates the teletext BT.656 ancillary data format and Figure 17 depicts the portion of the incoming teletext signal which is sliced and output as part of the ancillary data stream. The teletext data is present during the horizontal blanking interval after the line containing the teletext information. The actual BT.656 bytes that contain teletext data only contain 4 bits of the actual data packet. Note that only the data packet of Figure 18 is sent as ancillary data; the clock run-in is not included in the data stream. Detection of Teletext The teletext decoder monitors the scan lines, looking for the 16-bit clock run-in (sometimes referred to as the clock synchronization code) used by teletext. If found, it locks to the clock run-in, the teletext data is sampled and loaded into shift registers, and the data is then transferred to internal holding registers. If the clock run-in is not found, it is assumed the scan line contains video data unless other VBI information is detected, such as WSS. Intercast™ is a trademark of Intel Corporation. 18 HMP8115 CLOCK RUN-IN DATA PACKET Bit 0 MSB NOTES: 33. The MSB is bit number: 271 for system C, 279 for system B 525-line and 343 for system B 625-line. 34. The clock run-in is 16 bits wide for both systems and is not included in the BT.656 ancillary data stream. 35. The bit rate is 5.727272 Mbits/second for system B and C on 525/60 systems and 6.9375 and 5.734375 Mbits/second respectively for 625/50 systems. FIGURE 18. TELETEXT VBI VIDEO SIGNAL TABLE 7. OUTPUTTING THE SLICED TELETEXT DATA AS BT.656 ANCILLARY DATA PIXEL INPUT P15 P14 P13 P12 P11 P10 P9 P8 Preamble 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Data ID P14 ep 1 1 0 1 0 0 Data Block Number P14 ep 0 0 0 0 0 1 Data Word Count P14 ep 0 1 0 1 1 0 Teletext Data (B, 625-line = 43 bytes) (B, 525-line = 35 bytes) (C = 34 bytes) P14 ep 0 = 525-line 1 = 625-line 0 = system B 1 = system C bit 343 bit 342 bit 341 bit 340 P14 ep 0 0 bit 339 bit 338 bit 337 bit 336 : Reserved CRC P14 ep 0 0 bit 7 bit 6 bit 5 bit 4 P14 ep 0 0 bit 3 bit 2 bit 1 bit 0 P14 ep 0 0 0 0 0 0 P14 ep 0 0 0 0 0 0 P14 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 NOTES: 36. ep = even parity for P8-P13. 37. CRC = Sum of P8-P14 of Data ID through last user data word. Preset to all zeros, carry is ignored. 38. For 525-line system B, bits 280-343 are “0”. 39. For system C, bits 272-343 are “0”. 19 HMP8115 REAL TIME CONTROL INTERFACE The PSW bit is always a “0” for NTSC encoding. During PAL encoding, it indicates the sign of V (“0” = negative; “1” = positive) for that scan line. The Real Time Control Interface (RTCI) outputs timing information for a NTSC/PAL encoder as BT.656 ancillary data. This allows the encoder to generate “clean” output video. RTCI information via BT.656 ancillary data is shown in Table 8. If enabled, this transfer occurs once per line and is completed before the start of the SAV sequence. TABLE 8. OUTPUTTING RTCI AS BT.656 ANCILLARY DATA PIXEL INPUT P15 P14 P13 P12 P11 P10 P9 P8 Preamble 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Data ID P14 ep 1 1 0 1 0 1 Data Block Number P14 ep 0 0 0 0 0 1 Data Word Count P14 ep 0 0 0 0 1 1 HPLL Increment P14 ep 0 0 0 0 0 0 P14 ep 0 0 0 0 0 0 P14 ep 0 0 0 0 0 0 P14 ep 0 0 0 0 0 0 P14 ep PSW 0 bit 31 bit 30 bit 29 bit 28 P14 ep F2 = 0 F1 = 0 bit 27 bit 26 bit 25 bit 24 FSCPLL Increment : CRC P14 ep 0 0 bit 7 bit 6 bit 5 bit 4 P14 ep 0 0 bit 3 bit 2 bit 1 bit 0 P14 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 NOTES: 40. ep = even parity for P8-P13. 41. CRC = Sum of P8-P14 of Data ID through last user data word. Preset to all zeros, carry is ignored. 20 HMP8115 During I2C write cycles, the first data byte after the slave address is treated as the control register sub address and is written into the internal address register. Any remaining data bytes sent during an I2C write cycle are written to the control registers, beginning with the register specified by the address register as given in the first byte. The address register is then autoincremented after each additional data byte sent on the I2C bus during a write cycle. Writes to reserved bits within registers or reserved registers are ignored. Host Interface All internal registers may be written to or read by the host processor at any time, except for those bits identified as read-only. The bit descriptions of the control registers are listed in Tables 9-48. The HMP8115 supports the fast-mode (up to 400 kbps) I2C interface consisting of the SDA and SCL pins. The device acts as a slave for receiving and transmitting data over the serial interface. When the interface is not active, SCL and SDA must be pulled high using external 4kΩ pull-up resistors. The slave address for the HMP8115 is 88H. In order to perform a read from a specific control register within the HMP8115, an I2C bus write must first be performed to properly setup the address register. Then an I2C bus read can be performed to read from the desired control register(s). As a result of needing the write cycle for a read cycle there are actually two START conditions as shown in Figure 21. The address register is then autoincremented after each byte read during the I2C read cycle. Reserved registers return a value of 00H . Data is placed on the SDA line when the SCL line is low and held stable when the SCL line is pulled high. Changing the state of the SDA line while SCL is high will be interpreted as either an I2C bus START or STOP condition as indicated by Figure 20. tSU:DATA tBUF SDA tHD:DATA SCL tLOW tHIGH tR tF tSU:STOP FIGURE 19. I2C TIMING DIAGRAM SDA SCL 8 1-7 S START CONDITION 9 R/W ADDRESS 1-7 ACK 8 DATA 9 P ACK STOP CONDITION FIGURE 20. I2C SERIAL DATA FLOW DATA WRITE 1000 1000 S CHIP ADDR FROM MASTER A SUB ADDR A 0x88 DATA READ S DATA A REGISTER POINTED TO BY SUB ADDR DATA A P FROM HMP8115 OPTIONAL FRAME MAY BE REPEATED n TIMES 1000 1000 (R/W) CHIP ADDR 0x88 A SUB ADDR A S CHIP ADDR A 0x89 DATA REGISTER POINTED TO BY SUB ADDR A DATA OPTIONAL FRAME MAY BE REPEATED n TIMES FIGURE 21. REGISTER WRITE/READ FLOW 21 NA P S = START CYCLE P = STOP CYCLE A = ACKNOWLEDGE NA = NO ACKNOWLEDGE HMP8115 HMP8115 Control Registers TABLE 9. 8115 REGISTER SUMMARY SUBADDRESS CONTROL REGISTER RESET/ DEFAULT VALUE USE VALUE COMMENTS 00H PRODUCT ID 15H 01H INPUT FORMAT 18H Defaults to Autodetect of input video standard 02H OUTPUT FORMAT 00H Defaults to 16-bit YCbCr format 03H OUTPUT CONTROL 00H Defaults to outputs disabled 04H GENLOCK CONTROL 01H Defaults to 27MHz CLK2, Rectangular Pixel mode 05H ANALOG INPUT CONTROL 00H Defaults to input signal select = NTSC/PAL1 06H COLOR PROCESSING 52H 07H RESERVED 08H LUMA PROCESSING 09H Reserved 0AH SLICED VBI DATA ENABLE 00H 0BH SLICED VBI DATA OUTPUT 00H 0CH VBI DATA STATUS 00H 0DH Reserved 0EH VIDEO STATUS 00H 0FH INTERRUPT MASK 00H 10H INTERRUPT STATUS 00H 11H-17H Reserved 04H - - - 18H BRIGHTNESS 00H 19H CONTRAST 80H 1AH HUE 00H 1BH SATURATION 80H 1CH COLOR GAIN 40H 1DH Reserved 1EH SHARPNESS 10H 1FH HOST CONTROL 00H 20H CLOSED CAPTION_ODD_A 80H 21H CLOSED CAPTION_ODD_B 80H 22H CLOSED CAPTION_EVEN_A 80H 23H CLOSED CAPTION_EVEN_B 80H 24H WSS_ODD_A 00H 25H WSS_ODD_B 00H 26H WSS_CRC_ODD 00H 27H WSS_EVEN_A 00H 28H WSS_EVEN_B 00H 29H WSS_CRC_EVEN 00H 2AH-2FH Reserved - - 22 HMP8115 TABLE 9. 8115 REGISTER SUMMARY (Continued) SUBADDRESS CONTROL REGISTER RESET/ DEFAULT VALUE USE VALUE 30H START H_BLANK LOW 4AH Table 2 31H START H_BLANK HIGH 03H Table 2 32H END H_BLANK 7AH Table 2 33H START V_BLANK LOW 02H Table 2 34H START V_BLANK HIGH 01H Table 2 35H END V_BLANK 12H Table 2 36H END HSYNC 40H Table 2 37H HSYNC DETECT WINDOW FFH 20H 38H-3FH Reserved - 40H-7FH Test and Unused - COMMENTS Recommend HSYNC DETECT WINDOW= 20H TABLE 10. PRODUCT ID REGISTER SUB ADDRESS = 00H BIT NO. 7-0 FUNCTION Product ID DESCRIPTION This 8-bit register specifies the last two digits of the product number. It is a read-only register. Data written to it is ignored. RESET STATE 15H TABLE 11. INPUT FORMAT REGISTER SUB ADDRESS = 01H BIT NO. 7 6-5 FUNCTION DESCRIPTION Reserved RESET STATE 0B Video Timing Standard These bits are read only unless D4 = “0”. 00 = (M) NTSC 01 = (B, D, G, H, I, N) PAL 10 = (M) PAL 11 = Combination (N) PAL; also called (NC) PAL 00B 4 Auto Detect Video Standard 0 = Manual selection of video timing standard 1 = Auto detect of video timing standard 1B 3 Setup Select Typically, this bit should be a “1” during (M) NTSC and (M, N) PAL operation. Otherwise, it should be a “0”. 0 = Video source has a 0 IRE blanking pedestal 1 = Video source has a 7.5 IRE blanking pedestal 1B 2-0 Reserved 000B 23 HMP8115 TABLE 12. OUTPUT FORMAT REGISTER SUB ADDRESS = 02H BIT NO. FUNCTION DESCRIPTION RESET STATE 7-5 Output Color Format 000 = 16-bit 4:2:2 YCbCr 001 = 8-bit 4:2:2 YCbCr 010 = 8-bit BT.656 011 = 15-bit RGB 100 = 16-bit RGB 101 = reserved 110 = reserved 111 = reserved 000B 4-3 RGB Gamma Select These bits are ignored except during RGB output modes. 00 = Linear RGB (gamma of input source = 2.2) 01 = Linear RGB (gamma of input source = 2.8) 10 = Gamma-corrected RGB (gamma = gamma of input source) 11 = reserved 00B 2-1 Output Color Select 00 = Normal operation 01 = Output blue field 10 = Output black field 11 = Output 75% color bars 00B Vertical Pixel Siting This bit specifies whether or not the chrominance pixels have a halfline pixel offset from their associated luminance pixels. 0 = Half-line offset 1 = Aligned 0B 0 TABLE 13. OUTPUT CONTROL REGISTER SUB ADDRESS = 03H BIT NO. FUNCTION DESCRIPTION RESET STATE 7 Video Data Output Enable This bit is used to enable the P0-P15 outputs. 0 = Outputs 3-stated 1 = Outputs enabled 0B 6 Video Timing Output Enable This bit is used to enable the HSYNC, VSYNC, BLANK, FIELD, VBIVALID, DVALID, and INTREQ outputs. 0 = Outputs 3-stated 1 = Outputs enabled 0B 5 FIELD Polarity 0 = Active low (low during odd fields) 1 = Active high (high during odd fields) 0B 4 Polarity BLANK 0 = Active low (low during blanking) 1 = Active high (high during blanking) 0B 3 HSYNC Polarity 0 = Active low (low during horizontal sync) 1 = Active high (high during horizontal sync) 0B 2 VSYNC Polarity 0 = Active low (low during vertical sync) 1 = Active high (high during vertical sync) 0B 1 DVALID Polarity 0 = Active low (low during valid pixel data) 1 = Active high (high during valid pixel data) 0B 0 VBIVALID Polarity 0 = Active low (low during VBI data) 1 = Active high (high during VBI data) 0B 24 HMP8115 TABLE 14. GENLOCK CONTROL REGISTER SUB ADDRESS = 04H BIT NO. FUNCTION DESCRIPTION RESET STATE 7 Aspect Ratio Mode 0 = Rectangular (BT.601) pixels 1 = Square pixels 0B 6 Freeze Output Timing Enable Setting this bit to a “1” freezes the output timing at the end of the field. Resetting this bit to a “0” resumes normal operation at the start of the next field. 0 = Normal operation 1 = Freeze output timing 0B 5 DVALID Duty Cycle Control (DVLD_DCYC) This bit is ignored during the 8-bit YCbCr and BT.656 output modes. During 16-bit YCbCr, 15-bit RGB, or 16-bit RGB output modes, this bit is defined as: 0 = DVALID has 50/50 duty cycle at the pixel output datarate 1 = DVALID goes active based on linelock. This will cause DVALID to not have a 50/50 duty cycle. This bit is intended to be used in maintaining backward compatibilty with the HMP8112A DVALID output timing. 0B 4 DVALID Line Timing Control (DVLD_LTC) During 16-bit YCbCr, 15-bit RGB, or 16-bit RGB output modes, this bit is defined as: 0 = DVALID present only during active video time on active scan lines 1 = DVALID present the entire scan line time on all scan lines During the 8-bit YCbCr and BT.656 output modes, this bit defines the DVALID output signal as: 0 = Normal timing 1 = DVALID signal ANDed with CLK2 0B 3 Missing HSYNC Detect Select This bit specifies the number of missing horizontal sync pulses before the device goes into the horizontal lock acquisition mode. In mode “0”, the default value of the HPLL Adjust register should be used. In mode “1”, the typical values the HPLL Adjust register should be 10H to 20H. 0 = 12 pulses 1 = 1 pulse 0B 2 Missing VSYNC Detect Select This bit specifies the number of missing vertical sync pulses before the device goes into the vertical lock acquisition mode. 0 = 3 pulses 1 = 1 pulse 0B 1-0 CLK2 Frequency This bit indicates the frequency of the CLK2 input clock. 00 = 24.54MHz 01 = 27.0MHz 10 = 29.5MHz 11 = Reserved 01B TABLE 15. ANALOG INPUT CONTROL REGISTER SUB ADDRESS = 05H BIT NO. FUNCTION 7-3 Reserved 2-0 Video Signal Input Select DESCRIPTION RESET STATE 00000B 000 = NTSC/PAL 1 001 = NTSC/PAL 2 010 = NTSC/PAL 3 011 = S-video 100 = reserved 101 = reserved 110 = reserved 111 = reserved 000B 25 HMP8115 TABLE 16. COLOR PROCESSING REGISTER SUB ADDRESS = 06H BIT NO. FUNCTION DESCRIPTION RESET STATE 7-6 Color Gain Control Select If a value of “10”, the color gain adjust register is used to specify the amount of color gain to be applied. 00 = No gain control (gain = 1x) 01 = Automatic gain control 10 = Fixed gain control 11 = Freeze automatic gain control 01B 5-4 Color Killer Select 00 = Force color on 01 = Enable color killer 10 = reserved 11 = Force color off 01B 3-2 Color Coring Select Coring may be used to reduce low-level noise around zero (code 128) in the CbCr signals. 00 = No coring 01 = 1 code coring 10 = 2 code coring 11 = 3 code coring 00B 1 Contrast Control Select This bit specifies whether the contrast control affects just the Y data (“0”) or both the Y and CbCr data (“1”). To avoid color shifts when changing contrast, this bit should be a “1”. 0 = Contrast controls only Y data 1 = Contrast controls Y and CbCr data 1B 0 Color Lowpass Filter Select This bit selects the bandwidth of the CbCr data. 0 = 850kHz 1 = 1.5MHz 0B TABLE 17. LUMA PROCESSING REGISTER SUB ADDRESS = 08H BIT NO. FUNCTION DESCRIPTION RESET STATE 7-6 Y Filtering Select The chroma trap filter may be used to remove any residual color subcarrier information from the Y channel. During S-video operation, it should be disabled. During PAL operation, it should be enabled. The 3MHz lowpass filter may be used to remove high-frequency noise. 00 = No filtering 01 = Enable chroma trap filter 10 = Enable 3.0MHz lowpass filter 11 = reserved 00B 5-4 Black Level Y Coring Select Coring may be used to reduce low-level noise around black in the Y signal. 00 = No coring 01 = 1 code coring 10 = 2 code coring 11 = 3 code coring 00B 3-2 High Frequency Y Coring Select Coring may be used to reduce high-frequency low-level noise in the Y signal. 00 = No coring 01 = 1 code coring 10 = 2 code coring 11 = 3 code coring 01B 1-0 Sharpness Frequency Select If a value of “01” or “10”, the sharpness adjust register is used to specify the amount of sharpness to be applied. 00 = Bypass sharpness control 01 = Maximum gain at 2.6MHz 10 = Maximum gain at color subcarrier frequency 11 = reserved 00B 26 HMP8115 TABLE 18. SLICED VBI DATA ENABLE REGISTER SUB ADDRESS = 0AH BIT NO. RESET STATE FUNCTION DESCRIPTION 7-6 Sliced Closed Captioning Enable 00 = Closed caption disabled 01 = Closed caption enabled for odd fields: line 21 for NTSC, line 18 for (M) PAL, or line 22 for (B, D, G, H, I, N, NC) PAL 10 = Closed caption enabled for even fields: line 284 for NTSC, line 281 for (M) PAL, or line 335 for (B, D, G, H, I, N, NC) PAL 11 = Closed caption enabled for both odd and even fields 00B 5-4 Sliced WSS Enable 00 = WSS disabled 01 = WSS enabled for odd fields: line 20 for NTSC; line 17 for (M) PAL, or line 23 for (B, D, G, H, I, N, NC) PAL 10 = WSS enabled for even fields: line 283 for NTSC, line 280 for (M) PAL, or line 336 for (B, D, G, H, I, N, NC) PAL 11 = WSS enabled for both odd and even fields 00B 3-2 Sliced Teletext Enable 00 = Teletext disabled 01 = Teletext system B enabled 10 = Teletext system C enabled 11 = reserved 00B 1-0 Reserved 00B TABLE 19. SLICED VBI DATA OUTPUT REGISTER SUB ADDRESS = 0BH BIT NO. FUNCTION DESCRIPTION RESET STATE 7 Sliced Closed Caption BT.656 Output Enable This bit specifies whether or not to output the caption data registers as BT.656 ancillary data. It is ignored unless captioning is enabled. Access via the I2C interface is always available. 0 = Do not output as BT.656 ancillary data 1 = Output as BT.656 ancillary data 0B 6 Sliced WSS BT.656 Output Enable This bit specifies whether or not to output the WSS data registers as BT.656 ancillary data. It is ignored unless WSS is enabled. Access via the I2C interface is always available. 0 = Do not output as BT.656 ancillary data 1 = Output as BT.656 ancillary data 0B 5 Sliced Teletext BT.656 Output Enable This bit specifies whether or not to output teletext data as BT.656 ancillary data. It is ignored unless teletext is enabled. 0 = Do not output as BT.656 ancillary 1 = Output as BT.656 ancillary data 0B 4-1 0 Reserved RTCI BT.656 Output Enable 0000B This bit specifies whether or not to output RTCI data as BT.656 ancillary data. 0 = Do not output as BT.656 ancillary 1 = Output as BT.656 ancillary data 27 0B HMP8115 TABLE 20. VBI DATA STATUS REGISTER SUB ADDRESS = 0CH BIT NO. FUNCTION DESCRIPTION RESET STATE 7 Closed Captioning Odd Field Detect Status This bit is read-only. Data written to this bit is ignored. 0 = Closed captioning not detected 1 = Closed captioning detected 0B 6 Closed Captioning Even Field Detect Status This bit is read-only. Data written to this bit is ignored. 0 = Closed captioning not detected 1 = Closed captioning detected 0B 5 WSS Odd Field Detect Status This bit is read-only. Data written to this bit is ignored. 0 = WSS not detected 1 = WSS detected 0B 4 WSS Even Field Detect Status This bit is read-only. Data written to this bit is ignored. 0 = WSS not detected 1 = WSS detected 0B 3 VBI Teletext Detect Status This bit is read-only. Data written to this bit is ignored. 0 = Teletext not detected during vertical blanking interval 1 = Teletext detected during vertical blanking interval 0B 2-0 Reserved 000B TABLE 21. VIDEO STATUS REGISTER SUB ADDRESS = 0EH BIT NO. FUNCTION DESCRIPTION RESET STATE 7 Vertical Lock Status This bit is read-only. Data written to this bit is ignored. 0 = Not vertically locked 1 = Vertically locked 0B 6 Horizontal Lock Status This bit is read-only. Data written to this bit is ignored. 0 = Not horizontally locked 1 = Horizontally locked 0B 5 Color Lock Status This bit is read-only. Data written to this bit is ignored. 0 = Not color locked 1 = Color locked 0B 4 Input Video Detect Status This bit is read-only. Data written to this bit is ignored. 0 = Input video not detected on selected video input 1 = Input video detected on selected video input 0B 3-0 Reserved 0000B 28 HMP8115 TABLE 22. INTERRUPT MASK REGISTER SUB ADDRESS = 0FH BIT NO. FUNCTION DESCRIPTION RESET STATE 7 Genlock Loss Interrupt Mask If this bit is a “1”, an interrupt is generated when genlock is lost. 0 = Interrupt disabled 1 = Interrupt enabled 0B 6 Input Signal Loss Interrupt Mask If this bit is a “1”, an interrupt is generated when a video signal is no longer detected on the selected video input. 0 = Interrupt disabled 1 = Interrupt enabled 0B 5 Closed Caption Interrupt Mask If this bit is a “1”, an interrupt is generated when the Caption_ODD_A and Caption_ODD_B or the Caption_EVEN_A and Caption_EVEN_B data registers contain new data. 0 = Interrupt disabled 1 = Interrupt enabled 0B 4 WSS Interrupt Mask If this bit is a “1”, an interrupt is generated when the WSS_ODD_A and WSS_ODD_B or the WSS_EVEN_A and WSS_EVEN_B data registers contain new data. 0 = Interrupt disabled 1 = Interrupt enabled 0B 3 Teletext Interrupt Mask If this bit is a “1”, an interrupt is generated when teletext information is first detected at the beginning of each field. 0 = Interrupt disabled 1 = Interrupt enabled 0B 2-1 0 Reserved Vertical Sync Interrupt Mask 00B If this bit is a “1”, an interrupt is generated at the beginning of each field. 0 = Interrupt disabled 1 = Interrupt enabled 0B TABLE 23. INTERRUPT STATUS REGISTER SUB ADDRESS = 10H BIT NO. FUNCTION DESCRIPTION RESET STATE 7 Genlock Loss Interrupt Status If this bit is a “1”, the reason for the interrupt request was that genlock was lost. To clear the interrupt request, a “1” must be written to this bit. 0B 6 Input Signal Loss Interrupt Status If this bit is a “1”, the reason for the interrupt request was that the input video source is no longer present. To clear the interrupt request, a “1” must be written to this bit. 0B 5 Closed Caption Interrupt Status If this bit is a “1”, the reason for the interrupt request was that the Caption_ODD_A and Caption_ODD_B or the Caption_EVEN_A and Caption_EVEN_B data registers contain new data. To clear the interrupt request, a “1” must be written to this bit. 0B 4 WSS Interrupt Status If this bit is a “1”, the reason for the interrupt request was that the WSS_ODD_A and WSS_ODD_B or the WSS_EVEN_A and WSS_EVEN_B data registers contain new data. To clear the interrupt request, a “1” must be written to this bit. 0B 3 Teletext Interrupt Status If this bit is a “1”, the reason for the interrupt request was that teletext data has been detected in the current field. To clear the interrupt request, a “1” must be written to this bit. 0B 2-1 0 Reserved Vertical Sync Interrupt Status 00B If this bit is a “1”, the reason for the interrupt request was that a new field was started. To clear the interrupt request, a “1” must be written to this bit. 29 0B HMP8115 TABLE 24. BRIGHTNESS REGISTER SUB ADDRESS = 18H BIT NO. FUNCTION 7 Reserved 6-0 Brightness Adjust DESCRIPTION RESET STATE 0B These bits control the brightness. They may have a value of +63 (“011 1111”) to -64 (“100 0000”), with positive values increasing brightness. A value of 0 (“000 0000”) has no effect on the data. 0000000B TABLE 25. CONTRAST REGISTER SUB ADDRESS = 19H BIT NO. 7-0 FUNCTION Contrast Adjust DESCRIPTION These bits control the contrast. They may have a value of 0x (“0000 0000”) to 1.992x (“1111 1111”). A value of 1x (“1000 0000”) has no effect on the data. RESET STATE 80H TABLE 26. HUE REGISTER SUB ADDRESS = 1AH BIT NO. 7-0 FUNCTION Hue Adjust DESCRIPTION These bits control the color hue. They may have a value of +30 degrees (“0111 1111”) to -30 degrees (“1111 1111”). A value of 0 degrees (“0000 0000”) has no effect on the color data. RESET STATE 00H TABLE 27. SATURATION REGISTER SUB ADDRESS = 1BH BIT NO. 7-0 FUNCTION Saturation Adjust DESCRIPTION These bits control the color saturation. They may have a value of 0x (“0000 0000”) to 1.992x (“1111 1111”). A value of 1x (“1000 0000”) has no effect on the color data. A value of 0x (“0000 0000”) disables the color information. RESET STATE 80H TABLE 28. COLOR GAIN REGISTER SUB ADDRESS = 1CH BIT NO. 7-0 FUNCTION Color Gain Adjust DESCRIPTION These bits control the amount of gain control for the color difference (CbCr) signals. They may have a value of 0.5x (“0010 0000”) to 3.98x (“1111 1111”). A value of 1x (“0100 0000”) has no effect on the data. This register is ignored unless the color gain control mode selection is “fixed gain control”. RESET STATE 40H TABLE 29. SHARPNESS REGISTER SUB ADDRESS = 1EH BIT NO. FUNCTION 7-6 Reserved 5-0 Sharpness Adjust DESCRIPTION RESET STATE 00B These bits control the amount of gain control of high frequency luminance signals (either 2.6MHz or Fsc). They may have a value of +12dB (“11 1111”) to -12dB (“00 0100”). A value of 0dB (“01 0000”) has no effect on the data. This register is ignored if the sharpness mode selection is “disable sharpness control” or “reserved”. 30 010000B HMP8115 TABLE 30. HOST CONTROL REGISTER SUB ADDRESS = 1FH BIT NO. FUNCTION DESCRIPTION When this bit is set to 1, the entire device except the I2C bus is reset to a known state exactly like the RESET input going active. The software reset will initialize all register bits to their reset state. Once set this bit is self clearing after only 4 CLK periods. This bit is cleared on power-up by the external RESET pin. RESET STATE 7 Software Reset 6 Reserved 5 Closed Caption Odd Field Read Status This bit is read-only. Data written to this bit is ignored. The bit is cleared when the caption data has been read out via the I2C interface or as BT.656 ancillary data. 0 = No new caption data 1 = Caption_ODD_A and Caption_ODD_B data registers contain new data. 0B 4 Closed Caption Even Field Read Status This bit is read-only. Data written to this bit is ignored. The bit is cleared when the caption data has been read out via the I2C interface or as BT.656 ancillary data. 0 = No new caption data 1 = Caption_EVEN_A and Caption_EVEN_B data registers contain new data. 0B 3 WSS Odd Field Read Status This bit is read-only. Data written to this bit is ignored. The bit is cleared when the WSS data has been read out via the I2C interface or as BT.656 ancillary data. 0 = No new WSS data 1 = WSS_ODD_A and WSS_ODD_B data registers contain new data. 0B 2 WSS Even Field Read Status This bit is read-only. Data written to this bit is ignored. The bit is cleared when the WSS data has been read out via the I2C interface or as BT.656 ancillary data. 0 = No new WSS data 1 = WSS_EVEN_A and WSS_EVEN_B data registers contain new data. 0B 1-0 0B 0B Reserved 00B TABLE 31. CLOSED CAPTION_ODD_A DATA REGISTER SUB ADDRESS = 20H BIT NO. 7-0 FUNCTION Odd Field Caption Data DESCRIPTION If odd field captioning is enabled and present, this register is loaded with the first eight bits of caption data on line 18, 21, or 22. Bit 0 corresponds to the first bit of caption information. Data written to this register is ignored. RESET STATE 80H TABLE 32. CLOSED CAPTION_ODD_B DATA REGISTER SUB ADDRESS = 21H BIT NO. 15-8 FUNCTION Odd Field Caption Data DESCRIPTION If odd field captioning is enabled and present, this register is loaded with the second eight bits of caption data on line 18, 21, or 22. Data written to this register is ignored. RESET STATE 80H TABLE 33. CLOSED CAPTION_EVEN_A DATA REGISTER SUB ADDRESS = 22H BIT NO. 7-0 FUNCTION Even Field Caption Data DESCRIPTION If even field captioning is enabled and present, this register is loaded with the first eight bits of caption data on line 281, 284, or 335. Bit 0 corresponds to the first bit of caption information. Data written to this register is ignored. 31 RESET STATE 80H HMP8115 TABLE 34. CLOSED CAPTION_EVEN_B DATA REGISTER SUB ADDRESS = 23H BIT NO. 15-8 FUNCTION Even Field Caption Data DESCRIPTION If even field captioning is enabled and present, this register is loaded with the second eight bits of caption data on line 281, 284, or 335. Data written to this register is ignored. RESET STATE 80H TABLE 35. WSS_ODD_A DATA REGISTER SUB ADDRESS = 24H BIT NO. 7-0 FUNCTION Odd Field WSS Data DESCRIPTION If odd field WSS is enabled and present, this register is loaded with the first eight bits of WSS information on line 17, 20, or 23. Bit 0 corresponds to the first bit of WSS information. Data written to this register is ignored. RESET STATE 00H TABLE 36. WSS_ODD_B DATA REGISTER SUB ADDRESS = 25H BIT NO. FUNCTION 15-14 Reserved 13-8 Odd Field WSS Data DESCRIPTION RESET STATE 00B If odd field WSS is enabled and present, this register is loaded with the second six bits of WSS information on line 17, 20, or 23. Data written to this register is ignored. 000000B TABLE 37. WSS_CRC_ODD DATA REGISTER SUB ADDRESS = 26H BIT NO. FUNCTION 7-6 Reserved 5-0 Odd Field WSS CRC Data DESCRIPTION RESET STATE 00B If odd field WSS is enabled and present during NTSC operation, this register is loaded with the six bits of CRC information on line 20. It is always a “000000” during PAL operation. Data written to this register is ignored. 000000B TABLE 38. WSS_EVEN_A DATA REGISTER SUB ADDRESS = 27H BIT NO. 7-0 FUNCTION Even Field WSS Data DESCRIPTION If even field WSS is enabled and present, this register is loaded with the first eight bits of WSS information on line 280, 283, or 336. Bit 0 corresponds to the first bit of WSS information. Data written to this register is ignored. RESET STATE 00H TABLE 39. WSS_EVEN_B DATA REGISTER SUB ADDRESS = 28H BIT NO. FUNCTION 15-14 Reserved 13-8 Even Field WSS Data DESCRIPTION RESET STATE 00B If even field WSS is enabled and present, this register is loaded with the second six bits of WSS information on line 280, 283, or 336. Data written to this register is ignored. 32 000000B HMP8115 TABLE 40. WSS_CRC_EVEN DATA REGISTER SUB ADDRESS = 29H BIT NO. FUNCTION 7-6 Reserved 5-0 Even Field WSS CRC Data DESCRIPTION RESET STATE 00B If even field WSS is enabled and present during NTSC operation, this register is loaded with the six bits of CRC information on line 283. It is always a “000000” during PAL operation. Data written to this register is ignored. 000000B TABLE 41. START H_BLANK LOW REGISTER SUB ADDRESS = 30H BIT NO. 7-0 FUNCTION Assert BLANK Output Signal DESCRIPTION This 8-bit register is cascaded with Start H_Blank High Register to form a 10-bit start_horizontal_blank REGISTER. It specifies the horizontal count (in 1x clock cycles) at which to assert BLANK each scan line. Bit 0 is always a “0”, so the start of horizontal blanking may only be done with two pixel resolution. The leading edge of HSYNC is count 000H. RESET STATE 4AH TABLE 42. START H_BLANK HIGH REGISTER SUB ADDRESS = 31H BIT NO. 15-10 9-8 FUNCTION DESCRIPTION Reserved Assert BLANK Output Signal RESET STATE 000000B This 2-bit register is cascaded with Start H_Blank Low Register to form a 10-bit start_horizontal_blank register. It specifies the horizontal count (in 1x clock cycles) at which to assert BLANK each scan line. The leading edge of HSYNC is count 000H. 11B TABLE 43. END H_BLANK REGISTER SUB ADDRESS = 32H BIT NO. 7-0 FUNCTION Negate BLANK Output Signal DESCRIPTION This 8-bit register specifies the horizontal count (in 1x clock cycles) at which to negate BLANKeach scan line. Bit 0 is always a “0”, so the end of horizontal blanking may only be done with two pixel resolution. The leading edge of HSYNC is count 000H. RESET STATE 7AH TABLE 44. START V_BLANK LOW REGISTER SUB ADDRESS = 33H BIT NO. 7-0 FUNCTION Assert BLANK Output Signal DESCRIPTION This 8-bit register is cascaded with Start V_Blank High Register to form a 9-bit start_vertical_blank register. It specifies the line number to assert BLANK each field. For NTSC operation, it occurs on line (n + 5) on odd fields and line (n + 268) on even fields. For PAL operation, it occurs on line (n + 5) on odd fields and line (n + 318) on even fields. 33 RESET STATE 02H HMP8115 TABLE 45. START V_BLANK HIGH REGISTER SUB ADDRESS = 34H BIT NO. 15-9 8 FUNCTION DESCRIPTION Reserved Assert BLANK Output Signal RESET STATE 0000000B This 1-bit register is cascaded with Start V_Blank Low Register to form a 9-bit start_vertical_blank register. 1B TABLE 46. END V_BLANK REGISTER SUB ADDRESS = 35H BIT NO. 7-0 FUNCTION Negate BLANK Output Signal DESCRIPTION This 8-bit register specifies the line number to negate BLANK each field. RESET STATE 12H For NTSC operation, it occurs on line (n + 5) on odd fields and line (n + 268) on even fields. For PAL operation, it occurs on line (n + 5) on odd fields and line (n + 318) on even fields. TABLE 47. END HSYNC REGISTER SUB ADDRESS = 36H BIT NO. 7-0 FUNCTION Negate HSYNC Output Signal DESCRIPTION This 8-bit register specifies the horizontal count at which to negate HSYNC each scan line. Values may range from 0 (0000 0000) to 510 (1111 1111) CLK2 cycles. The leading edge of HSYNC is count 00H. RESET STATE 40H TABLE 48. HSYNC DETECT WINDOW REGISTER SUB ADDRESS = 37H BIT NO. 7-0 FUNCTION Horizontal Sync Detect Window DESCRIPTION This 8-bit register specifies the width of the window (in 1x clock samples) to look for horizontal sync pulses each line. The window is centered about where the horizontal sync pulse should be located. If the horizontal sync pulse falls inside this window, the digital PLL will lock to it. If the horizontal sync pulse falls outside this window, the digital PLL is immediately reset to have the same timing. Recommend using a value of 20H to optimize the response time of the digital PLL. 34 RESET STATE FFH HMP8115 Pinout GND VCC DEC_T LAGC_CAP LCAP VCC NC NC GND HSYNC VSYNC GND VCC FIELD DVALID BLANK 80 LEAD PQFP TOP VIEW 80 79 78 77 76 75 74 73 72 71 70 69 68 67 66 65 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 64 63 62 61 60 59 58 57 56 55 54 53 52 51 50 49 48 47 46 45 44 43 42 41 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 GND VCC WPE GAIN_CNTL CCAP DEC_L VCC NC GND RESET GND GND VCC CLK2 GND SDA AGND VAA AGND NC NTSC/PAL3 NTSC/PAL2 NTSC/PAL1 YIN YOUT AGND AGND VAA CLK2 VAA AGND AGND A/D_TEST NC C NC AGND AGND AGND AGND 35 P15 P14 GND VBIVALID P13 VCC P12 P11 P10 P9 P8 GND VCC P7 P6 P5 P4 P3 GND P2 INTREQ P1 P0 SCL HMP8115 Pin Description PIN NAME PIN NUMBER INPUT/ OUTPUT P0-P15 42, 43, 45, 47-51, 54-58, 60, 63, 64 O Pixel output pins. See Table 3. HSYNC 71 O Horizontal sync output. HSYNC is asserted during the horizontal sync intervals. The polarity of HSYNC is programmable. This pin is three-stated after a RESET or software reset and should be pulled high through a 10K resistor. VSYNC 70 O Vertical sync output. VSYNC is asserted during the vertical sync intervals. The polarity of VSYNC is programmable. This pin is three-stated after a RESET or software reset and should be pulled high through a 10K resistor. FIELD 67 O FIELD output. The polarity of FIELD is programmable. This pin is three-stated after a RESET or software reset and should be pulled high through a 10K resistor. BLANK 65 O Composite blanking output. BLANK is asserted during the horizontal and vertical blanking intervals. The polarity of is programmable. This pin is three-stated after a RESET or software reset and should be pulled high through a 10K resistor. DVALID 66 O Data valid output. DVALID is asserted during CLK2 cycles that contain valid pixel data. This pin is three-stated after a RESET or software reset and should be pulled high through a 10K resistor. CLK2 38, 13 I 2x pixel clock inputs. All CLK2 pins must be connected together. This clock must be a continuous, free-running clock. RESET 34 I Reset control input. A logical zero for a minimum of four CLK2 cycles resets the device. RESET must be a logical one for normal operation. SDA 40 I/O SCL 41 I I2C interface clock input. WPE 27 I White Peak Enable. When enabled (“1”), the video gain is reduced when the A/D output code exceeds 248. When disabled (“0”), the video amplifier will clip when the A/D output code reaches code 255. VBIVALID 61 O Vertical Blanking Interval Valid output. VBIVALID is asserted during CLK2 cycles that contain valid VBI (Vertical Blanking Interval) data such as Closed Captioning, Teletext, and Wide Screen Signalling data. The polarity of VBIVALID is programmable. This pin is three-stated after a RESET or software reset and should be pulled high through a 10K resistor. INTREQ 44 O Interrupt Request Output. This is an open-drain output and requires an external 10K pull-up resistor to VCC. NTSC/PAL 1 7 I Composite Video Input. This input must be AC-coupled to the video signal (using a 1µF capacitor) and terminated with a 75Ω resistor, as shown in the Applications section. These components should be as close to this pin as possible for best performance. If not used, this pin should be connected to AGND through a 0.1µF capacitor. NTSC/PAL 2 6 I Composite Video Input. This input must be AC-coupled to the video signal (using a 1µF capacitor) and terminated with a 75Ω resistor, as shown in the Applications section. These components should be as close to this pin as possible for best performance. If not used, this pin should be connected to AGND through a 0.1µF capacitor. NTSC/PAL 3 (Y) 5 I Composite video or Luminance (Y) video input. This input must be AC-coupled to the video signal (using a 1µF capacitor) and terminated with a 75Ω resistor, as shown in the Applications section. These components should be as close to this pin as possible for best performance. If not used, this pin should be connected to AGND through a 0.1µF capacitor. DESCRIPTION I2C interface data input/output. 36 HMP8115 Pin Description (Continued) PIN NAME PIN NUMBER INPUT/ OUTPUT C 19 I Chrominance (C) video input. This input must be AC-coupled to the video signal (using a 1µF capacitor) and terminated with a 75Ω resistor, as shown in the Applications section. These components, and the corresponding anti-aliasing lowpass filter, should be as close to this pin as possible for best performance. If not used, this pin should be connected to AGND through a 0.1µF capacitor. YOUT 9 O Analog output of the video multiplexer. This output should be lowpass filtered and input via the YIN pin, as shown in the Applications section. The anti-aliasing lowpass filter should be as close to YOUT and YIN as possible for best performance. YIN 8 I Analog input to the ADC. GAIN_CTRL 28 I Gain Control Input. A DC voltage is used to set the video amplifier’s gain, as shown in Figure 2. The reference circuit should be as close to this pin as possible for best performance. DEC_T 78 I Decoupling for A/D Converter Reference. A 0.1µF capacitor should be connected between this pin and AGND, as shown in the Applications section. This capacitor should be as close to this pin as possible for best performance. DEC_L 30 I Decoupling for A/D Converter Reference. A 0.1µF capacitor should be connected between this pin and AGND, as shown in the Applications section. This capacitor should be as close to this pin as possible for best performance. LAGC_CAP 77 I Capacitor connection for Luminance AGC Circuit. Controls the AGC loop time constant. A 0.01µF capacitor should be connected between this pin and AGND, as shown in the Applications section. This capacitor should be as close to this pin as possible for best performance. LCAP 76 I Capacitor connection for Luminance Clamp Circuit. Controls the clamp loop time constant. A 0.047µF capacitor should be connected between this pin and AGND, as shown in the Applications section. This capacitor should be as close to this pin as possible for best performance. CCAP 29 I Capacitor connection for Chrominance Clamp Circuit. Controls the clamp loop time constant. A 0.047µF capacitor should be connected between this pin and AGND, as shown in the Applications section. This capacitor should be as close to this pin as possible for best performance. VCC 26, 31,37, 52, 59, 68, 75, 79 I Digital power supply pins. All VCC pins must be connected together. GND 25, 33, 35, 36, 39, 46, 53, 62, 69, 72, 80 I Digital ground pins. All GND pins must be connected together. VAA 2, 12,14 I Analog power supply pins. All VAA pins must be connected together. AGND 1, 3, 10, 11, 15,16, 21, 22, 23, 24 I Analog ground pins. All AGND pins must be connected together. A/D TEST 17 O A/D test pin. This pin must be left floating for proper operation. NC 4, 18, 20, 32, 73, 74 DESCRIPTION No Connect pins. These pins must be left floating for proper operation. 37 HMP8115 Applications Information PCB LAYOUT CONSIDERATIONS Analog Signals A PCB board with a minimum of 4 layers is recommended, with layers 1 and 4 (top and bottom) for signals and layers 2 and 3 for power and ground. The PCB layout should implement the lowest possible noise on the power and ground planes by providing excellent decoupling. Traces containing digital signals should not be routed over, under, or adjacent to the analog output traces to minimize crosstalk. If this is not possible, coupling can be minimized by routing the digital signals at a 90 degree angle to the analog signals. The analog input traces should also not overlay the VAA power plane to maximize high-frequency power supply rejection. The optimum layout places the HMP8115 as close as possible to the power supply connector and the video input connector. EVALUATION BOARD Component Placement HMPVIDEVAL/ISA External components should be positioned as close as possible to the appropriate pin, ideally such that traces can be connected point to point. Chip capacitors are recommended where possible, with radial lead ceramic capacitors the second-best choice. The HMPVIDEVAL/ISA evaluation board allows connecting the HMP8115 into a PC ISA slot for evaluation. It includes the HMP8115 NTSC/PAL decoder, 3MB of VRAM, and a NTSC/PAL encoder. The board accepts composite or S-video input and displays video on a standard TV. The ISA bus and evaluation software allow easy performance evaluation of the HMP8115 using tools such as the Tektronix VM700 video test system. Power supply decoupling should be done using a 0.1µF ceramic capacitor in parallel with a 0.01µF chip capacitor for each group of VAA and VCC pins to ground. These capacitors should be located as close to the power and ground pins as possible, using short, wide traces. RELATED APPLICATION NOTES Digital Ground Plane Application Notes are also available on the Intersil Multimedia web site at http://www.semi.harris.com/mmedia. All GND pins on the HMP8115 should be connected to the digital ground plane of the board. AN9644: Composite Video Separation Techniques AN9716: Widescreen Signalling (WSS) Analog Ground Plane AN9717: YCbCr to RGB Considerations A separate analog ground plane for the HMP8115 is recommended. All AGND pins on the HMP8115 should be connected to the analog ground plane. This analog ground plane should be connected to the board’s digital ground plane at a single point. AN9728: BT.656 Video Interface for ICs AN9738: Video Module Interface (VMI) for ICs Analog Power Plane The HMP8115 should have its own VAA power plane that is isolated from the common power plane of the board, with a gap between the two power planes of at least 1/8 inch. All VAA pins on the HMP8115 must be connected to this analog power plane. The analog power plane should be connected to the board’s normal VCC power plane at a single point though a low-resistance ferrite bead, such as a Ferroxcube 5659065-3B, Fair-Rite 2743001111, or TDK BF45-4001. The ferrite bead provides resistance to switching currents, improving the performance of HMP8115. A single 47µF capacitor should also be used between the analog power plane and the ground plane to control low-frequency power supply ripple. If a separate linear regulator is used to provide power to the analog power plane, the power-up sequence should be designed to ensure latchup will not occur. A separate linear regulator is recommended if the power supply noise on the VAA pins exceeds 200mV. 38 HMP8115 U1 NTSC-PAL1 C1 NTSC-PAL2 C2 NTSC-PAL3/Y C3 7 1.0µF 6 1.0µF NTSC/PAL 1 NTSC/PAL 2 5 NTSC/PAL 3 1.0µF R3 75 R2 75 R1 75 ANTI-ALIAS FILTER CHROMA R4 75 C4 19 1.0µF ANTI-ALIAS FILTER 8 9 77 C5 0.01µF C 76 C6 0.047µF 29 C7 0.047µF YIN YOUT LAGC_CAP LCAP CCAP P[7..0] P7 P6 P5 P4 P3 P2 P1 P0 P15 P14 P13 P12 P11 P10 P9 P8 BLANK DVALID FIELD HSYNC VSYNC VBIVALID INTREQ 78 C8 0.1µF C9 0.01µF C10 0.1µF 30 DEC_T DEC_L C11 0.01µF VAA R5 1K R6 750 28 RESET SDA SCL CLK2 CLK2 TEST 51 50 49 48 47 45 43 42 64 63 60 58 57 56 55 54 VCC R16 4K BLANK HSYNC VSYNC VBIVALID 44 34 40 41 38 13 36 27 R17 4K DVALID FIELD INTREQ RESET 27MHz VAA R7 10K JP1 JUMPER FIGURE 22. HMP8115 REFERENCE SCHEMATICS 39 VCC VCC RP1 10K 65 66 67 71 70 61 GAIN_CNTL WPE P[15..8] R8 50 C12 15pF SDA SCL 27MHz HMP8115 Absolute Maximum Ratings Thermal Information Digital Supply Voltage (VCC to GND) . . . . . . . . . . . . . . . . . . . . 7.0V Analog Supply Voltage (VAA to GND) . . . . . . . . . . . . . . . . . . . . 7.0V Digital Input Voltages . . . . . . . . . . . . . . . GND - 0.5V to VCC + 0.5V ESD Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Class 1 Thermal Resistance (Typical, See Note 42) θJA (oC/W) PQFP Package . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Maximum Power Dissipation HMP8115CN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.9W Maximum Storage Temperature Range . . . . . . . . . .-65oC to 150oC Maximum Junction Temperatures . . . . . . . . . . . . . . . . . . . . . 150oC Maximum Lead Temperature (Soldering 10s) . . . . . . . . . . . . 300oC Operating Temperature Range HMP8115CN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0oC to 70oC CAUTION: Stresses above those listed in “Absolute Maximum Ratings” may cause permanent damage to the device. This is a stress only rating and operation of the device at these or any other conditions above those indicated in the operational sections of this specification is not implied. NOTE: 42. θJA is measured with the component mounted on an evaluation PC board in free air. Dissipation rating assumes device is mounted with all leads soldered to printed circuit board. Electrical Specifications VCC = VAA = 5.0V, TA = 25oC PARAMETER SYMBOL TEST CONDITION MIN TYP MAX UNITS 4.75 5 5.25 V - 280 315 mA - 105 115 mA - 175 200 mA - 1.47 1.66 W POWER SUPPLY CHARACTERISTICS Power Supply Voltage Range VCC, VAA Total Power Supply Current ITOT Digital Power Supply Current ICC Analog Power Supply Current IAA Total Power Dissipation PTOT (Note 43) CLK2 = 29.5MHz, VCC = VAA = 5.25V Outputs Not Loaded CLK2 = 29.5MHz, VCC = VAA = 5.25V, Outputs Not Loaded DC CHARACTERISTICS: DIGITAL I/O (EXCEPT CLK2 and I2C INTERFACE) Input Logic High Voltage VIH VCC = Max 2.0 - - V Input Logic Low Voltage VIL VCC = Min - - 0.8 V 2.4 - - V Output Logic High Voltage VOH IOH = -4mA, VCC = Max Output Logic Low Voltage VOL IOL = 4mA, VCC = Min - - 0.4 V VCC = Max Input = 0V or 5V - - 10 µA f = 1MHz, (Note 43) All Measurements Referenced to Ground, TA = 25oC - 8 - pF - - 10 µA Input Leakage Current Input/Output Capacitance Three-State Output Current Leakage IIH, IIL CIN, COUT IOZ DC CHARACTERISTICS: CLK2 DIGITAL INPUT Input Logic High Voltage VIH VCC = Max 0.7xVCC - - V Input Logic Low Voltage VIL VCC = Min - - 0.3xVCC V Input Leakage Current IIH VCC = Max Input = 0V or VCC - - 10 µA - 450 - - µA - 8 - pF IIL Input Capacitance CIN CLK2 = 1MHz, (Note 43) All Measurements Referenced to Ground, TA = 25oC 40 HMP8115 Electrical Specifications PARAMETER VCC = VAA = 5.0V, TA = 25oC (Continued) SYMBOL TEST CONDITION MIN TYP MAX UNITS DC CHARACTERISTICS: I2C INTERFACE Input Logic High Voltage VIH VCC = Max 0.7xVCC - - V Input Logic Low Voltage VIL VCC = Min - - 0.3xVCC V 3.0 - - V Output Logic High Voltage VOH IOH = -1mA, VCC = Max Output Logic Low Voltage VOL IOL = 3mA, VCC = Min 0 - 0.4 V VCC = Max Input = 0V or 5V - - 10 µA SCL = 400kHz, (Note 43) All Measurements Referenced to GND, TA = 25oC - 8 - pF 20 - 29.5 MHz 40 - 60 % Input Leakage Current Input/Output Capacitance IIH, IIL CIN, COUT AC CHARACTERISTICS: DIGITAL I/O (EXCEPT I2C INTERFACE) CLK2 Frequency CLK2 Waveform Symmetry (Note 43) CLK2 Pulse Width High tPWH 13 - - ns CLK2 Pulse Width Low tPWL 13 - - ns 10 - - ns (Note 44) Data and Control Setup Time tSU Data and Control Hold Time tHD 0 - - ns tDVLD 0 5 8 ns - 2 6 ns CLK2 to Output Delay tr, tf Data and Control Rise/Fall Time (Note 43) AC CHARACTERISTICS: I2C INTERFACE SCL Clock Frequency fSCL 0 - 400 kHz SCL Pulse Width Low tLOW 1.3 - - µs SCL Pulse Width High tHIGH 0.6 - - µs Data Hold Time tHD:DATA 0 - - ns Data Setup Time tSU:DATA 100 - - ns - - 300 ns - - 300 ns SDA, SCL Rise Time tR SDA, SCL Fall Time tF (Note43) ANALOG INPUT PERFORMANCE Composite Video Input Amplitude (Sync Tip to White Level) Input Termination of 75Ω and 1.0µF AC-Coupled 0.5 1.0 2.0 VP-P Luminance (Y) Video Input Amplitude (Sync Tip to White Level) Input Termination of 75Ω and 1.0µF AC-Coupled 0.5 1.0 2.0 VP-P Chrominance (C) Video Input Amplitude (Burst Amplitude) Input Termination of 75Ω and 1.0µF AC-Coupled, (Note 43) 0.143 0.286 0.6 VP-P 200 - - kΩ Video Input Impedance RAIN Note 43 41 HMP8115 Electrical Specifications PARAMETER Video Input Bandwidth ADC Input Range VCC = VAA = 5.0V, TA = 25oC (Continued) SYMBOL MIN TYP MAX UNITS 5 - - MHz AIN FULL SCALE - 1 - VP-P AIN OFFSET - 1.5 - V - 2 - LSB - 0.35 - LSB - 2 - % - 1 - Deg. - 2 - Deg. - 2 - % NTC-7 Composite (Note 43) - 2 - % SNRL WEIGHTED Pedestal Input (Note 43) - 50 - dB tLOCK Time from Initial Lock Acquisition to an Error of 1 Pixel. (Note 43) 2 3 - Fields - 5 % 1 or 12 1 or 12 1 or 12 Hsyncs 1 or 3 1 or 3 1 or 3 Vsyncs BW ADC Integral Nonlinearity INL ADC Differential Nonlinearity DNL TEST CONDITION 1VP-P Sine Wave Input to -3dBc Reduction, (Note 43) Best Fit Linearity VIDEO PERFORMANCE Differential Gain DG Differential Phase DP Hue Accuracy Modulated Ramp (Note 43) 75% Color Bars (Note 43) Color Saturation Accuracy Luminance Nonlinearity SNR GENLOCK PERFORMANCE Horizontal Locking Time Long-Term horizontal Sync Lock Range Range over specified pixel jitter is maintained. Assumes line time changes by amount indicated slowly between over one field. (Note 43) Number of Missing Horizontal Syncs Before Lost Lock Declared HSYNC LOST Number of Missing Vertical Syncs Before Lost Lock Declared VSYNC LOST Programmable via register 04H (Note 43) - Long-Term Color Subcarrier Lock Range Range over color subcarrier locking time and accuracy specifications are maintained. Subcarrier frequency changes by amount indicated slowly over 24 hours. (Note 43) - ±200 ±400 Hz Vertical Sample Alignment (Notes 43, 45) - 1/8 - Pixel - 10 - ns NOTES: 43. Guaranteed by design or characterization. 44. Test performed with CL = 40pF, IOL = 4mA, IOH = -4mA. Input reference level is 1.5V for all inputs. VIH = 3.0V, VIL = 0V. 45. This should not be confused with Clock Jitter, since the HMP8115 does not generate the sample clock. Thus, clock jitter is solely dependent on the source of the CLK2 signal. The Vertical Sample Alignment parameter specifies how accurately samples align vertically from one scan line to the next. 42 HMP8115 Metric Plastic Quad Flatpack Packages (MQFP/PQFP) Q80.14x20 (JEDEC MO-108CB-1 ISSUE A) D 80 LEAD METRIC PLASTIC QUAD FLATPACK PACKAGE D1 -D- -B- -AE E1 e PIN 1 SEATING A PLANE -H- 0.10 0.004 0.40 0.016 MIN -C- 5o-16o 0.20 A-B S 0.008 M C 0o MIN A2 A1 0o-7o L 5o-16o INCHES MILLIMETERS SYMBOL MIN MAX MIN MAX NOTES A - 0.134 - 3.40 - A1 0.010 - 0.25 - - A2 0.100 0.120 2.55 3.05 - B 0.012 0.018 0.30 0.45 6 B1 0.012 0.016 0.30 0.40 - D 0.904 0.923 22.95 23.45 3 D1 0.783 0.791 19.90 20.10 4, 5 E 0.667 0.687 16.95 17.45 3 E1 0.547 0.555 13.90 14.10 4, 5 L 0.026 0.037 0.65 0.95 N 80 80 7 e 0.032 BSC 0.80 BSC - ND 24 24 - NE 16 16 Rev. 0 1/94 NOTES: D S 1. Controlling dimension: MILLIMETER. Converted inch dimensions are not necessarily exact. B 2. All dimensions and tolerances per ANSI Y14.5M-1982. B1 3. Dimensions D and E to be determined at seating plane -C- . 4. Dimensions D1 and E1 to be determined at datum plane -H- . 0.13/0.17 0.005/0.007 5. Dimensions D1 and E1 do not include mold protrusion. Allowable protrusion is 0.25mm (0.010 inch) per side. BASE METAL WITH PLATING 6. Dimension B does not include dambar protrusion. Allowable dambar protrusion shall be 0.08mm (0.003 inch) total. 0.13/0.23 0.005/0.009 7. “N” is the number of terminal positions. All Intersil semiconductor products are manufactured, assembled and tested under ISO9000 quality systems certification. Intersil products are sold by description only. Intersil Corporation reserves the right to make changes in circuit design and/or specifications at any time without notice. Accordingly, the reader is cautioned to verify that data sheets are current before placing orders. Information furnished by Intersil is believed to be accurate and reliable. However, no responsibility is assumed by Intersil or its subsidiaries for its use; nor for any infringements of patents or other rights of third parties which may result from its use. No license is granted by implication or otherwise under any patent or patent rights of Intersil or its subsidiaries. For information regarding Intersil Corporation and its products, see web site http://www.intersil.com Sales Office Headquarters NORTH AMERICA Intersil Corporation P. O. Box 883, Mail Stop 53-204 Melbourne, FL 32902 TEL: (407) 724-7000 FAX: (407) 724-7240 EUROPE Intersil SA Mercure Center 100, Rue de la Fusee 1130 Brussels, Belgium TEL: (32) 2.724.2111 FAX: (32) 2.724.22.05 43 ASIA Intersil (Taiwan) Ltd. Taiwan Limited 7F-6, No. 101 Fu Hsing North Road Taipei, Taiwan Republic of China TEL: (886) 2 2716 9310 FAX: (886) 2 2715 3029