S5D0127X01 Data Sheet MULTIMEDIA VIDEO MULTISTANDARD VIDEO DECODER/SCALER The S5D0127X01 converts analog NTSC, PAL or SECAM video in composite, S-video, or component format to digitized component video. Output data can be selected for CCIR 601 or square pixel sample rates in either YCbCr or RGB formats. The digital video can be scaled down in both the horizontal and vertical directions. The S5D0127X01 also decodes Intercast, Teletext, Closed Caption, and WSS data with a built-in bit data slicer. Digitized CVBS data can be output directly during VBI for external processing. 100 PQFP FEATURES • Accepts NTSC-M/N/4.43, PAL-M/N/B/G/H/I/D/K/L and SECAM formats with auto detection • 6 analog inputs: 3 S-video, 6 composite, or 1 3-wire YCbCr component video ORDERING INFORMATION • 2-line luma and chroma comb filters including adaptive luma comb for NTSC Device Package • Programmable luma bandwidth, contrast, brightness, and edge enhancement S5D0127X01Q0R0 100 PQFP Temperature Range -20°~+70°C • Programmable chroma bandwidth, hue, and saturation • High quality horizontal and vertical down scaler • Intercast, Teletext and Closed Caption decoding with built-in bit slicer • Digital Video • Direct output of digitized CVBS during VBI for Intercast application • Video Capture/Editing • Analog square pixel or CCIR 601 sample rates RELATED PRODUCTS • Output in 4:4:4, 4:2:2, or 4:1:1 YCbCr component, or 24-bit or 16-bit RGB formats with dithering • S5D0123X01 ENCODER MULTISTANDARD VIDEO • YCbCr 4:2:2 output can be 8 or 16 bits wide with embedded timing reference code support for 8-bit mode • Simultaneous scaled and non-scaled digital output ports outputs for 8-bit mode. • Direct access to scaler via bi-directional digital port. • Programmable Gamma correction table • Programmable timing signals • Industry standard IIC interface APPLICATIONS • Multimedia Modified on May/04/2000 ELECTRONICS PAGE 1 OF 96 ELECTRONICS Modified on May/04/2000 SCLK SDAT AEX1,0 AY[2:0] AC[2:0] COMP2 VRT VRB HS1 VS XTALI XTALO IIC Host Interface Gain/Offset Control 6 input MUX 2-AGCs 2-ADCs Reference Generator/ Compensation EXV Direct. Control Digital Input Timing VBI Data Decoding Chroma Processing Luma Processing Comb Filter Sliced VBI or Raw CVBS Variable LPF Vertical Scaler SRAM Gamma LUT Color Space Converter Horizontal Scaler Non Scaler Output Scaler Input OR Simultaneous Non Scaler Output OR 24-bit Output Line Lock, Timing Generation Data Output Format CCEN CCDAT OEN C[7:0] Y[7:0] EXV[7:0] EHAV EVAV HS1 HS2 VS HAV VAV CK CK2 SCH PID S5D0127X01 Data Sheet MULTIMEDIA VIDEO BLOCK DIAGRAM PAGE 2 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO PIN ASSIGNMENT - 100 PQFP NCP Y7 Y6 Y5 Y4 NCP PORTA TESTEN VSS VDD EXV3 VDD3 VSS VSS EXV5 EXV4 AEX0 EXV6 VDD3 VRT HS2(IIC) SCLK CCEN CCDAT SDAT EXV7 AEX1 NCP NCP Y3 Y2 Y1 Y0 C7 VDD3 VDD3 VSS VSS C6 C5 S5D0127X01 C1 C0 NCP NCP 50 49 48 47 46 45 44 43 42 41 40 39 38 37 36 35 34 33 32 31 EXV1 EXV2 NCP NCP CK2 ODD VS SCH(PORTB) HAV HS1 VSS OEN EXV0 PID CK VSS VDD VDD3 VDD3 VSS VSS XTALO XTALI VDDA1 RST C4 C3 C2 NCP NCP VAV(OENC0) EVAV(OENC1) EHAV 81 NCP 82 NCP 83 VSS 84 AY0 85 VDDA 86 AY1 87 VSS 88 AY2 89 VDDA 90 AC0 91 VSS 92 AC1 93 VDDA 94 AC2 95 VSS 96 TEST 97 COMP2 98 VDDA 99 NCP 100 NCP NCP VRB NCP 80 79 78 77 76 75 74 73 72 71 70 69 68 67 66 65 64 63 62 61 60 59 58 57 56 55 54 53 52 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Modified on May/04/2000 ELECTRONICS PAGE 3 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO PIN DESCRIPTION Pin Name Pin # Type Description INPUT AY0 84 I 1 of 6 analog CVBS or 1of 3 S-video Y inputs. AY1 86 I 1 of 6 analog CVBS or 1of 3 S-video Y inputs. AY2 88 I 1 of 6 analog CVBS input or 1 of 3 S-video Y inputs or Y input for 3 wire component input AC0 90 I 1 of 6 analog CVBS or 1 of 3 S-video C inputs. AC1 92 I 1 of 6 analog CVBS or 1 of 3 S-video C inputs or Cb input for 3 wire component input AC2 94 I 1 of 6 analog CVBS or 1 of 3 S-video C inputs or Cr input for 3 wire component input XTALI 8 I Pin 1 for an external crystal or TTL clock input. XTALO 7 O Pin 2 for an external crystal. RST 10 I Chip reset. Active low signal. OUTPUT (All output pins can be selectively three-stated) Y0 - Y7, C0 - C7 45-48,53-56,3339,44 O Digital video outputs. EXV0 - EXV7 16,27,28,61-63, 68,71 I/O Expanded digital video I/O port. Can be configured as an additional 8-bit output port (no scaling), or additional outputs of the main digital output stream for 24 bit output modes, as an 8-bit input for direct digital access of the down scaler. HS1 26 I/O Programmable horizontal timing signal. One pulse every video line. When the EXV port is configured as an input, this pin can be programmed as an input. HS2(IIC) 76 I/O Programmable horizontal timing signal. One pulse every video line. At power up, this pin needs a 10 kΩ pull-down resistor to configure the chip to operate in IIC mode. VS 23 I/O Programmable vertical timing signal. When the EXV port is configured as an input, this pin can be programmed as an input. HAV 25 O Programmable horizontal active video flag. VAV(OENC0) 3 I/O Programmable vertical active video flag. During reset, the pin is an input and the logic state of this pin is latched into the OENC[0] register bit. Use a 10 kΩ resistor for pull-up or pull-down. EHAV 5 O Valid pixel data flag. Polarity is programmable. Active when output video data is valid. Modified on May/04/2000 ELECTRONICS PAGE 4 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO PIN DESCRIPTION (Continued) Pin Name Pin # Type Description EVAV(OENC1) 4 I/O Valid line flag. Polarity is programmable. Active when output video line is valid. During reset, the pin is an input and the logic state of this pin is latched into the OENC[1]register bit. Use a 10 kΩ resistor for pull-up or pull-down. ODD 22 O Odd field flag. Polarity is programmable. Active for fields 1 and 3. PID 17 O PAL ID flag. High for phase alternating line. OEN 15 I Digital video data, timing and clock output 3-state control. CK 18 I/O Pixel clock. In normal decoding mode, this is an output. When the EXV port is used as an input, this can be programmed as an input pixel clock. CK2 21 O Pixel output clock (rate is one half of CK) aligned to HAV signal. CCDAT 73 O Sliced VBI data output. Data can be from Closed Caption, Teletext, Intercast, or WSS type encoded data. CCEN 74 O When high, this pin indicates that valid VBI data is being clocked out at the CCDAT pin or at the digital video output. MULTI-PURPOSE I/O PORTS AND TEST ENABLE PORTA 58 I/O Multi-purpose I/O port. SCH(PORTB) 24 I/O Multi-purpose I/O port. TESTEN 57 I When tied to VDD, the chip is put into the test mode. For normal use, this pin should be connected to VSS. TEST 96 I When tied to VDD, the chip is put into the test mode. For normal use, this pin should be connected to VSS. REFERENCE AND COMPENSATION VRT 77 I/O ADC VRT compensation (requires an external 0.1 µF capacitor connected to VSS). VRB 78 I/O ADC VRB compensation (requires an external 0.1 µF capacitor connected to VSS). COMP2 97 I/O Internal 1.3 V reference (requires an external 0.1 µF capacitor connected to VSS). HOST INTERFACE SCLK 75 I Serial clock for IIC host interface. SDAT 72 I/O Serial data for IIC host interface. AEX0 - AEX1 69 - 70 I Device ID selection for IIC host interface. Modified on May/04/2000 ELECTRONICS PAGE 5 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO PIN DESCRIPTION (Continued) Pin Name Pin # Type Description POWER AND GROUND VDD 20,59 PWR Digital power supply for output buffers. The voltage can be +5V or 3.3V depending on interface requirement. VDD3 11,12,42,43,66, 67 +3.3V Digital power supply for internal logic. VDDA 85,89,93,98 +5V Analog power supply for ADC, AGC and reference circuits. VDDA1 9 +5V Analog power supply for clock generation circuitry. VSS 6,13,14,19,40, 41,60,64,65,83, 87,91,95 GND Common ground. NC NCP 1,2,29-32,49-52, 79-82,99,100 - These pins are directly connected to the die substrate. If electrical connect is desired (not required) only connection to VSS is allowed. Modified on May/04/2000 ELECTRONICS PAGE 6 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO PIN CROSS REFERENCE: NUMERICAL ORDER BY PIN NUMBER Pin # Pin Name Pin # Pin Name Pin # Pin Name Pin # Pin Name 1 NCP 26 HS1 51 NCP 76 HS2(IIC) 2 NCP 27 EXV1 52 NCP 77 VRT 3 VAV(OENC0) 28 EXV2 53 Y4 78 VRB 4 EVAV(OENC1) 29 NCP 54 Y5 79 NCP 5 EHAV 30 NCP 55 Y6 80 NCP 6 VSS 31 NCP 56 Y7 81 NCP 7 XTALO 32 NCP 57 TESTEN 82 NCP 8 XTALI 33 C0 58 PORTA 83 VSS 9 VDDA1 34 C1 59 VDD 84 AY0 10 RST 35 C2 60 VSS 85 VDDA 11 VDD3 36 C3 61 EXV3 86 AY1 12 VDD3 37 C4 62 EXV4 87 VSS 13 VSS 38 C5 63 EXV5 88 AY2 14 VSS 39 C6 64 VSS 89 VDDA 15 OEN 40 VSS 65 VSS 90 AC0 16 EXV0 41 VSS 66 VDD3 91 VSS 17 PID 42 VDD3 67 VDD3 92 AC1 18 CK 43 VDD3 68 EXV6 93 VDDA 19 VSS 44 C7 69 AEX0 94 AC2 20 VDD 45 Y0 70 AEX1 95 VSS 21 CK2 46 Y1 71 EXV7 96 TEST 22 ODD 47 Y2 72 SDAT 97 COMP2 23 VS 48 Y3 73 CCDAT 98 VDDA 24 SCH(PORTB) 49 NCP 74 CCEN 99 NCP 25 HAV 50 NCP 75 SCLK 100 NCP Modified on May/04/2000 ELECTRONICS PAGE 7 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO 1. FUNCTIONAL DESCRIPTION 1.1. VIDEO INPUT The S5D0127X01 supports complete video decoding of many analog video standards. In addition, the chip can support direct 8-bit YCbCr input for high quality video scaling and other processing. 1.1.1. Analog Video Input Figure 1 shows the detailed block diagram of the analog front end. Up to six composite video sources, three S-video sources, one 3-wire YCbCr component video source, or any combination can be selected. The allowed inputs are selected using the INSEL[3:0] bits in the CMDB register. Table 1 lists all possible input selections. The front end has two paths each containing an analog gain control, a clamping control, and an 8-bit ADC. Composite video input uses only the luma path. S-video and analog component YCbCr inputs utilize both the luma and chroma paths. The ADC digital data is used to calculate the correct gain and clamp values. The data is feedback to the analog clamping and gain control. This architecture eliminates any offset and gain mismatch in the analog front end. OFFSET LPF AC1 MUX AC0 To Chroma Processing 8 Bit ADC AGC AC2 VRT COMP1 VRB AY0 To Luma Processing MUX AY1 8 Bit ADC AGC AY2 LPF 8-bit Digital Input To Timing Generation GAIN OFFSET Figure 1. Analog Front End The analog inputs must be AC coupled through an external 0.1 µF capacitor clamp. Due to the high sampling rate of the ADC’s inside the S5D0127X01, most video sources will not require a low-pass filter for alias reduction. For those video sources with harmonics above 13 MHz, a simple single order pole at 6 MHz will provide sufficient high frequency signal reduction. This can be implemented with a 400 pF capacitor in parallel with the 75 Ω load. Analog Video S5D0127X01 0.1 µF 75 Ω Figure 2. Typical Analog Video Input Modified on May/04/2000 ELECTRONICS PAGE 8 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Table 1: Analog Video Input selections INSEL[3:0](hex) Selected Input(s) Video Type 0 AY0 Composite 1 AY1 Composite 2 AY2 Composite 4 AC0 Composite 5 AC1 Composite 6 AC2 Composite 8 AY0, AC0 S-Video 9 AY1, AC1 S-Video A AY2, AC2 S-Video F AY2(Y), AC1(Cb), AC2(Cr) YCbCr component video 1.1.2. Digital AGC Control The AGC normally references to the ADC code difference between sync tip and back porch. Two sets of sync tip-back porch ADC values are available for different AGC gain requirements: if AGCGN = 0, the sync tip locks to code 2, and the back porch locks to code 70; when AGCGN = 1, the sync tip locks to 16, and the back porch locks to code 70. Video signal with abnormal sync tip or very bright saturated colors may cause the ADC to limit the maximum value. This situation can be corrected by enabling the AGCOVF bit in the CMDB register to force the gain tracking loop to reduce AGC when maximum limiting conditions occur. The AGC may also be programmed to freeze the AGC at the current value by setting the AGCFRZ bit in the CMDB register. Once the AGC is frozen, the gain can be manually adjusted with the AGC register. The tracking time constant for the AGC can be controlled with the AGC_LPG[1:0] bits in the TRACKB register. In addition, the AGC tracking time constant can be configured as 2X faster during acquisition via the AGC_LKG. 1.1.3. Digital Video Input The high quality digital video down scaler in the S5D0127X01 can be directly accessed via the EXV bi-directional port. The S5D0127X01 accepts CCIR 656 compliant 8-bit YCbCr digital video input with embedded or external timing. Video timing may also be generated by the S5D0127X01. Data path for 8-bit YCbCr input is shown in Figure 3. Selection of analog video input or digital CCIR 656 data is with the INPSL[1:0] register bits. The S5D0127X01 can operate in master or slave timing mode when the chip is programmed for digital video input. 1.1.4. Pixel Clock and Timing Mode Selection for Digital Video Input Pixel clock and synchronization timing can be individually selected to either come from an external generator or be generated internally. In addition, if synchronization is provided by an external source, the S5D0127X01 supports embedded syncs as defined in CCIR 656, or TTL HS and VS inputs. Selection of pixel clock is via CKDIR bit in Modified on May/04/2000 ELECTRONICS PAGE 9 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO CMDD register. Timing selection is through either SYNDIR or EAV bit. HS1 VS Timing Control Y EXV[7:0] Data Demux Luma Processing To Luma Scaler Chroma Processing To Chroma Scaler C From Luma ADC From Chroma ADC Figure 3. 8-bit YCbCr Input Data Path By using an external pixel clock, the reference clock input at XTALI is no longer required. Additional register bits have to be programmed for different selections of pixel clock and timing, which are detailed in Table 2. The following register/bit-settings are required for digital video input: INSEL[3:0] = 8, 9, A, or F. TSTCGN = 1. DMCTL[1:0] = 2 or 3. UGAIN = 238. BRT = 34. SAT = 229. RGBH = UNIT = PED = 1. Table 2: Digital Video Input Pixel Clock and Timing Selection Pixel Clock TTL Timing Embedded Timing Additional Register Programming CKDIR*1 SYNDIR*2 EAV*3 VMEN 0 0 0 1 0 1 3 0 0 1 0 1 1 3 0 1 0 0 1 1 1 1 0 0 1 0 1 3 1 0 1 0 1 1 1 1 1 0 0 1 1 1 TSTGPH TSTGEN TSTGFR PIXSEL 0 if input data is at square pixel rate. 1 if input is at CCIR 601 rate. MNFMT IFMT 1 0 if input is 50 Hz video. 1 if input is 60 Hz video. 1 1 1 1 1 *1 : CKDIR = 0 - CK is output and is internally generated. CKDIR = 1 - CK is input from an external source. : SYNDIR = 0 - HS1 and VS are output. SYNDIR = 1 - HS1 and VS are inputs from external sources. *3 : EAV = 0 - chip will not sync to embedded timing. EAV = 1 - chip will sync to embedded timing. Note: the combination X11 for CKDIR, SYNDIR, EAV is not valid. *2 Modified on May/04/2000 ELECTRONICS PAGE 10 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO When in digital input mode, all programmable timing registers (such as HAVB,HAVE, HS2B etc.) are still functional. If HS1 and VS are programmed as inputs, the associated output timing controls such as HS1B,E will have no effect. An example of horizontal timing for digital input is shown in Figure 4. HS1 Programmable, when an output - Any input phase is acceptable This HS1 location can also come From a 656 SAV code Constant to internal counter reference EXV[7:0] 80 10 80 10 80 10 U0 Y0 V0 Y1 U2 Y2 V2 Y3 U4 Y4 V4 Y5 U6 Y6 V6 Y7 Ux Yx Vx Yx Ux Yx Vx Yx Vx Data group delay through chip -- Y[7:0] 80 10 80 10 80 10 U0 Y0 V0 Y1 U2 Y2 V2 Y3 U4 Y4 V4 Y5 U6 Y6 V6 Y7 Ux Yx Vx Yx Ux Y output for OFMT=2 is shown, any 8 or 16 bit output format is allowed. Fully programmable HAVB location based on internal counter HAV -- fully programmable, Defines location of first, last pixel and defines Cb,Y,Cr data location Fully programmable HAVE location HAV CK CK2 CK can be input or output The CK2 output clock phasing is aligned to the HAV leading edge Figure 4. Horizontal Timing for EXV Port as Digital Input 1.1.5. Additional Information for Analog Component Video Input For the S5D0127X01 to correctly set the V component phase in analog component video input mode, PID (pin 17) and PORTA (pin 58) need to be connected together. PORTA has to be configured as input (DIRA = 0) and connected to the internal CBG signal (DATAA[2:0] = 3). It is also recommended that external clamp circuit be used for Cb and Cr inputs (before the coupling caps) and the internal chroma clamp be disabled (COFFENB = 1) due to slight Cb/Cr leakage. Modified on May/04/2000 ELECTRONICS PAGE 11 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO 1.2. VIDEO TRACKING AND TIMING GENERATION When the S5D0127X01 is configured for analog video input, the chip tracks the video input and generates a sampling clock that is line locked to the input video. The S5D0127X01 requires an external reference clock for video tracking. This reference can be supplied via a crystal using the on chip crystal interface or any TTL compatible source. These configurations are shown in Figure 5 1.2.1. Clock Input Timing Reference The S5D0127X01 can use either a 24.576 MHz or a 26.8 MHz reference. However, it is recommended that the 24.576 MHz reference be used for CCIR 601 operation, and the 26.8 MHz reference be used for square pixel or dual mode operation. Other specifications for the crystal are: • Fundamental or third overtone • Load capacitance of ~20 pF • Series resistance of 40 Ω or less • Frequency deviation of 50 ppm or less over operating temperature range 22 pF 8 22 pF XTALI 8 24.576 MHz TTL Clock S5D0127X01 S5D0127X01 24.576 MHz 7 XTALI N. C. XTALO 7 XTALO 5.7 µH Optional for 3rd harmonic crystal 391 pF Using a Crystal Figure 5. Using a Clock Standard Clock Configurations 1.2.2. The Sampling Clock The sampling clock is generated by multiplying the line rate by N. This ensures that samples are aligned horizontally, vertically and in time. The required N factor for the S5D0127X01 is based upon the field rate (60 Hz or 50 Hz) and the desired sampling rates (CCIR 601 or square pixel). Field rate can be automatically detected and can be monitored with the FFRDET bit in the STAT register. Manual control of the field rate can be controlled with the MNFMT and IFMT bits. The PIXSEL bit in register CMDA selects CCIR 601 or square pixel. Table 3 shows the constants for the various combinations of input formats and output pixel rates. Modified on May/04/2000 ELECTRONICS PAGE 12 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Table 3: Timing for Different Pixel Rates CCIR 601 Data Rates Square Pixel Data Rates Units M N,B,G,H,I,D,K,K1,L M N,B,G,H,I,D,K,K1,L Field Rate 60 50 60 50 Hz Pixels/Line (N) 858 864 780 944 Pixels Active Pixels/Line 720 720 640 768 Pixels Active Lines/Frame 480 580 480 580 Lines Pixel Rate 13.5 13.5 12.27 14.75 MHz 27 27 24.54 29.5 MHz ADC Sampling Rate The time constants for the pixel clock tracking loop can be adjusted with the HFSEL[1:0] bits. In addition to providing the pixel clock, the S5D0127X01 also outputs various timing signals to indicate the beginning of a line, a field, and for field and frame identification. All the timing and clock pins may be optionally put into high impedance state. Three-state of these pins are software controlled and initial state of these pins at power up is controlled via two configuration pins: 3 and 4. The S5D0127X01 can generate all the video timing without video input. This enables the S5D0127X01 to be used as a video timing generator for a system that contains both the S5D0127X01 for live video input and a MPEG decoder which requires a video timing generator. 1.2.3. Horizontal Timing The S5D0127X01 creates many internal timing signals aligned to the horizontal sync tip (mid-way of the falling edge of horizontal sync, typically ADC code 36). These include locations of color burst (CBG, CBGW) used in chrominance processing, back porch (BPG), and sync tip timing signals (SLICE, FS_PULSE) used for AGC and clamp functions. SLICE is low whenever the input is below half way level of horizontal sync (typically ADC code 36). FS_PULSE is a single clock pulse coincide with the start of SLICE. One of these internal signals can be made available at the PORTA or PORTB pin at any time. The chip outputs two horizontal synchronization signals: HS1 and HS2. The start and stop locations for these signals are fully programmable. Offset programmed to HSxB, HSxE, and HSxBE0 are added to the default edge locations as shown in Table 4. Note that there are different modulo numbers for different input video standards and output pixel rates. Modified on May/04/2000 ELECTRONICS PAGE 13 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Table 4: Horizontal Timing Signal Edge Locations (in # of CK) 60 Hz Description Signal Chip delay 50 Hz CCIR 601 Square Pixel CCIR 601 Square Pixel (modulo 1716) (modulo 1560) (modulo 1728) (modulo 1888) 120 120 120 120 Sync gate (1-CK pulse) SYG 72 72 72 72 Back porch gate BPG [147 222] [129 204] [154 234] [168 254] Color burst gate (1-CK pulse) CBG 222 204 234 254 CBGW [159 254] [147 233] [173 254] [186 277] Two pulses per line (1-CK each pulse) FH2 42, 900 42, 822 42, 906 42, 986 Default one pulse per line HS1 [65 238] [45 220] [69 250] [65 270] Default one pulse per line HS2 [65 238] [45 220] [69 250] [65 270] Default horizontal cropping HAV [351 75] [334 58] [379 91] [415 59] Wide color burst gate An additional signal, HAV, is provided for horizontal video cropping. This signal has programmable polarity, start and stop locations. Two 11-bit registers, HAVB and HAVE, are used to define the first and last pixel locations of the horizontal portion of the cropped video. Numbers programmed into these registers are used as offset to the default locations as shown in Table 4. Note that even though HAVB and HAVE have 1-CK resolution, the difference between them should be maintained at multiple of 4 CKs for correct output. Table 4 shows the default edge locations relative to the midway of the falling edge of the analog horizontal sync. Note the numbers shown are in multiple of CK clocks. Figure 6 shows the approximate locations for the horizontal timing signals. Horizontal timing signals used for scaling will be described in Section 1.6.1. Modified on May/04/2000 ELECTRONICS PAGE 14 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Analog video input Digital video output Active video Blank Active video Chip delay HAVE HAVB HAV HSE HSB HS1,2 FS_PULSE SLICE SYG BPG CBG CBGW FH2 Figure 6. Approximate Locations for the Horizontal Timing Signals 1.2.4. Vertical Timing The vertical timing signals include VS, VAV, ODD, SCH, and PID. The VS is used for identifying the first line of video in the vertical position. The VS leading edge can be programmed to either track the incoming video’s serration pulses or to be aligned to the beginning of the video line or half way, as shown in Figure 36 and Figure 37. If VALIGN = 0, the VS leading edge is based on the output of an internal low pass filter, and its location is dependent on the noise conditions of the video input. The trailing edge of VS is locked to either the beginning of the video line or half way. The half way location relative to the beginning of the video line changes depending on current input standard and output format. If VALIGN = 1, the leading edge of the VS is aligned to the beginning of the video line or half way. The trailing edge is always aligned to the beginning of the video line. The VSE bit in the CMDA register can be programmed to shorten the VS falling edge by one horizontal line. The VAV signal is used for vertical cropping. The start and stop lines for VAV are programmable through the VAVB and VAVE registers, respectively. The ODD signal signifies the current field number. When ODD is active, the current field is 1 or 3 (or 5 or 7 if in PAL mode). The leading and trailing edges of ODD can be aligned to either the leading edge of VS (VALIGN = 1) or the trailing edge of VS (VALIGN = 0). The signal may be used in conjunction with SCH and PID to exactly identify the Modified on May/04/2000 ELECTRONICS PAGE 15 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO current field. To distinguish between fields 1, 2 verse fields 3, 4 (or fields 1, 2, 3, 4 verse fields 5, 6, 7, 8 for PAL) the phase of the color burst relative to the sync tip must be measured. That information is provided by the SCH pin. The S5D0127X01 provides the output of a comparator that measures whether the current color burst phase relative to the falling edge of the sync is greater or less than a predetermined constant. This constant is controlled with SCHCMP[3:0]. The polarity of the SCH output pin depends on the current SCHCMP[3:0] value. The SCH signal changes every video line. The SCH for line 260 is held for the entire vertical blanking period. By using the SCH signal for the same line from each field, proper field identification can be determined. Figure 8 shows field identification values for SCHCMP[3:0]=0. It is important to note that the SCH value is only valid for video signals that have a constant sync tip to color burst relationship. This is not the case with consumer VCRs. . HS1 (default) ODD FIELD and VALIGN = 0 VS 60 Hz - CCIR 601 = 885, Square = 708 50 Hz - CCIR 601 = 891, Square = 971 0, except 60 Hz Square = 2 ODD EVEN FIELD or VALIGN = 1 27, except 60 Hz Square = 28 VS 0, except 60 Hz Square = 2 ODD 15, except 60 Hz Square = 14 VAV HAV default width for each input standard and output mode 60 Hz - 261 50 Hz - CCIR 601 = 273, Square = 341 EVAV Note: Numbers shown are in CK. Active high polarities are used. Timing shown for VAV and EVAV are with qualifier off. Figure 7. Short Term Vertical Timing Modified on May/04/2000 ELECTRONICS PAGE 16 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO VS ODD 2 1 3 4 SCH Truth Table FIELD 1 2 3 4 ODD H L H L SCH H L L H Figure 8. Note: ODD and SCH are measured at the trailing edge of VS. NTSC Vertical Timing Signals The PID pin is used to identify whether the current V-axis is inverted in PAL mode. This signal changes at the color burst. By noting this value at the same line of each field, a determination of whether a field is from {1-4} or {5-8} can be made. As with the SCH pin, the S5D0127X01 is designed to hold the line 260 PID measurement for the entire vertical blank period. This allows easy sampling of the PID or current field identification. The ODD, SCH and PID signals change at different times and more than once within the video fields. Proper data for field identification is determined by latching all three signals at the trailing edge of VS. Figure 9 shows the VS, ODD, SCH, and PID signals and their latched values for each of the 8 possible fields. Figure 10 is the line to line timing diagram for these signals in PAL mode. VS 2 1 ODD 3 6 5 4 7 8 SCH PID Truth Table FIELD 1 2 3 4 5 6 7 8 ODD H L H L H L H L SCH H L H L L H L L PID H L L H H L L H Figure 9. PAL Vertical Timing Signals Modified on May/04/2000 ELECTRONICS Note: ODD, SCH and PID are measured at the trailing edge of VS (VALIGN = 0). PAGE 17 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO SCH ODD PID VS HS Figure 10. Line to Line VS, SCH and PID Timing (PAL input) 1.3. HORIZONTAL LUMA PROCESSING A simplified block diagram for the luma path is shown in Figure 11. Contrast Control Decimation Filter Horizontal Peaking Chroma Trap CONT HYBWR HYBWI HYPK CTRAP Programmable Low Pass Filter Brightness Control FROM ADC Figure 11. HYLPF BRT Horizontal Luma Processing Unit 1.3.1. Luminance DC Gain The S5D0127X01 can accommodate CCIR 624 M/N/H/G standards, which fall into categories of -40 or -43 sync tip and inclusion or exclusion of 7.5 setup. The S5D0127X01 can produce correct CCIR 601 luminance output levels by controlling the gain and offset in the luminance path via PED. This register should be set for the appropriate input standard. The programmable CONT and BRT registers provide the user with additional flexibility to create non-standard luminance gain and offset values. Modified on May/04/2000 ELECTRONICS PAGE 18 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Max Input Peak White Black Level Blanking Level Sync Tip Figure 12. Luminance Signal Luminance levels produced by the S5D0127X01 for different broadcast standards (assuming AGCGN=0, CONT=0 and BRT=0) are summarized in Table 5. Table 5: Luminance Digital Level Code M/N PED=1 M/N PED=0 B/G/H PED=1 Level (IRE) ADC (CVBS) Y[7:0] Level (IRE) ADC (CVBS) Y[7:0] Level (IRE) ADC (CVBS) Y[7:0] Max Input 109 255 255 109 255 255 117 255 255 Peak White 100 240 235 100 240 235 100 229 235 Black 7.5 83 16 0 70 16 0 70 16 Blank 0 70 1 0 70 16 0 70 Sync -40 2 1 -40 2 1 -43 2 Signal KS0127B Data Path Equation C Y = 1.37CVBS – 100 C Y = 1.288CVBS – 74 CY = 1.37CVBS – 80 When digital component output is desired in RGB mode, the RGBH register can be programmed to increase the 0-100% values from standard CCIR 601 levels to full range levels.The gain variations are shown in Table 6. Table 6: RGB Output Range RGB normal gain (RGBH=0) RGB high gain (RGBH=1) Signal Cy RGB (U,V=0) Cy RGB (U,V=0) Peak White 235 235 255 255 Black 16 16 0 0 Modified on May/04/2000 ELECTRONICS PAGE 19 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO For CCIR 601 digital video input (INPSL[1:0] = 1), register UNIT must be set to 1 to produce unit gain. 1.3.2. Horizontal Luma Frequency Shaping The luma path contains many programmable filters for different purposes. The combination of these filters will give different frequency characteristics. The over sampled video data from the ADC pass through a decimation filter. The decimation filter has user programmable bandwidth. Three registers are used to control the decimation filter characteristics and each is designed for certain purposes. The HYBWI, when set to “1”, provides extra bandwidth for very high quality video source. The HYBWR, when set to “1”, reduces the bandwidth so high frequency noise can be eliminated. The 3-bit register HYLPF[2:0] provides the necessary bandwidth reduction for horizontal scaling. When all three registers are programmed to “0”, the decimation filter has the bandwidth of the normal video. The S5D0127X01 provides the option of bypassing the decimation filter. This option should be used only when the input video is band limited and with low high frequency noise. For composite video input, the notch filter can be enabled (CTRAP set to “1”) to extract the luminance. The notch filter has different center frequencies for different input video format. User selectable peaking function is included for edge enhancement. The notch filter should be bypassed for S-video and component video input, or if luma comb filter is enabled. The luminance filter characteristics have been designed to be very similar for all combinations of 60/50 Hz video and CCIR 601/square pixel sampling rates. Figure 13 and Figure 14 show the output characteristics of the luminance path with different filter combinations for the supported input standards and output pixel rates. Modified on May/04/2000 ELECTRONICS PAGE 20 OF 96 S5D0127X01 Data Sheet Figure 13. Figure 14. MULTIMEDIA VIDEO Medium to High Frequency Luma Filter Characteristics (CTRAP=0) Medium to Low Frequency Luma Filter Characteristics (NTSC, CTRAP=1) Modified on May/04/2000 ELECTRONICS PAGE 21 OF 96 S5D0127X01 Data Sheet Figure 15. MULTIMEDIA VIDEO Medium to Low Frequency Luma Filter Characteristic (PAL, CTRAP=1) Figure 16. Luma Filter Characteristic with Peaking On (NTSC, CTRAP=1) Modified on May/04/2000 ELECTRONICS PAGE 22 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO 1.4. HORIZONTAL CHROMA PROCESSING A simplified block diagram for the horizontal chroma processing unit is shown in Figure 17. COS SIN Frequency Tracking RTCO Control CFTC HUE, TSTCFR Low Pass RTC_DTO RTC_PID V Gain / PAL Control Saturation Control SECAM Frequency Differentiator CBW,FSEC Low Pass FROM ADC Color Killer UGAIN VGAIN U CKILL CBW,FSEC Auto Detect Gain Tracking SAT CGTC CKILL TSTCGN Figure 17. Offset Control UOFFST VOFFST Coring Control CORE Horizontal Chroma Processing Unit The S5D0127X01 supports chroma input in NTSC, PAL, SECAM and component formats. The color standard is automatically detected and the various chroma processing blocks are enabled as required for the given chroma standard. Details of the various chroma processing blocks follow. 1.4.1. IF Compensation For improved chroma demodulation when the input video is from a mis-tuned IF source, an IF compensation filter is included that has variable gain for the upper chroma side band. This is controlled by the CIFCMP[1:0] bits at location CDEM. The frequency response is shown in Figure 18. For convenience, all plots are normalized to the NTSC modulation frequency. Modified on May/04/2000 ELECTRONICS PAGE 23 OF 96 S5D0127X01 Data Sheet Figure 18. MULTIMEDIA VIDEO Chroma IF Compensation Frequency Response 1.4.2. Demodulation Gain The demodulation gain block is controlled by feedback from the gain tracking block. For NTSC and PAL type inputs, the gain constant is derived from a programmable reference compared against the U component of the input video. This reference is controlled by the SAT register. The default value “0” is the correct gain (saturation for nominal output). For SECAM type input, the feedback is calculated such that proper frequency demodulation is obtained. When external calibration is desired, the gain feed back loop can be “opened” by setting TSTCGN=1. The SAT then controls bits 8 through 1 of a 10 bit multiplier. For standard auto tracking applications, it is recommended that the SAT register be used as an end user saturation control. This register is 2’s compliment. 1.4.3. Demodulation Low Pass Filter The demodulation circuit also contains a programmable low pass filter and a coring function for noise reduction. The chroma low pass filter frequency response for the demodulation circuit for the various video standards are shown in Figure 19 Modified on May/04/2000 ELECTRONICS PAGE 24 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Figure 19. Chroma Low Pass Filter Frequency Response 1.4.4. SECAM Demodulation SECAM processing includes a frequency differentiator, a Cloche and a de-emphasis filter. Frequency response for the filters are shown in Figure 20 and Figure 21. Figure 20. Cloche Filter Frequency Characteristic Modified on May/04/2000 ELECTRONICS PAGE 25 OF 96 S5D0127X01 Data Sheet Figure 21. MULTIMEDIA VIDEO De-emphasis Filter Frequency Response 1.4.5. Additional Chroma Functions S5D0127X01 has many built in auto detection circuits. These allow S5D0127X01 to track any type of video standard input automatically. For analog component video input, the demodulation function is not enabled. The low pass filter provides a group delay for Cb and Cr alignment. This enables the two components to be sampled by one ADC. An RTCO serial output is provided that encodes the current chroma and pixel frequency of the decoder. This information can be used by an Encoder running off of the decoder clock to produce proper color output. The horizontal position of the serial signal is controlled by the HS2 location. The phasing of the DTO and the Encoder can be reset using the RTC_DTO bit. For PAL mode, the PID polarity can be controlled with the RTC_PID bit. 1.5. COMB FILTER Comb filters provide superior Y/C separation for composite NTSC and PAL than simple chroma trap filter. The S5D0127X01 contains on-chip separate 2-line stored luma and chroma comb filters. An internal signal COMB controls for what lines the comb function is enabled. This signal is available through the PORTB pin. Combing is part of the vertical processing which also includes vertical scaling, which is discussed in Section 1.6. A block diagram for the vertical processing section is shown in Figure 22. Modified on May/04/2000 ELECTRONICS PAGE 26 OF 96 S5D0127X01 Data Sheet Luma Chroma MULTIMEDIA VIDEO Horizontal Scaler Horizontal Scaler Luma Vertical Scaler Luma Adaptive Comb Vertical BW Retention Y Chroma Vertical Scaler Sum C Chroma Comb Figure 22. Vertical Processing 1.5.1. Luma Comb Filter The luma comb filter reduces high frequency chroma leakage into the luminance path. The S5D0127X01 uses 2-line stored luma data for combing. Filter coefficients for different video input standards are provided and can be selected automatically based on the video input. Filter coefficients may also be set manually. An optional active comb is employed for NTSC video. Selection of luma comb coefficients is based on line-to-line chroma correlation. Provision is made to disable luma comb for S-video, component, or digital video input. This is achieved by programming the luma comb control register MNYCMB to “1”, and by choosing the value 3 or 4 for YCMBCO[2:0]. This will result either a 1-line or 2-line luma delay. Care must be exercised when disabling the luma comb so that luma line delay matches the chroma path line delay. Special filtering is applied to ensure that high vertical bandwidth is retained for the luma path. 1.5.2. Chroma Comb Filter The chroma comb filter provides further color separation from the composite video. Filter coefficients can be automatically selected based on the input video standard or manually set using NMCCMB and CCMBCO[2:0]. Modified on May/04/2000 ELECTRONICS PAGE 27 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO 1.6. SCALING The S5D0127X01 includes a high quality down scaler. The video images can be down scaled in both horizontal and vertical direction to an arbitrary size. 1.6.1. Horizontal Scaler The horizontal scaler uses a 5-tap 32-phase interpolation filter for luma, and a 3-tap 8-phase interpolation filter for chroma. Scaled pixel data are stored in an on-chip FIFO so they can be sent out in a continuous stream. Horizontal scaling ratio is programmed via the 15-bit register HSCL. The timing signal EHAV is used to indicate when scaled pixel data is available at the video output port. EHAV can be programmed so that it is active for every line regardless of vertical cropping and scaling. Or it can be programmed to be active only for valid video lines. For example, Figure 23 shows the timing for CIF output assuming HAV is programmed to be active for 720 pixels. The HSCL register is programmed with the value 4000 (hex). The trailing edge of EHAV is either aligned with the trailing edge of HAV if the total number of scaled pixels is even, or is one pixel clock earlier if the number is odd. CK2 720 HAV 360 EHAV Y[7:0] Y0 Y1 Y2 Y3 - - Y356 Y357 Y358 Y359 C[7:0] U0 V0 U2 V2 - - U356 V356 U358 V358 Figure 23. Horizontal Scaler Timing for CIF Output (CCIR 601 Pixel Rate) Frequency response and group delay for the luma scaler are shown in Figure 24 and Figure 25, respectively. The luma interpolation filter is designed to achieve relatively flat frequency response and minimal group delay up to the normal video bandwidth. A flat full data path frequency response may be obtained with the help of the luma peaking control register HYPK[1:0]. The high quality filter ensures minimal artifacts for any scaling ratio. Modified on May/04/2000 ELECTRONICS PAGE 28 OF 96 S5D0127X01 Data Sheet Figure 24. MULTIMEDIA VIDEO Horizontal Luma Scaler Interpolation Filter Frequency Response 3 group delay (13.5MHz Clocks) 2.5 2 1.5 1 Figure 25. Horizontal Luma Scaler Interpolation Filter Group Delay Because of the limited bandwidth of the chroma data, a simpler interpolation filter is used for the horizontal chroma scaler. The frequency response and group delay for this filter are shown in Figure 26 and Figure 27, respectively. Modified on May/04/2000 ELECTRONICS PAGE 29 OF 96 S5D0127X01 Data Sheet Figure 26. MULTIMEDIA VIDEO Horizontal Chroma Scaler Interpolation Filter Frequency Response Group Delay 1.5 1.0 0.5 Figure 27. Horizontal Chroma Scaler Interpolation Filter Group Delay 1.6.2. Luma Vertical Scaler Vertical luma scaling uses either a 3-tap or 5-tap 8-phase interpolation filter depending on the horizontal scaling Modified on May/04/2000 ELECTRONICS PAGE 30 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO ratio. Vertical scaling ratio is programmed via the 14-bit register VSCL. A valid scaled line is indicated by the timing signal EVAV being active. The EVAV can be programmed to be internally gated by the VAV signal so it can only be valid within the vertically cropped region. Luma horizontal scaling can use either a 3-tap or a 5-tap interpolation filter depending on the horizontal scaling ration. If the scaled horizontal line has less than or equal to 384 pixels, the 5-tap luma interpolation filter can be turned on by programming the VRT2X bit to a “1”. Otherwise, the VRT2X bit should be set to “0” and the 3-tap filter be used. The VYBW bit provides additional vertical bandwidth control for vertical scaling. Typically, when the vertical scaling ratio is less than 1/2, this bit should be set to “1” to eliminate any aliasing effect. Luma vertical scaler interpolation filter frequency response is shown in Figure 28. Figure 28. Luma Vertical Scaler Interpolation Filter Frequency Response In vertical scaling, the start of signal VAV controls the phase of the vertical scaler interpolation filter. If VAVB, VAVE, VAVOD0, VAVEV0, and VSCL are programmed such that the vertical interpolation filter has the same phase and scaling ratio as that of a memory controller (most memory controller has simple line dropping vertical scaling), it is possible to interface the S5D0127X01 to the memory controller without using EVAV. 1.6.3. Chroma Vertical Scaling Chroma vertical scaling uses different algorithms depending on video input standard and horizontal scaling ratio. If horizontal scaling results in line with less than or equal to 384 pixels, and the VRT2X is set to a “1”, a 5-tap interpolation filter will be used for all video inputs. Otherwise, for NTSC, a 3-tap interpolation filter will be used for NTSC input, and decimation (line dropping without filtering) will be used for PAL and SECAM. Filter characteristics for the 3-tap and 5-tap filters are shown in Figure 29. Modified on May/04/2000 ELECTRONICS PAGE 31 OF 96 S5D0127X01 Data Sheet Figure 29. MULTIMEDIA VIDEO Chroma Vertical Scaler Interpolation Filter Frequency Response Modified on May/04/2000 ELECTRONICS PAGE 32 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO 1.7. VBI DATA PROCESSING The S5D0127X01 VBI data processing is very flexible in that it supports VBI data formats of: • Closed Caption Line 21 Data Service (EIA-608) • 525 line / 60Hz Teletext systems B,C,D (ITU-R BT.653-2) • 625 line / 50Hz Teletext systems A,B,C,D (ITU-R BT.653-2) • Copy Generation Management System (EIA/IS-702) • Wide Screen Signalling (WSS ETS 300 294). Note that the SMPTE data slicing is removed for the S5D0127X01 and replaced with the WSS / CGMS processing. This data can be accessed from the part via four different methods: • Enabling the “Raw un-processed 27MHz” Y ADC samples to be output for the appropriate lines in place of the normal YUV data. • Slicing the data (creating a clock and comparing the data to a threshold at the clock) and bursting this data out on Y output. • Reading the sliced data from two internal registers via the IIC bus. • Via 2 external pins that output the sliced VBI data and the time at which the slice is valid. A simplified block diagram for the VBI section is shown in Figure 30. Normal Decoded Y Data MUX / VBI Format From Y ADC 8 Y[7:0] FRAME ALIGNMENT SLICER TTFRAM FIFO ODD ODDEN EVENEN ODDOS VBIL VYFMT VBINSRT CCEN MODE CTRL LOGIC CLOCK GENERATOR PIXSEL IFMT SECAM Figure 30. CCDATA VBI Decoder Block Diagram Modified on May/04/2000 ELECTRONICS PAGE 33 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Table 7 lists all the video standards that the VBI data slicer supports. Some of these modes are auto detected based on the current video input standard, Table 7: Video Standards Supported by VBI Decoder Value of Chip Detec- Required Values of Characteristics of the Stantion Bits Registers to enable dard Standard Mode Format SECAM 60Hz Teletext system C (NTSC / Intercast) 1 0 2 0 5.727272 272 (34) 50Hz Teletext system B (PAL) 0 0 2 0 6.9375 344 (43) 50Hz Teletext system A (SECAM) 0 1 2 0 6.203125 304 (38) 60Hz Teletext system B 0 1 2 1 5.727272 280 (35) 50Hz Teletext system C 0 1 2 2 5.734375 272 (34) 50Hz Teletext system D 0 1 2 3 5.6457875 280 (35) 60Hz Teletext system D 0 1 2 3 5.727272 280 (35) N/A N/A 1 N/A 0.5035 16 (2) CGMS (NTSC 60Hz) 1 N/A 3 N/A 0.447443 20 (3) WSS (PAL 50Hz) 0 N/A 3 N/A 5.0000 84 Closed Caption NTSC 601 VBIL0-15 TT_SYS Data Rate (MHz) Number of Bits (bytes) Configuring the VBI processing consists of many different steps which are individually explained below. 1.7.1. Enabling the VBI Processor The VBI processor can be enabled independently for the ODD or EVEN fields with the ODDEN and EVENEN bits. Some VBI data is only present on line in 1 of the 2 fields, These independent field enables allow control of the total VBI data output from the chip. These controls apply to all VBI Lines, Thus it is not possible to enable Closed caption line 21 for the Even field and line 19 Teletext for both the odd and even field. 1.7.2. Selecting the Type of Output Data As previously mentioned, there are 4 different ways the VBI data can be extracted. Three of these are selected as shown in the table, the fourth method (CCEN and CCDAT pins) is always available if VBI processing is enabled. Modified on May/04/2000 ELECTRONICS PAGE 34 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Table 8: VBI Data Output Mode (VBILn != 0) VBCVBS 0 VBINSRT 0 0 1 1 1 0 1 Output Mode The VBI data is available via the internal registers CCDAT1 and CCDAT2. Only the last 2 extracted bytes are stored in these registers. Thus, this mode is only useful for extraction of Closed Caption data. This mode enables output of the sliced VBI data. This mode enables output of direct data from the ADC. This mode is invalid. The S5D0127X01 adds an additional output mode and flexibility to vary the modes from line to line. If VBCVBS=0 and VBINSRT=1 S5D0127X01 will output sliced data on enabled lines. By setting VBIMID to 1, any line for which VBIL=3 will output raw ADC data instead of WSS or CGMS. This mode allows a mixture of sliced and raw data. This can be used to output raw data from a teletext line and sliced data from a closed caption line. 1.7.3. Select Individual Lines Enabled for VBI Processing The S5D0127X01 allows programmable selection of processing for the various video lines. For example Teletext/Intercast data can be sliced for lines 14 - 17, and closed caption for line 21. Each 2-bit register VBIL0 through VBIL15 defines how a specific VBI line is processed. As can be seen in Figure 36 for 60 Hz and Figure 37 for 50 Hz video, the following alignments exist: Table 9: VBI Line(s) Selection VBIL number VBIL0 VBIL1 VBIL2 VBIL3 VBIL4 VBIL5 VBIL6 VBIL7 VBIL8 VBIL9 VBIL10 VBIL11 VBIL12 VBIL13 VBIL14 VBIL15 Line Number That the VBIL Processing command applies to (Assuming ODDOS=1) Odd Field Even Field Odd Field Even Field 60 Hz 60 Hz 50 Hz 50Hz All Lines All Lines All lines All lines Except Except 10-24 Except Except 7-21 320 - 334 273-287 9&10 272&273 6&7 319&320 11 274 8 321 12 275 9 322 13 276 10 323 14 277 11 324 15 278 12 325 16 279 13 326 17 280 14 327 18 281 15 328 19 282 16 329 20 283 17 330 21 284 18 331 22 285 19 332 23 286 20 333 24&25 287&288 21&22 334&335 Modified on May/04/2000 ELECTRONICS PAGE 35 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO The ODDOS[1:0] bits allow offset between the odd and even fields. Thus VBIL9 can be lines 17,18 or 19 for ODD fields while VBIL9 is still line 281 for EVEN fields. This extra controls account for variations of VBI data locations from ODD and EVEN fields. When Intercast or Teletext data is selected, an 8-bit user programmable register (TTFRAM) is provided for the framing byte. The frame alignment processor uses this information to properly locate the first data bit on each line 1.7.4. Raw CVBS Data Output Format When raw ADC data is selected as output in place of the normal YUV or RGB data. The following rules apply: • For 656 type 8 bit outputs, The ADC data outputs with successive data points in place of the Cb, Y, Cr, Y data stream. • For 16 bit or 24 bit outputs, The ADC data is output on the Y[7:0] and C[7:0] output pins. At any CK2 clock 2 bytes of ADC data are output. The Y[7:0] bus represents data N while C[7:0] is data N+1. • ADC data is only output during the region that HAV is active. • All ADC outputs are limited to the range 1-254, thus a 0 or 255 value will not be output. For the line selected mode described above using VBCVBS and VBIL, data is from the luma ADC only. If C ADC data or the entire video line is required, configure OFMT bits. 1.7.5. Sliced Data Output Formats While sliced data is available for many of the output formats, the target application is 656 output format. The description of data format is limited to this mode. The S5D0127X01 allows this data to be output during active video. Figure 31 shows the timing diagram for VYFMT[1:0]=3. Video input HAV CCEN -VBINSRT=0 CCEN -VBINSRT=1 Y[7:0] -- VYFMT=3 VBINSRT=1 10h 10h Zoom In - Random data Y[7:0] -- VYFMT=3 VBINSRT=1 10h Figure 31. 10h 10h 10h 84h E4h 85h 45h 2Dh A4h 45h 23h 54h 10h VBI Insertion Timing for VYFMT[1:0]=3 Digitized CVBS data can also be output on the video output port (except for output format 1, 5 and 7). CVBS is Modified on May/04/2000 ELECTRONICS PAGE 36 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO always digitized at the CK clock rate. CVBS data is available when HAV is active. Raw CVBS data is output in a similar fashion as decoded video. For 8-bit output format, data is output at CK rate using the same 8-bit port as the decoded video. For 16-bit and 24-bit output format, data is output at CK2 rate using Y and C ports. The sequence of data output is CVBS2n on Y, and CVBS2n+1 on C (note that EXV port is not used in 24-bit format for outputting raw CVBS data). For Closed Caption data, two read-only registers, CCDAT1 and CCDAT2, are provided so the Closed Caption data can be read via the host interface. The VBIFLG bit can be polled to see if data captured in the two registers can be safely read. 1.8. COLOR SPACE CONVERTER The color space converter processes the video data as YCbCr 4:4:4 when converting to RGB. A programmable limiter (YCRANG) can be imposed on the Y/C data to limit the ranges. One can choose to limit the Y/C to 1 - 254, or Y to 16 - 235 and C to 16 - 240. When selected, YCbCr 4:4:4 is converted to 24 bit RGB according to the following equations: R = C Y + 1.375C R G = C Y – 0.703C R – 0.328C B B = C Y + 1.734C B For 16-bit RGB output, truncation with dithering is used to convert the data from 24 bit to 16 bit. 1.9. GAMMA CORRECTION The S5D0127X01 programmable gamma tables allows the customer to apply many different type of corrections. These corrections can be a standard 2.2 factor for NTSC or 2.8 for PAL. These factors can be applied in the RGB or YUV domains. A basic standard gamma equation of R = R' 2.2 when applied to the R, G, or B signals, generates the response shown as the upper curve below. It is the inverse of the monitor response and thus compensates to produce a linear response. Modified on May/04/2000 ELECTRONICS PAGE 37 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Gamma correction Effective output TV tube characteristic Figure 32. RGB Gamma Correction 1.9.1. Programming the S5D0127X01 The previous response is easily programmed into the S5D0127X01 loading the 0, 8, 16, 24 etc. values into the GAMMA0,1,2,3 locations. Thus every 8th value is stored. The S5D0127X01 will use linear interpolation to generate the values between every 8th points. This is shown in the following figure. Output GAMMA7 GAMMA6 GAMMA5 GAMMA4 GAMMA3 GAMMAD2 GAMMA2 GAMMAD1 GAMMA1 GAMMAD0 GAMMA0 0 8 16 24 32 40 48 56 64 192 200 208 216 224 232 240 248 Input Figure 33. Gamma LUT Programming Modified on May/04/2000 ELECTRONICS PAGE 38 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO For ease of design, the difference between adjacent points must also be loaded. The complete data values for the previous gamma factor of 1/2.2 is shown in the table below. Table 10: RGB Gamma LUT Values Offset GAMMA program at index Offset+40h GAMMAD program at index Offset+60h 0 0 53 1 53 20 2 73 14 3 87 12 4 99 11 5 110 10 6 120 8 7 128 8 8 136 8 9 144 7 A 151 7 B 158 6 C 164 6 D 170 6 E 176 5 F 181 6 10 187 5 11 192 5 12 197 5 13 202 5 14 207 4 15 211 5 16 216 4 17 220 5 18 225 4 19 229 4 1A 233 4 1B 237 4 1C 241 4 1D 245 4 1E 249 3 1F 252 4 The flexibility of this architecture is shown in the following example. Here it is assumed that the S5D0127X01 is operating in a YUV output mode but some form of Gamma correction is required. By converting the RGB gamma correction function back to the YUV color space, the following function can be applied to the U and V signals for improved color performance. This flexibility can be extended in software to produce many type of customer defined transfer functions. Modified on May/04/2000 ELECTRONICS PAGE 39 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Figure 34. Gamma Correction for Cb and Cr Modified on May/04/2000 ELECTRONICS PAGE 40 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO 1.10. DIGITAL VIDEO OUTPUT The S5D0127X01 can output digital video data in various formats, which are tabulated in Table 11. All 8-bit output Table 11: Digital Video Output Format Clock CK2 OFMT 0 1 Type YCbCr 4:2:2 YCbCr 4:1:1 Pin 2N +1 4N +1 CK 4 +2 5 6 7 2, 3 YCbCr RGB RGB RGB 4:4:4 565 888 888 +3 N N N N YCbCr 4:2:2 4N +1 +2 +3 C0 Cb0 Cr0 Cb0 B0 B0 B3 C1 Cb1 Cr1 Cb1 B1 B1 B4 C2 Cb2 Cr2 Cb2 B2 B2 B5 C3 Cb3 Cr3 Cb3 B3 B3 B6 C4 Cb4 Cr4 Cr6 Cr4 Cr2 Cr0 Cb4 B4 B4 B7 C5 Cb5 Cr5 Cr7 Cr5 Cr3 Cr1 Cb5 G0 B5 G2 C6 Cb6 Cr6 Cb6 Cb4 Cb2 Cb0 Cb6 G1 B6 G3 C7 Cb7 Cr7 Cb7 Cb5 Cb3 Cb1 Cb7 G2 B7 G4 Y0 Y0 Y0 Y0 Y0 Y0 Y0 Y0 G3 G0 G5 Cb0 Y0 Cr0 Y0 Y1 Y1 Y1 Y1 Y1 Y1 Y1 Y1 G4 G1 G6 Cb1 Y1 Cr1 Y1 Y2 Y2 Y2 Y2 Y2 Y2 Y2 Y2 G5 G2 G7 Cb2 Y2 Cr2 Y2 Y3 Y3 Y3 Y3 Y3 Y3 Y3 Y3 R0 G3 R3 Cb3 Y3 Cr3 Y3 Y4 Y4 Y4 Y4 Y4 Y4 Y4 Y4 R1 G4 R4 Cb4 Y4 Cr4 Y4 Y5 Y5 Y5 Y5 Y5 Y5 Y5 Y5 R2 G5 R5 Cb5 Y5 Cr5 Y5 Y6 Y6 Y6 Y6 Y6 Y6 Y6 Y6 R3 G6 R6 Cb6 Y6 Cr6 Y6 Y7 Y7 Y7 Y7 Y7 Y7 Y7 Y7 R4 G7 R7 Cb7 Y7 Cr7 Y7 EXV0 Cr0 R0 B0 EXV1 Cr1 R1 B1 EXV2 Cr2 R2 B2 EXV3 Cr3 R3 G0 EXV4 Cr4 R4 G1 EXV5 Cr5 R5 R0 EXV6 Cr6 R6 R1 EXV7 Cr7 R7 R2 Modified on May/04/2000 ELECTRONICS PAGE 41 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO formats use CK as pixel clock; the other formats use CK2 as pixel clock. The first pixel is always aligned to the leading edge of the HAV signal. 1.10.1. Validation Code Insertion S5D0127X01 inserts validation codes during inactive video (HAV is inactive), and invalid video (HAV is active but EHAV is inactive) to assist in recognition of scaled data and VBI data. Table 12 lists the available codes, when they are inserted, and related programming registers. Table 12: Invalid and Unused Code Insertion Code Description INVALY This user programmed code is inserted in the Y or G output stream in scaling operation when HAV is active while EHAV is inactive. Insertion of this code is independent of the output format. Related register is INVALY. INVALU This user programmed code is inserted in the U or B output stream in scaling operation when HAV is active while EHAV is inactive. Insertion of this code is independent of the output format. Related register is INVALU. INVALV This user programmed code is inserted in the V or R output stream in scaling operation when HAV is active while EHAV is inactive. Insertion of this code is independent of the output format. Related register is INVALV. UNUSEY This user programmed code is inserted in the Y or G output stream when HAV is inactive and no other reference code is inserted. Insertion of this code is independent of the output format. Related register is UNUSEY. UNUSEU This user programmed code is inserted in the U or B output stream when HAV is inactive and no other reference code is inserted. Insertion of this code is independent of the output format. Related register is UNUSEU. UNUSEV This user programmed code is inserted in the V or R output stream when HAV is inactive and no other reference code is inserted. Insertion of this code is independent of the output format. Related register is UNUSEV. An example timing diagram for some of the programmable modes is shown in Figure 35. In this diagram, The field rate is 60 Hz, A CCIR 601 sampling rate has been selected thus giving 720 active pixels. The horizontal scaling ratio has been selected for an output of 718 out of 720 pixels. Modified on May/04/2000 ELECTRONICS PAGE 42 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Legend U0 Luma Data with pixel number Chroma (Cb) Data with pixel number V0 Chroma (Cr) Data with pixel number YI UI Programmable INVALY data (index 0x32) Programmable INVALU data (index 0x33) VI Programmable INVALV data (index 0x34) YU UU Programmable UNUSEY data (index 0x35) Programmable UNUSEU data (index 0x36) VU Programmable UNUSEV data (index 0x37) Y0 Invalid data (During Active Video but Scaling Has made picture smaller and right justified) Unused Data If Horizontal Scaling is disabled, EHAV will be the same as HAV and there will be no Invalid data. For this Example, the last Data would be Y719 Unused Data HAV EHAV CK CK2 OFMT=0 (16 bits @ 13.5 Mhz 4:2:2 - 601) Y[7:0] YU YU YU YI YI YI YI Y0 Y1 Y2 Y3 Y0 C[7:0] VU UU VU UI VI UI VI U0 V0 U2 V2 Y0 Y714 U714 Y715 YU YU V714 UU VU OFMT=2 (8 bits @ 27 Mhz 656 data no SAV EAV) Y[7:0] VU YU UU YU VU YU UI YI VI OFMT= 3 YI UI YI VI YI U0 Y0 V0 Y1 U2 Y2 V2 Y3 Y0 Y0 Ux Yx Vx Yx UU YU VU SAV -- see Table 12 for details A B C D EAV -- see table for details Same Format as OFMT=2 A B C D OFMT=4 (24 bits @ 13.5 Mhz YUV 4:4:4) Y[7:0] YU YU YU YI YI YI YI Y0 Y1 Y2 Y3 Y0 YT714 YY715 YU YU C[7:0] UU UU UU UI UI UI UI U0 U1 U2 U3 Y0 U714 Y715 UU UU EXV[7:0] VU VU VU VI VI VI VI V0 V1 V2 V3 V0- V714 V715 VU VU Figure 35. Horizontal Data Timing for Various Output Modes 1.10.2. 656 Op Codes The S5D0127X01 supports timing synchronization through embedded (656) timing reference codes in the output video data stream. This mode is available for output format 3 (OFMT[3:0] = 3). The 656 Op Codes follow the CCIR 656 standard. An optional set of 656 Op Codes can be enabled to identify VBI data using the TASKB bit. The (A,B,C,D) inserted codes for 656 output modes are explained below. Locations in the data stream are shown in Figure 35. The D’ data is substituted for the standard codes shown in column D if TASKB bit is set and the current line is processing VBI data (sliced or raw ADC data format). Modified on May/04/2000 ELECTRONICS PAGE 43 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Fields 60 Hz ODD FIELD 1,3 525 1 2 3 4 524 525 1 2 3 5 6 7 8 9 10 5 6 7 8 9 20 21 22 23 24 21 22 23 Analog Input 4 10 20 Digital output VS VSE=0 VALIGN=0 VSE=1 VSE=1 VSE=0 VS VSE=0 VALIGN=1 VSE=1 VSE=1 VALIGN=0 ODD VSE=1 ODD VSE=0 VSE=0 VALIGN=1 EAV SAV 656 SAV EAV Codes for VSE=0, VALIGN=1 Y[0..7] DAC7 DAC7 F1EC F1 EC F1EC B6AB B6AB B6AB B6AB B6AB B6AB 9D80 9D80 9D80 Y[0..7] TASK B VIP 656 SAV EAV Codes for VSE=0, VALIGN=1, VBIL12-VBIL1=1, TASKB=1 DAC7 DAC7 F1 EC F1 EC F1EC B6AB B6AB B6AB B6AB B6AB B624 13 0E 13 0E 13 0E 13 80 VBIL[N] 0 0 0 0 0 0 0 0 0 0 1 267 268 269 270 271 272 273 266 267 268 269 270 271 272 1 - 11 9D80 9D80 9D80 9D80 13 80 9D80 9D80 12 13 14 15 Fields 2,4 60 Hz EVEN FIELD 263 264 265 262 263 264 266 283 284 285 286 287 283 284 285 286 Analog Input 265 273 Digital Output VSE=0 VALIGN=0 VSE=1 VS VSE=1 VSE=0 VS VSE=1 VSE=0 VALIGN=1 VALIGN=0 ODD VALIGN=1 VSE=0 VSE=1 ODD VSE=0 VSE=1 EAV SAV 656 SAV EAV Codes for VSE=0, VALIGN=1 Y[0..7] 9D80 9D80 B6AB B6AB F1 EC F1 EC F1 EC F1EC F1EC F1EC F1 EC F1 C7 DAC7 DAC7 DAC7 DAC7 DAC7 DAC7 TASK B VIP 656 SAV EAV Codes for VSE=0, VALIGN=1, VBIL12-VBIL1=1, TASKB=1 Y[0..7] 9D80 9D80 B6AB B6AB F1EC F1EC F1 EC F1 EC F1 EC F1EC F162 7F 62 7F62 7F 62 7F 62 7F C7 DAC7 DAC7 VBIL[N] 0 0 0 0 0 0 Figure 36. 0 0 0 0 1 - 11 12 13 14 15 Vertical Timing for 60 Hz Video Modified on May/04/2000 ELECTRONICS 1 PAGE 44 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Fields 1,3 50 Hz ODD FIELD 622 623 624 625 1 2 621 622 623 624 625 3 4 5 6 7 2 3 4 5 6 20 21 22 23 24 25 20 21 22 23 24 Analog Input Digital Output 1 VS VALIGN=0 VSE=1 VS VSE=0 VSE=0, ALT656=1 VSE=1 VALIGN=0 ODD VSE=0 ODD VSE=0 VSE=1 VALIGN=1 VSE=1 7 VSE=0, ALT656=0 VSE=0 VSE=1 VALIGN=1 EAV SAV 656 SAV EAV Codes for VSE=0, VALIGN=1 Y[0..7] DAC7 DAC7 DAC7 F1EC F1EC B6AB B6AB B6AB B6AB B6AB B6AB B6AB B6AB B6AB B6AB B6 AB 9D 80 9D 80 9D TASK B VIP 656 SAV EAV Codes for VSE=0, VALIGN=1, ALT656=1, VBIL15-VBIL1=1, TASKB=1 Y[0..7] DAC7 DAC7 DAC7 DAC7 DAC7 F1EC F1EC B6AB B6AB B6AB B6AB B624 13 0E 130E 13 0E 130E 13 80 9D 80 VBIL[N] 0 0 0 50 Hz EVEN FIELD 310 0 0 0 0 0 0 0 1 318 319 320 1 - 14 15 15 0 334 335 336 337 333 334 335 Fields 2,4 311 312 313 314 315 316 317 310 311 312 313 314 315 316 333 Analog Input 317 318 319 320 336 Digital Output VALIGN=0 VS VS VSE=1 VALIGN=1 VSE=0 ODD ODD VSE=0 VSE=1 VSE=1 VALIGN=0 VSE=0 VSE=1 VSE=-0 VSE=1 VSE=0, ALT656=0 VSE=0, ALT656=1 VALIGN=1 EAV SAV 656 SAV EAV Codes for VSE=0, VALIGN=1 Y[0..7] 9D80 9D80 B6AB B6AB F1EC F1EC F1EC F1EC F1EC F1EC F1EC F1EC F1EC F1EC F1EC F1 EC DAC7 DA C7 DA TASK B VIP 656 SAV EAV Codes for VSE=0, VALIGN=1, ALT656=1, VBIL15-VBIL1=1, TASKB=1 Y[0..7] 9D80 9D80 B6AB B6AB F1EC F1EC F1EC F1EC F1EC F1EC F162 54 49 5449 54 49 54 49 54 49 54 C7 DA C7 VBIL[N] 0 0 0 0 0 0 Figure 37. 0 0 0 0 1 - 14 15 15 0 Vertical Timing For 50 Hz Video Modified on May/04/2000 ELECTRONICS 1 PAGE 45 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Table 13: 656 and TASKB 656 Op Codes SAV / EAV Output Sequence -Reference Output timing pictures Condition Field Vertical Horizontal 656 FVH values A B C D D’ F V H Field 2 Vertical Blank End Active Video FFh 00h 00h F1h 7Fh 1 1 1 Field 2 Vertical Blank Start Active Video FFh 00h 00h ECh 62h 1 1 0 Field 2 Vertical Active End Active Video FFh 00h 00h DAh 54h 1 0 1 Field 2 Vertical Active Start Active Video FFh 00h 00h C7h 49h 1 0 0 Field 1 Vertical Blank End Active Video FFh 00h 00h B6h 38h 0 1 1 Field 1 Vertical Blank Start Active Video FFh 00h 00h ABh 24h 0 1 0 Field 1 Vertical Active End Active Video FFh 00h 00h 9Dh 13h 0 0 1 Field 1 Vertical Active Start Active Video FFh 00h 00h 80h 0Eh 0 0 0 1.10.3. 656 Op Code Vertical Transitions The vertical transition locations of the various 656 Op Codes are shown in Figure 36 and Figure 37. Note that for proper transition locations of the SAV and EAV Op Codes VSE=0 and VALIGN=1. 1.11. HOST INTERFACE The S5D0127X01 supports the IIC serial interface for programming the chip registers. 1.11.1. IIC Interface The two wire interface consists of the SCLK and SDAT signals. Data can be written to or read from the S5D0127X01. For both read and write, each byte is transferred MSB first, and the data bit is valid when the SCLK is high. To write to the slave device, the host initiates a transfer cycle with a START signal. The START signal is HIGH to LOW transition on the SDAT while the SCLK is high. The host then sends a byte consisting of the 7-bit slave device ID and a 0 in the R/W bit. The arrangement for the slave device ID and the R/W bit is depicted in Figure 38. AEX1 and AEX0 are configuration pins used to configure the S5D0127X01 to use one of the four addresses. Up to four S5D0127X01’s can be used in one system each with a unique address. msb 1 lsb 1 0 1 1 AEX1 AEX0 R/W slave device ID Figure 38. IIC Slave Device ID and R/W Byte Modified on May/04/2000 ELECTRONICS PAGE 46 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO The second byte the host sends is the base register index. The host then sends the data. The S5D0127X01 increments the index automatically after each byte of data is sent. Therefore, the host can write multiple bytes to the slave if they are in sequential order. The host completes the transfer cycle with a STOP signal which is a LOW to HIGH transition when the SCLK is high. Each byte transfer consists of 9 clocks. When writing to the S5D0127X01, an acknowledge signal is asserted by the salve device during the 9th clock. SCLK device ID SDAT START index data data ACK ACK STOP ACK Figure 39. IIC Data Write A read cycle takes two START-STOP phases. The first phase is a write to the index register. The second phase is the read from the data register. The host initiates the first phase by sending the START signal. It then sends the slave device ID along with a 0 in the R/W position. The index is then sent followed by the STOP signal. The second phase also starts with the START signal. It then sends the slave device ID but with a 1 in the R/W position to indicate data is to be read from the slave device. The host uses the SCLK to shift data out from the S5D0127X01. A typical second phase in a read transaction is depicted in Figure 40. Auto index increment is supported in Read mode. SCLK index device ID SDAT START ACK STOP ACK SCLK data device ID SDAT STOP START NACK Figure 40. IIC Data Read Modified on May/04/2000 ELECTRONICS PAGE 47 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO 2. CONTROL REGISTER DESCRIPTION This section contains information concerning the programmable control registers. Table 14 provides the default power up values for each index, and a bit map for each register. The following pages describe each register in detail and the possible programing values (an * indicates the power-on default). Gamma correction registers are write only. When the index register points to any of the Gamma correction register, the Gamma look-up table is put into a program mode. Normal operation resumes when the index is outside the range from 0x40 to 0xFF. Table 14: Register Summary Bit Map Index Mnemonic Default 7 6 5 NOVID 4 3 0x00 STAT RO CHIPID VBIFLG FFRDET PALDET 0x01 CMDA 2C POWDN VSE 0x02 CMDB 20 AGCGN 0x03 CMDC 00 VMEN TSTGE1 0 TSTGPK TSTGPH 0x04 CMDD 00 EAV 0 CKDIR INPSL[1:0] 0x05 HAVB 00 HAVB[7:0] 0x06 HAVE 00 HAVE[7:0] 0x07 HS1B 00 HS1B[8:1] 0x08 HS1E 00 HS1E[8:1] 0x09 HS2B 00 HS2B[8:1] 0x0A HS2E 00 HS2E[8:1] 0x0B AGC 50 0x0C HXTRA 00 0x0D CDEM 00 OUTHIZ 0x0E PORTAB 00 DIRB 0x0F LUMA 00 0 0x10 CON 00 CONT[7:0] 0x11 BRT 00 BRT[7:0] 0x12 CHROMA 08 0x13 CHROMB 00 CDLY[3:0] 0x14 DEMOD 00 FSCDET SECDET CDMLPF CTRACK 0x15 SAT 00 SAT[7:0] 0x16 HUE 00 HUE[7:0] 0x17 VERTIA 00 0x18 VERTIB 00 0x19 VERTIC 03 0x1A HSCLL 00 0x1B HSCLH 00 0x1C VSCLL FC 0x1D VSCLH FF 0x1E OFMTA - 0x1F OFMTB 00 HFSEL[1:0] XT24 VALIGN AGCOVF AGCFRZ 1 0 CDET HLOCK CLOCK PIXSEL MNFMT IFMT INSEL[3:0] TSTGFR[1:0] SYNDIR TSTGEN Y1MHZ GPPORT AGC[7:0] HAVB[10:8] FSEC HAVE[10:8] 0 CIFCMP[1:0] DATAB[2:0] UNIT ACCFRZ RGBH PALM MNYCMB PED PALN 0 HYBWR CBW 0 0 DATAA[2:0] CTRAP CORE[1:0] HYPK[1:0] CKILL[1:0] SCHCMP[3:0] YCMBCO[2:0] HYLPF[2:0] MNCCMB HS1BE0 HS2BE0 DIRA MNFSC[1:0] VRT2X HYBWI CCMBCO[2:0] HYDEC ACMBEN MNSECAM[1:0] VCTRL[2:0] VSCLEN[1:0] VYBW 0 EVAVEV EVAVOD HSCL[6:0] CMBMOD HSCL[14:7] VSCL[5:0] ACMBCO ACMBRE VSCL[13:6] GAMEN[1:0] VSVAV OENC[1:0] EVAND[1:0] EVHS1 OFMT[3:0] EVHAV Modified on May/04/2000 ELECTRONICS 2 EVEHAV EVAVG EVANDL PAGE 48 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Table 14: Register Summary Bit Map Index Mnemonic Default 7 6 5 4 2 0 VBICTL 00 VBCVBS 0x21 CCDAT1 RO b0 0x22 CCDAT2 RO b0 0x23 VBIL30 00 VBIL3 VBIL2 VBIL1 VBIL0 0x24 VBIL74 00 VBIL7 VBIL6 VBIL5 VBIL4 0x25 VBIL118 00 VBIL11 VBIL10 VBIL9 VBIL8 0x26 VBIL1512 00 VBIL15 VBIL14 VBIL13 VBIL12 0x27 TTFRAM 00 0x28 TESTA 00 0x29 UVOFFH 00 0x2A UVOFFL 33 0x2B UGAIN 00 UGAIN[7:0] 0x2C VGAIN 00 VGAIN[7:0] 0x2D VAVB 23 0x2E VAVE 82 0x2F CTRACK 00 0 0 0x30 POLCTL 00 EVAVPL VSPL ODDPL HAVPL EHAVPL HS2PL VAVPL HS1PL 0x31 REFCOD 00 YCRANG 0 0 0 0 0 0 0 0x32 INVALY 10 INVALY[7:0] 0x33 INVALU 80 INVALU[7:0] 0x34 INVALV 80 INVALV[7:0] 0x35 UNUSEY 10 UNUSEY[7:0] 0x36 UNUSEU 80 UNUSEU[7:0] 0x37 UNUSEV 80 UNUSEV[7:0] 0x38 EXCTRL 00 0 0 CLEVEL 0x39 TRACKA 00 0x3A VBICTLB 00 VBISWAP 0x3B TRACKB 00 ALT656 0x3C RTC 00 0x3D CMDE 09 0x3E VSDEL 00 0x3F CMDF 00 - 0x60-7F GAMMAD - 0xC0-DF GAMUV - 0xE0-FF GAMUVD - b1 b2 b1 b2 VBINSRT ODDEN EVENEN 1 0x20 0x40-5F GAMMA VYFMT[1:0] 3 b3 b4 b3 b4 b5 b6 b5 b6 P1 P2 TTFRAM[7:0] 0 0 TSTCLC TSTCGN 0 0 0 TSTCFR 0 0 UOFFST[5:4] UOFFST[3:0] 0 0 VOFFST[5:4] VOFFST[3:0] VAVB[6:1] VAVOD0 VAVEV0 VAVE[8:1] DMCTL[1:0] 0 ENINCST 0 STCTRL MAC_DETVCR_DET TT_SYS[1:0] VBI_PH RTC_DTO RTC_PID ODFST VSALG TR_MS NOVIDC CTRAPFSC VIPMODE - CGTC[1:0] AUCPWD VCR_LEV[1:0] VBI_FR PH_CTRL VNOISCT 0 CFTC[1:0] ATCTRAP VCTRAP AGCLSB VBIMID NEW_CC CC_OVFL YOFFENB COFFENB TDMOD HCORE[1:0] 0 AGC_LPG[1:0] 0 0 AGC_LKG 0 CHIPREVID VSDEC[5:0] EVAVY UVDLEN UVDLSL REGUD TASKB CBWI GAMMA0[7:0] - GAMMA31[7:0] - - - - GAMMAD0[5:0] - GAMMAD31[5:0] GAMUV0[7:0] - GAMUV31[7:0] GAMUVD0[5:0] - GAMUVD31[5:0] Modified on May/04/2000 ELECTRONICS ODDOS[1:0] PAGE 49 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Read Only Status Bits Index Mnemonic bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 00h STAT CHIPID VBIFLG NOVID FFRDET PALDET CDET HLOCK CLOCK CLOCK HLOCK CDET PALDET FFRDET NOVID VBIFLG CHIPID Status for color lock. 0 Not locked. 1 Color lock achieved. Status for current line tracking mode. 0 Chip is in initial tracking mode. 1 Chip is in steady state tracking mode. Status for detection of color. 0 No color signal is detected. 1 Color signal is detected. Status for current detected color format. Information contained in this bit is valid only if CLOCK is 1. 0 NTSC color format. 1 PAL color format. Status for current detected field frequency. 0 50 Hz field frequency, i.e. N,B,G,H,I,D,K,K1,L system. 1 60 Hz field frequency, i.e. M system. Video detect flag. This bit should not be used for detecting the presence of a TV channel from the output of a TV tuner. 0 Sync has been detected for the last 32 lines. 1 No sync has been detected. Vertical blanking interval flag. 0 Video is in active region. 1 Video is in vertical blanking region. Chip version ID. Please refer to the CHIPREVID bits for additional information 0 KS0122S. 1 S5D0124D01. Modified on May/04/2000 ELECTRONICS PAGE 50 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Control Register A Index Mnemonic bit 7 bit 6 01h CMDA POWDN VSE IFMT HFSEL[1:0] bit 3 bit 2 bit 1 bit 0 XT24 PIXSEL MNFMT IFMT 0 Chip is forced to assume input is 50 Hz.* 1 Chip is forced to assume input is 60 Hz. Manual input format control override. When this bit is 1 the IFMT bit is enabled. PIXSEL 0 The chip determines the input video standard based on the detected field rate:* NTSC if 60 Hz. PAL/SECAM if 50 Hz. 1 Input video standard is selected with the IFMT bit. Select pixel sampling rate. XT24 0 Output data is at square pixel rate. 1 Output data is at CCIR 601 rate.* Select the external clock reference frequency. HFSEL[1:0] POWDN bit 4 Manual video input standard select. Standard selection can be controlled automatically if MNFMT=0. MNFMT VSE bit 5 0 External clock is 26.8 MHz. 1 External clock is 24.576 MHz.* Horizontal tracking loop frequency select. 0 Force loop to very fast. 1 Force loop to fast. 2 Force loop to VCR time constant.* 3 Force loop to TV time constant. Change the vertical end location of the VS. 0 Line 10/10.5.* 1 Line 9/9.5. Power down mode. 0 Normal operation.* 1 All chip functions except microprocessor interface and CK/CK2 generation are disabled. The output of the CK/CK2 pins retains the most recent frequency when the power down mode is enabled. Modified on May/04/2000 ELECTRONICS PAGE 51 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Control Register B Index Mnemonic bit 7 bit 6 bit 5 bit 4 02h CMDB AGCGN VALIGN AGCOVF AGCFRZ INSEL[3:0] AGCFRZ AGCOVF VALIGN AGCGN bit 3 bit 2 bit 1 bit 0 INSEL[3:0] Analog input channel select. 0 AY0 is composite input.* 1 AY1 is composite input. 2 AY2 is composite input. 4 AC0 is composite input. 5 AC1 is composite input. 6 AC2 is composite input. 8 AY0 is luminance input, AC0 is chrominance input. 9 AY1 is luminance input, AC1 is chrominance input. A AY2 is luminance input, AC2 is chrominance input. F AY2 is luminance input, AC1 is Cb input, AC2 is Cr input. Freeze the analog AGC for the Y and C paths at their current values. 0 AGC is running. Reading AGC register returns the current AGC gain.* 1 AGC is frozen. Gain can be changed or read with AGC register. AGC gain control mode. 0 AGC gain tracks to sync tip and back porch delta. 1 If ADC overflows, AGC gain will be reduced (this has higher priority over normal sync tip - back porch tracking).* VS edge alignment control. 0 VS leading edge occurs during serration pulses (typically within the first serration pulse). VS trailing edge is aligned to half line or beginning of the line depending on the field.* 1 VS leading edge is aligned to half line or beginning of the line depending on the field. VS trailing edge is always aligned to beginning of the line. AGC gain calculation. 0 Normal mode. AGC gain calculation is based on sync tip to back porch difference equal to 68 ADC code.* 1 AGC gain calculation is base on sync tip to back porch difference equal to 54 ADC code. This will reduce the AGC gain by a factor of 1/1.25 compare to normal mode. When used in conjunction with PED and RGBH, this effectively increases the input dynamic range. Modified on May/04/2000 ELECTRONICS PAGE 52 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Control register C Index Mnemonic bit 7 bit 6 bit 5 bit 4 bit 3 03h CMDC VMEN TSTGE1 0 TSTGPK TSTGPH TSTGEN TSTGPK TSTGE1 VMEN bit 1 bit 0 TSTGFR[1:0] TSTGEN Enable manual control of horizontal phase and frequency tracking. TSTGFR[1:0] TSTGPH bit 2 0 Auto phase and frequency tracking.* 1 Enable manual control of horizontal phase and frequency with TSTGFR[1:0] and TSTGPH. When TSTGEN == 1, these two bits control the horizontal frequency tracking. 00 Stop frequency tracking and freeze the frequency at the current value.* 01 Horizontal frequency tracks the input. 1X Horizontal frequency tracking ignores video input and runs at nominal value based on the field rate and output pixel rate selected by IFMT and PIXEL bits. When TSTGEN == 1, this bit controls the horizontal phase tracking. 0 No phase tracking.* 1 Horizontal phase tracks the input video or HS1 input if in slave mode. If TSTGE1 == 1, this bit controls AGC. 0 AGC clamps to back porch and gain is set based on sync tip-back porch difference.* 1 AGC clamps to sync tip and gain is set based on peak-valley difference. Enables the function of TSTGPK. 0 Disables TSTGPK.* 1 Enables TSTGPK. Vertical master mode. 0 Normal vertical sync operation.* 1 Vertical sync ignores input and free runs at 50 Hz or 60 Hz. This mode can be used to generate video timing for a slave device. Modified on May/04/2000 ELECTRONICS PAGE 53 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Control Register D Index Mnemonic bit 7 bit 6 bit 5 04h CMDD EAV 0 CKDIR bit 4 bit 3 INPSL[1:0] bit 2 bit 1 bit 0 SYNDIR Y1MHZ GPPORT GPPORT General purpose port. This register is useful only if DATAA[2:0] == 7. If DIRA == 0, this bit is read only and reflects the logic state at PORTA pin. If DIRA == 1, any value written to this bit will appear at PORTA pin. Y1MHZ Luma bandwidth control. SYNDIR INPSL[1:0] CKDIR EAV 0 Luma bandwidth is controlled by other luma filters in the luma path.* 1 Luma data is low pass filtered to 1MHz bandwidth. HS1 and VS pin direction control. 0 HS1 and VS are output.* 1 HS1 and VS are input. Video input and clock source select. 0 Video source is analog and connected to the chip’s analog input. Clock is internally generated.* 1 Video source is 8-bit digital CbYCr and connected to EXV0 through EXV7 pins. 3 Video source is 8-bit digitized CVBS and connected to EXV0 through EXV7 pins. Clock select. 0 Clock is from internal clock generator. A reference clock at XTALI pin is required.* 1 Clock is from CK pin. When this is selected, the CK pin automatically becomes an input. In 8-bit digital CbYCr input mode, this bit selects the sync source. 0 Horizontal and vertical syncs are from HS1 and VS pins, respectively.* 1 Syncs are embedded in the 8-bit digital data stream (CCIR 656 compatible). Modified on May/04/2000 ELECTRONICS PAGE 54 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO HAV Start Control Index Mnemonic 05h HAVB 0Ch HXTRA HAVB[10:0] bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 HAVB[7:0] HAVB[10:8] HAVE[10:8] HS1BE0 HS2BE0 This 11-bit register is used to define the start location of the HAV signal relative to the sync tip (for CVBS input, this is the composite video sync tip. For 8-bit CbYCr input, this is the leading edge of the HS1 or EAV). The content of this register is a 2’s complement number which is used as an offset to the default. The resolution for this register is 1 CK clock. HAV End Control Index Mnemonic 06h HAVE 0Ch HXTRA HAVE[10:0] bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 HAVE[7:0] HAVB[10:8] HAVE[10:8] HS1BE0 HS2BE0 This 11-bit register is used to define the end location of the HAV signal relative to the sync tip. The content of this register is a 2’s complement number which is used as an offset to the default The resolution for this register is 1 CK clock. HS1 Start Control Index Mnemonic 07h HS1B 0Ch HXTRA HS1B[8:1] HS1BE0 bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 HS1B[8:1] HAVB[10:8] HAVE[10:8] HS1BE0 HS2BE0 If HS1 is programmed as an output, this 9-bit register defines the start location of the HS1 signal. The content of this register is a 2’s complement number which is used as an offset to the default. The resolution for this register is 1 CK clock. Modified on May/04/2000 ELECTRONICS PAGE 55 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO HS1 End Control Index Mnemonic 08h HS1E 0Ch HXTRA HS1E[8:1] HS1BE0 bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 HS1E[8:1] HAVB[10:8] HAVE[10:8] HS1BE0 HS2BE0 If HS1 is programmed as an output, this 9-bit register defines the end location of the HS1 signal. The content of this register is a 2’s complement number which is used as an offset to the default. The resolution for this register is 1 CK clock. HS2 Start Control Index Mnemonic 09h HS2B 0Ch HXTRA HS2B[8:1] HS2BE0 bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 HS2B[8:1] HAVB[10:8] HAVE[10:8] HS1BE0 HS2BE0 This 9-bit register defines the start location of the HS2 signal. The content of this register is a 2’s complement number which is used as an offset to the default HS2B location. The resolution for this register is 1 CK clock. HS2 End Control Index Mnemonic 0Ah HS2E 0Ch HXTRA HS2E[8:1] HS2BE0 bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 HS2E[8:1] HAVB[10:8] HAVE[10:8] HS1BE0 HS2BE0 This 9-bit register defines the end location of the HS2 signal. The content of this register is a 2’s complement number which is used as an offset to the default HS2E location. The resolution for this register is 1 CK clock. Modified on May/04/2000 ELECTRONICS PAGE 56 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO AGC Control Index Mnemonic 0x0B AGC AGC[7:0] bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 AGC[7:0] This register is used to manually set AGC when AGCFRZ is set to “1”. The content in the register is unsigned. Chroma Demodulation Control Index Mnemonic bit 7 bit 6 bit 5 0Dh CDEM OUTHIZ FSEC 0 CIFCMP[1:0] FSEC OUTHIZ bit 4 bit 3 CIFCMP[1:0] bit 2 bit 1 bit 0 0 0 0 IF compensation for the chroma path. 0 No compensation.* 1 Upper chroma side band is 1 dB higher than lower side band. 2 Upper chroma side band is 3 dB higher than lower side band. 3 Upper chroma side band is 6 dB higher than lower side band. Chroma frequency demodulation filter select for SECAM video. 0 Select SECAM chroma frequency demodulation filter if SECAM video is detected.* 1 Always use SECAM chroma frequency demodulation filter. This is the software output three-state control bit. If this bit is set to a “1”, or the OEN pin is at a logic LOW level, output pins can be selectively put in the high impedance state using the additional software control bits OENC[1:0]. 0 This is default setting.* 1 This will enable the output pins to be three-stated regardless the state of the OEN pin. Modified on May/04/2000 ELECTRONICS PAGE 57 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Port A and B Control Index Mnemonic bit 7 0Eh PORTAB DIRB DATAA[2:0] DIRA DATAB[2:0] DIRB bit 6 bit 5 bit 4 DATAB[2:0] bit 3 bit 2 DIRA bit 1 bit 0 DATAA[2:0] Port A data select. For internal gate signal locations. 0 Port A is disconnected from the internal signal path.* 1 Port A is connected to the BPG (back porch gate) signal. 2 Port A is connected to the SYG (sync tip gate) signal. 3 Port A is connected to the CBG (color burst gate) signal. 4 Port A is connected to the CBGW (color burst gate wide) signal. The CBGW is high for the entire color burst period. 5 Port A is connected to the SLICE (mid way of the sync tip) signal. 6 Port A is connected to the VBI (vertical blanking interval) signal. 7 Port A is connect to the GPPORT bit. Port A direction control. 0 Port A is configured as input. The input is connected directly to the signal path selected by DATAA[2:0]. The internally generated gate signal is disconnected from the signal path.* 1 Port A is an output and is driven by the internally generated signal as selected by DATAA[2:0]. Port B data select. For internal gate signal locations. 0 Port B is disconnected from the internal signal path.* 1 Port B is connected to the SCH (sync tip to color burst phase) signal. 2 Port B is connected to the FH2 (twice per line pulses) signal. 3 Port B is connected to the FS_PULSE (falling edge of the sync tip) signal. 4 Port B is connected to the VBI_CVBS (VBI raw ADC) signal. This signal is high for those lines that output data directly from the ADC (not YUV or RGB data). 5 Port B is connected to the VBI_PROC (VBI sliced) signal. This signal is high for those video lines that output sliced VBI data. 6 Port B is connected to the VS (Vertical Sync) signal. 7 Port B is configured as the RTCO (Real Time Control Output). This single pin serial interface transmits phase and frequency information to a video encoder so that it may operate directly from the S5D0127X01 output clock. Port B direction control. 0 Port B is configured as input. The input is connected directly to the signal path selected by DATAB[2:0]. The internally generated gate signal is disconnected from the signal path.* 1 Port B is an output and is driven by the internally generated signal as selected by DATAB[2:0]. Modified on May/04/2000 ELECTRONICS PAGE 58 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Luma Control Register Index Mnemonic bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 0x0F LUMA 0 UNIT RGBH PED HYBWR CTRAP HYPK[1:0] CTRAP HYBWR PED RGBH UNIT bit 1 bit 0 HYPK[1:0] Luminance horizontal peaking control around 3 MHz. 0 Less than nominal peaking.* 1 Nominal peaking. 2 Increased peaking. 3 Maximum peaking. Chroma trap (notch filter) in the luma path. 0 No chroma trap. This mode is recommended for S-video or component video input.* 1 Chroma trap is enabled. Luminance horizontal bandwidth reduction control. 0 Full bandwidth.* 1 Reduced bandwidth. Enable gain correction for 7.5 blank-to-black setup (pedestal). 0 No pedestal. 0% = Y code 16. 100% = Y code 235.* 1 Gain adjusted for 7.5% blank-to-black setup (pedestal). 7.5% = Y code 16. 7.5% 100% input produce Y code 16 - 235. High gain to produce full range Y for 0% (or 7.5% if PED = 1) to 100% input. 0 Black (0% or 7.5%) to peak white(100%) input produce Y code 16 to 235.*. 1 Black (0% or 7.5%) to peak white(100%) input produce Y code 0 to 255. When PED and RGBH are both set to a “1”, setting this bit to a “1” produces a unit gain for CCIR 601 digital input (INPSL[1:0] = 1). 0 Luma DC gain is controlled by PED and RGBH as described for each bit.* 1 Luma DC gain is unity for CCIR 601 digital input. Modified on May/04/2000 ELECTRONICS PAGE 59 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Contrast Control Index Mnemonic 0x10 CON CON[7:0] bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 bit 1 bit 0 CON[7:0] This 8-bit register contains a 2’s compliment number for contrast control. Brightness Control Index Mnemonic 0x11 BRT BRT[7:0] bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 BRT[7:0] Brightness control register. The number contained in the register is 2’s compliment. Modified on May/04/2000 ELECTRONICS PAGE 60 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Chroma Control Register A Index Mnemonic bit 7 bit 6 bit 5 bit 4 0x12 CHROMA ACCFRZ PALM PALN CBW CKILL[1:0] CORE[1:0] CBW PALN PALM ACCFRZ bit 3 bit 2 CORE[1:0] bit 1 bit 0 CKILL[1:0] Color kill. 0 Auto detect mode. If color burst is too low or no color burst, chroma data is forced to code 128.* 2 Chroma is always ON. 3 Chroma data is always forced to code 128. Chroma data coring. 0 No coring. 1 Chroma data within the range 128+/-1, inclusive, will be force to 128. 2 Chroma data within the range 128+/-3, inclusive, will be force to 128.* 3 Chroma data within the range 128+/-7, inclusive, will be force to 128. Chroma bandwidth control. 0 Chroma 3 dB bandwidth is 850 kHz.* 1 Chroma 3 dB bandwidth is 550 kHz. Select color tracking for PAL-N, or NTSC-N when input field rate is 50 Hz and Fsc is 3.58 MHz. 0 Select NTSC-N.* 1 Select PAL-N. Select color tracking for PAL-M or NTSC-M when input field rate is 60 Hz. 0 Select color tracking for NTSC-M.* 1 Select color tracking for PAL-M. Chroma gain tracking freeze control. 0 Chroma gain tracks the input. Color saturation can be adjusted with SAT.* 1 Chroma gain freezes at the current saturation level. Modified on May/04/2000 ELECTRONICS PAGE 61 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Chroma Control Register B Index Mnemonic 0x13 CHROMB bit 7 bit 6 bit 5 bit 4 bit 3 CDLY[3:0] bit 2 bit 1 bit 0 SCHCMP[3:0] SCHCMP[3:0] Phase constant compare value for color burst phase relative to sync tip. Each step is 22.5 degrees with the value of 0 equal to 0 degree. CDLY[3:0] Chroma path group delay relative to the luma path (in unit of CK): 0 No delay.* 1 -0.5 2 1 3 0.5 4 2 5 1.5 6 3 7 2.5 8 -4 9 -4.5 A -3 B -3.5 C -2 D -2.5 E -1 F -1.5 Modified on May/04/2000 ELECTRONICS PAGE 62 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Chroma Demodulation Control and Status Index Mnemonic 0x14 DEMOD bit 7 bit 6 bit 5 bit 4 bit 3 FSCDET SECDET CDMLPF CTRACK bit 2 MNFSC[1:0] bit 1 bit 0 MNSECAM[1:0] MNSECAM[1:0] Enable manual SECAM input detection. MNFSC[1:0] CTRACK CDMLPF SECDET FSCDET 0 Detection of SECAM input is automatic.* 2 Force the chip to assume input is not SECAM. 3 Force the chip to assume input is SECAM. Enable manual Fsc detection. 0 Detection of Fsc frequency is automatic.* 2 Force chip to assume input Fsc is 4.43 MHz or 4.286 MHz. 3 Force chip to assume input Fsc is 3.58 MHz. Chroma frequency tracking mode. 0 Chroma frequency tracking is based on the field rate and Fsc.* 1 Chroma frequency tracking is based on field rate only. Bypass the LPF in the chroma demodulator. 0 Chroma data pass through the LPF for color demodulation.* 1 Chroma data bypass the LPF. This setting is used for component video input. SECAM detection (read only). 0 Chip did not detect SECAM input. 1 Chip detected SECAM input. Color subcarrier detection (read only). 0 Chip detected 4.43 MHz or 4.286 MHz Fsc. 1 Chip detected 3.58 MHz Fsc. Modified on May/04/2000 ELECTRONICS PAGE 63 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Color Saturation Control Index Mnemonic 0x15 SAT SAT[7:0] bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 SAT[7:0] Color saturation control register. Register content is in 2’s compliment if TSTCGN=0. 0 value corresponds to nominal saturation. Hue Control Index Mnemonic 0x16 HUE HUE[7:0] bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 HUE[7:0] Hue control register. The register content is in 2’s compliment format. It covers the range from -180° to +178.59° degree. The resolution is 1.41°/LSB. Modified on May/04/2000 ELECTRONICS PAGE 64 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Vertical Processing Control A Index Mnemonic bit 7 0x17 VERTIA MNYCMB VCTRL[2:0] bit 5 bit 4 YCMBCO[2:0] bit 3 bit 2 VRT2X bit 1 bit 0 VCTRL[2:0] Luminance vertical filter control. VRT2X 0 Scaler uses LPF path, comb uses HPF.* 1 Scaler uses full bandwidth, comb is disabled. 2 Scaler is disabled, comb uses full bandwidth. 3 Scaler uses LPF, comb is disabled. 4 Scaler is disabled, comb uses HPF. 3/5-tap vertical scaler filter select. YCMBCO[2:0] MNYCMB bit 6 0 Select 3-tap vertical scaler filter.* 1 Select 5-tap vertical scaler filter. This option can be used only if horizontally cropped line is less than or equal to 384 pixels. Luma comb filter coefficients selection when the MNYCMB is set to “1”. 0 [1/4 1/2 1/4].* 1 [3/8 1/2 1/8]. 2 [1/2 1/2 0]. 3 [1 0 0]. 4 [0 1 0]. 5 [1/2 0 1/2]. 6 [0 1/2 1/2]. 7 [1/8 1/2 3/8]. Select between auto and manual luma comb filter coefficients. 0 Luma comb filter coefficients are automatically selected based on input video standard.* 1 Luma comb filter coefficients are selected with YCMBCO[2:0]. Modified on May/04/2000 ELECTRONICS PAGE 65 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Vertical Processing Control B Index Mnemonic 0x18 VERTIB VSCLEN[1:0] HYDEC HYBWI HYLPF[2:0] bit 7 bit 6 bit 5 HYLPF[2:0] bit 4 bit 3 HYBWI HYDEC bit 2 bit 1 VSCLEN[1:0] bit 0 0 Vertical scaling enable. 0 Vertical scaling is enabled.* 1 Vertical scaling is disabled. 2 Vertical scaling is disabled. Video is 1-line delayed. 3 Vertical scaling is disabled. Video is 2-line delayed. Luma path decimation filter enable. 0 Luma path decimation is enabled.* 1 Luma path decimation is disabled. Luma path decimation filter bandwidth select. 0 Normal bandwidth.* 1 Bandwidth is 1 MHz higher. Horizontal luma LPF bandwidth control. 0 Full bandwidth.* 1 4.5 MHz bandwidth. 2 3.5 MHz bandwidth. 3 2.5 MHz bandwidth. 4 1.5 MHz bandwidth. Modified on May/04/2000 ELECTRONICS PAGE 66 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Vertical Processing Control C Index Mnemonic bit 7 0x19 VERTIC MNCCMB EVAVOD bit 5 bit 4 CCMBCO[2:0] bit 3 bit 2 bit 1 bit 0 ACMBEN VYBW EVAVEV EVAVOD Enable VAV signal output during ODD field. EVAVEV 0 VAV signal is disabled (always inactive) during ODD field. 1 VAV signal is enabled during ODD field.* Enable VAV signal output during EVEN field. VYBW 0 VAV signal is disabled (always inactive) during EVEN field. 1 VAV signal is enabled during EVEN field.* Luma vertical bandwidth control. ACMBEN 0 Full bandwidth.* 1 Reduced bandwidth. Enable luma active comb for NTSC. CCMBCO[2:0] MNCCMB bit 6 0 Active comb is disabled.* 1 Active comb is enabled. Manual chorma comb filter coefficients select. 0 Select the coefficient set [1/2 1/2 0] (if VRT2X = 0).* 1 Select the coefficient set [1/4 1/2 1/4] (if VRT2X = 0). 2 Select the coefficient set [0 1/2 1/2 0 0] (if VRT2X = 1). 3 Select the coefficient set [0 1/4 1/2 1/4 0] (if VRT2X = 1). 4 Select the coefficient set [1 0 0]. 5 Select the coefficient set [0 1 0]. 6 Select the coefficient set [0 0 1]. 7 No output (disabled). Chroma comb filter coefficients are selected automatically or manually. 0 Filter coefficients are automatically selected based on the selected video input standard. SECAM must use this value.* 1 Filter coefficients are selected manually with CCMBCO[2:0]. Modified on May/04/2000 ELECTRONICS PAGE 67 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Horizontal Scaling Ratio Index Mnemonic 0x1A HSCLL 0x1B HSCLH CMBMOD bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 HSCL[6:0] bit 0 CMBMOD HSCL[14:7] This bit controls when comb is enabled internally. HSCL[14:0] 0 Comb is enabled by the internal signal COMB_EN.* 1 Comb is enabled when VAV is active. The 15-bit register defines a horizontal scaling ratio of HSCL[14:0]/215. Any change to this value will become effective during the next vertical sync. Vertical Scaling Ratio Index Mnemonic 0x1C VSCLL 0x1D VSCLH ACMBRE ACMBCO VSCL[13:0] bit 7 bit 6 bit 5 bit 4 bit 3 VSCL[5:0] bit 2 bit 1 bit 0 ACMBCO ACMBRE VSCL[13:6] Active comb filter threshold select. 0 High threshold.* 1 Low threshold. Active comb filter coefficient set select. 0 Use the set of coefficients for 100% comb.* 1 Use the set of coefficients for 75% comb. The 14-bit register defines a vertical scaling ratio of VSCL[13:0]/214. Any change to this value will become effective during the next vertical sync. Modified on May/04/2000 ELECTRONICS PAGE 68 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Output Control A Index Mnemonic 0x1E OFMTA OFMT[3:0] bit 7 bit 6 bit 5 GAMEN[1:0] bit 4 bit 3 bit 2 OENC[1:0] bit 1 bit 0 OFMT[3:0] Digital video output format select. 16 and 24 bit data are output at CK2 clock rate. 8 bit data are output at CK clock rate. OENC[1:0] 0 16-bit YCbCr 4:2:2 output on the Y and C output ports.* 1 12-bit YCbCr 4:1:1 output on the Y and C output ports. 2 8-bit YCbCr 4:2:2 without embedded timing reference codes. 3 8-bit YCbCr 4:2:2 with embedded timing reference codes. 4 24-bit YCbCr 4:4:4. 5 16-bit RGB 565. 6 24-bit RGB 888 with linear bit ordering. 7 24-bit RGB 888, bit ordering is an extension of the 16-bit RGB 565 format. 8 Same as mode 2 with the additional of 8-bit YCbCr 4:2:2 data output on the EXV port. While the Y port can be scaled down, the EXV port will always be a full size picture. 9 Same as 8 with the addition of SAV and EAV codes. A output Y ADC data all the time (including syncs) on the Y port, C port is non-scaled 656 data with no timing codes. B Output Y ADC data all the time (including syncs) on the Y port, Output C ADC data all the time (including syncs) on the C port, When either the OEN pin is low or the OUTHIZ is a “1”, these two bits will determine which output pins are three-stated. GAMEN[1:0] 0 All video pins are three-stated. 1 All video pins, plus HAV, VAV, EVAV, EHAV, PID, ODD, HS1, HS2, VS, and SCH are three-stated. 2 All pins listed above, plus CK and CK2 are three-stated. 3 Always output data. Gamma correction enable. 0 No gamma correction.* 1 Gamma correction is applied to Y/G data. 2 Gamma correction is applied to U/B and V/R data. 3 Gamma correction is applied to Y/G, U/B, and V/R data. Modified on May/04/2000 ELECTRONICS PAGE 69 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Output Control B Index Mnemonic bit 7 0x1F OFMTB VSVAV EVAVG bit 5 EVAND[1:0] bit 4 bit 3 bit 2 bit 1 bit 0 EVHS1 EVHAV EVEHAV EVAVG EVANDL Gate EVAV with VAV before sending to output. EVEHAV 0 EVAV is not gated with VAV. EVAV may be active outside of active VAV region.* 1 EVAV is gated with VAV. EVAV can be active only when VAV is active. Additional qualifier for EHAV. EVHAV 0 No additional qualifier.* 1 EHAV uses qualifier from EVAND[1:0]. Additional qualifier for HAV. EVHS1 0 No additional qualifier.* 1 HAV uses qualifier from EVAND[1:0]. Additional qualifier for HS1. EVAND[1:0], EVANDL VSVAV bit 6 0 No additional qualifier.* 1 HS1 uses qualifier from EVAND[1:0]. Qualifier signal that defines active video lines. This control enables 656 codes, HAV, EHAV and HS1. EVANDL is the EVAND LSB. These three bits are grouped together and explained below 0 Qualifier is active for all lines.* 1 Qualifier is EVAV. -- Any line during vertical active or blank will be output. 2 Qualifier is EVAV and VAV -- All lines during vertical blank (VAV==0) and all lines when EVAV is active during vertical active will be output. 3 Qualifier is VAV -- All lines during vertical active will be output. 4 Qualifier is EVAV and VAV -- All lines that EVAV is active during vertical active will be output. 5 Qualifier is EVAV, VAV and VBI_RAW_EN -- All lines as in option 4 plus the VBI RAW ADC lines. 6 Qualifier is EVAV, VAV, VBI_RAW_EN and VBI_SLC_EN-- All lines as in option 5 plus the VBI Sliced Lines. 7 All VBI sliced and VBI RAW Lines only. Enable VAV to be output to VS. 0 Output normal VS.* 1 VS has the same output as VAV Modified on May/04/2000 ELECTRONICS PAGE 70 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO VBI Decoder Control Index Mnemonic bit 7 0x20 VBICTL VBCVBS ODDOS[1:0] EVENEN ODDEN VBINSRT VYFMT[1:0] VBCVBS bit 6 bit 5 VYFMT[1:0] bit 4 bit 3 bit 2 VBINSRT ODDEN EVENEN bit 1 bit 0 ODDOS[1:0] Line offset for ODD field. See also VBIL[15:0]. 0 ODD field line offset is -1 compared to EVEN field.* 1 No offset. 2 ODD field line offset is 1 compared to EVEN field. 3 ODD field line offset is 2 compared to EVEN field. VBI data processing for EVEN field. 0 No processing.* 1 VBI processing is enabled for EVEN field. VBI data processing for ODD field. 0 No processing.* 1 VBI processing is enabled for ODD field. Enable VBI data to be output on the Y bus. 0 VBI data is not output on the Y bus.* 1 VBI data is output on the Y bus. When VBINSRT = 1, these bits control how VBI data are output on the Y bus. 0 1 bit on Y7 per CK2 clock.* 1 1 bit on Y7 plus a “1” on Y3 per CK2 clock. 2 4 bits on Y7..Y4, with first bit on Y7, last bit on Y4, plus a “1” on Y3 per CK2 clock. 3 8 bits on Y7..Y0, with first bit on Y7, last bit on Y0, per CK2 clock. Enable digitized CVBS data from ADC to be output for the selected VBI line instead of sliced VBI data. The new VBIMID bit allows simultaneous output (line by line bases) of sliced data and raw ADC data. 0 Output sliced VBI data for any line whose VBIL value ~= 0.* 1 Output digitized CVBS data for any line whose VBIL value ~=0. Modified on May/04/2000 ELECTRONICS PAGE 71 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO First Decoded Close-Caption Data Byte (Read Only) Index Mnemonic bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 0x21 CCDAT1 b0 b1 b2 b3 b4 b5 b6 P1 CCDAT1 This byte contains the first byte of the decoded close-caption data as defined in EIA-608. In order for this register to receive the CC data, VBINSRT must be programmed to a “1”, and VYFMT[1:0] must be programmed with the value 3. The same applies to CCDAT2. For normal NTSC Closed Caption decoding, ODDEN should be set to a “1”, VBIL12 should be programmed with the value 1. Second Decoded Close-Caption Data Byte (Read Only) Index Mnemonic bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 0x22 CCDAT2 b0 b1 b2 b3 b4 b5 b6 P2 CCDAT2 This byte contains the second byte of the decoded close-caption data as defined in EIA-608. VBI Data Decoding Index Mnemonic 0x23 VBIL30 VBIL3 VBIL2 VBIL1 VBIL0 0x24 VBIL74 VBIL7 VBIL6 VBIL5 VBIL4 0x25 VBIL118 VBIL11 VBIL10 VBIL9 VBIL8 0x26 VBIL1512 VBIL15 VBIL14 VBIL13 VBIL12 VBIL0..VBLI15 bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 These 16 2-bit numbers select how the chip should decode the VBI data for each VBI line. For 60 Hz video, VBIL1 through VBIL15 correspond to lines 10 through 24 in the ODD field, and lines 273 through 286 in the EVEN filed for NTSC (refer to NTSC line numbering convention). For 50 Hz video, VBIL1 corresponds to line 7 in the ODD field, and line 320 in the EVEN field. VBIL0 is used for all other lines not covered by VBIL1 through VBIL15. 0 Decode normal video.* 1 Decode Closed Caption data. 2 Decode Teletext data. 3 Decode WSS data. Modified on May/04/2000 ELECTRONICS PAGE 72 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Teletext Frame Alignment Pattern Index Mnemonic 0x27 TTFRAM TTFRAM[7:0] bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 bit 2 bit 1 bit 0 TTFRAM[7:0] User programmable Teletext frame alignment pattern. UV Offset Adjustment Index Mnemonic bit 7 bit 6 bit 5 bit 4 0x29 UVOFFH TSTCLC TSTCGN 0 TSTCFR 0x2A UVOFFL bit 3 UOFFST[5:4] UOFFST[3:0] VOFFST[5:4] VOFFST[3:0] VOFFST[5:0], UOFFST[5:0] These two 6-bit 2’s compliment values are for offset adjustment to the U and V components of the chroma data. The resolution is 1/4 LSB of the 8-bit U and V. TSTCFR Chroma frequency tracking control. TSTCGN TSTCLC 0 Chroma frequency tracking is enabled.* 1 Chroma frequency tracking is open loop. Chroma gain control. 0 Chroma gain tracks input.* 1 Chroma gain is controlled by SAT only. Cloche filter bypass. 0 Cloche filter is enabled for SECAM input.* 1 DC bypass of the cloche filter. Modified on May/04/2000 ELECTRONICS PAGE 73 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO U Component Gain Adjustment Index Mnemonic 0x2B UGAIN UGAIN[7:0] bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 bit 2 bit 1 bit 0 bit 2 bit 1 bit 0 VAVOD0 VAVEV0 UGAIN[7:0] U component gain adjustment. The nominal value is 0. V Component Gain Adjustment Index Mnemonic 0x2C VGAIN VGAIN[7:0] bit 7 bit 6 bit 5 bit 4 bit 3 VGAIN[7:0] V component gain adjustment. The nominal value is 0. VAV Begin Index Mnemonic 0x2D VAVB bit 7 bit 6 bit 5 bit 4 bit 3 VAVB[6:1] VAVEV0 The LSB for VAVB and VAVE for the even field. VAVOD0 The LSB for VAVB and VAVE for the odd field. VAVB[6:1] The 6 MSB’s of a 7-bit unsigned number which defines the start of VAV. The value “0” corresponds to line 4. VAV End Index Mnemonic 0x2E VAVE VAVE[8:1] bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 VAVE[8:1] The 8 MSB’s of a 9-bit unsigned number which defines the end of VAV. The value “0” corresponds to line 4. Modified on May/04/2000 ELECTRONICS PAGE 74 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Chroma Tracking Control Register Index Mnemonic bit 7 bit 6 0x2F CTRACK 0 0 CFTC[1:0] bit 5 bit 4 DMCTL[1:0] bit 3 bit 2 CGTC[1:0] bit 1 bit 0 CFTC[1:0] Chroma frequency tracking time constant. CGTC[1:0] 0 Slower.* 1 Slow. 2 Fast. 3 Faster. Chroma gain tracking time constant. DMCTL[1:0] 0 Slower.* 1 Slow. 2 Fast. 3 Faster. Chroma demodulation bypass mode. 0 Chroma demodulation is enabled.* 1 Chroma demodulation is bypassed for digital YCbCr input. 2 Chroma demodulation is bypassed for analog YCbCr input. Cb path is phase delayed by one half of CK2 clock period. 3 Chroma demodulation is bypassed for analog YCbCr input. Cr path is phase delayed by one half of CK2 clock period. Modified on May/04/2000 ELECTRONICS PAGE 75 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Timing Signal Polarity Control Index Mnemonic bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 0x30 POLCTL EVAVPL VSPL ODDPL HAVPL EHAVPL HS2PL VAVPL HS1PL HS1PL VAVPL HS2PL EHAVPL HAVPL ODDPL VSPL EVAVPL HS1 polarity. 0 Active high.* 1 Active low. VAV polarity. 0 Active high.* 1 Active low. HS2 polarity. 0 Active high.* 1 Active low.* EHAV polarity. 0 Active high.* 1 Active low. HAV polarity. 0 Active high.* 1 Active low. ODD polarity (this also affects the F bit in 656 code). 0 Active high.* 1 Active low. VS polarity (this also affect the V bit in 656 code). 0 Active high.* 1 Active low. EVAV polarity. 0 Active high.* 1 Active low. Modified on May/04/2000 ELECTRONICS PAGE 76 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Reference Code Insertion Control Index Mnemonic bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 0x31 REFCOD YCRANG 0 0 0 0 0 0 0 YCRANG Digital video output range control. 0 Y and C ranges are limited to 1 - 254; R, G, and B ranges are limited to 1 - 254.* 1 Y range is limited to 16 - 235; C range is limited to 16 - 240; R, G, and B ranges are limited to 16 - 240. Invalid Y Code Index Mnemonic 0x32 INVALY INVALY[7:0] bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 INVALY[7:0] User programmed code to be output for Y data when HAV is active but EHAV is inactive. Invalid U Code Index Mnemonic 0x33 INVALU INVALU[7:0] bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 INVALU[7:0] User programmed code to be output for U data when HAV is active but EHAV is inactive. Invalid V Code Index Mnemonic 0x34 INVALV INVALV[7:0] bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 INVALV[7:0] User programmed code to be output for V data when HAV is active but EHAV is inactive. Modified on May/04/2000 ELECTRONICS PAGE 77 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Unused Y Code Index Mnemonic 0x35 UNUSEY UNUSEY[7:0] bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 bit 1 bit 0 bit 1 bit 0 UNUSEY[7:0] User programmed code to be output for Y data when HAV is inactive. Unused U Code Index Mnemonic 0x36 UNUSEU UNUSEU[7:0] bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 UNUSEU[7:0] User programmed code to be output for U data when HAV is inactive. Unused V Code Index Mnemonic 0x37 UNUSEV UNUSEV[7:0] bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 UNUSEV[7:0] User programmed code to be output for V data when HAV is inactive. Modified on May/04/2000 ELECTRONICS PAGE 78 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Extra Control Bits for the S5D0127X01 Version Index Mnemonic bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 0x38 EXCTRL 0 ENINCST 0 - AUCPWD 0 0 CLEVEL CLEVEL AUCPWD ENINCST Programmable CKILL burst level select bit. 0 Burst peak level is 11 IRE.* 1 Burst peak level is 5.5 IRE. Auto chroma ADC power down mode enabled when appropriate input format is selected. 0 Chroma ADC powered down only during entire chip power down mode.* 1 Chroma ADC is power downed when CVBS input or case ‘0’condition. Scaler enable control bit during VBI. 0 Scaler on during VBI interval (defined by VAV).* 1 Scaler off during VBI interval. Modified on May/04/2000 ELECTRONICS PAGE 79 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Tracking Configuration Controls A Index Mnemonic 0x39 TRACKA AGCLSB VBCTRAP ATCTRAP VCR_LEV VCR_DET MAC_DET STCTRL bit 7 bit 6 bit 5 STCTRL MAC_DET VCR_DET bit 4 bit 3 VCR_LEV[1:0] bit 2 bit 1 bit 0 ATCTRAP VBCTRAP AGCLSB AGC LSB for control of the 9 bit AGC gain value. This bit only write to AGC when AGCFRZ is 1. 0 Write ‘0’to AGC 9 bit control LSB if AGCFRZ = 1.* 1 Write ‘1’to AGC 9 bit control LSB if AGCFRZ = 1. Chroma trap enabled during the VBI. 0 Chroma trap is controlled by CTRAP only.* 1 Chroma trap enabled during VBI. Auto Chroma Trap on luma path when VCR input is detected. 0 Chroma trap is controlled by CTRAP only.* 1 If VCR type input is detected, then CTRAP is enabled. Set the Fh variation from nominal for detection of VCR type input. 0 50 PPM.* 1 100 PPM. 2 200 PPM. 3 400 PPM. Status bit. Detect input that is not SCH locked such as consumer type VCR (Read only). 0 SCH locked video. 1 Color burst not locked to Fh (VCR). Status bit. Macrovision Encoded Data detected as input video source (Read only). 0 Standard video detected. 1 Macrovision Encoded data detected. State machine transition control. 0 Normal state machine transitions.* 1 Steady state sync level removed as condition for lock. Modified on May/04/2000 ELECTRONICS PAGE 80 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO VBI Control Register B Index Mnemonic bit 7 0x3A VBICTLB VBISWAP COFFENB YOFFENB CC_OVFL NEW_CC VBIMID TT_SYS VBISWAP bit 6 bit 5 TT_SYS[1:0] bit 4 VBIMID bit 3 bit 2 bit 1 bit 0 NEW_CC CC_OVFL YOFFENB COFFENB Disable control for the C-path clamp control. 0 C-path clamp works as normal.* 1 C-path clamp disabled. Disable control for the Y-path clamp control.* 0 Y-path clamp works as normal.* 1 Y-path clamp disabled. Defines when the current CCDAT1,2 data has over written previous data that was not read. (Read Only) 0 Current data has not generated an overflow condition. 1 Current data as written over data that was not read. Defines when new Closed Caption data is ready for reading from the CCDAT1,2 bytes. (Read Only) 0 Current data in CCDAT1,2 has already been read. 1 Current data in CCDAT1,2 is new. Changes function of WSS enable (per line bases during VBI) to a raw CVBS enable. 0 When VBIL(0-15) = 3, current line is enabled for WSS slicing.* 1 When VBIL(0-15) = 3, current line is enabled for raw ADC output. Select Teletext input system when auto detect is not possible. 0 Auto Teletext Select.* 1 Teletext System B. 2 Teletext System C. 3 Teletext System D. Reverse the bit order for data output from the closed caption or Teletext slicer. 0 Same as S5D0124D01 -- First bit sliced is located in MSB position.* 1 First bit sliced (in time) is located in LSB position. Modified on May/04/2000 ELECTRONICS PAGE 81 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Tracking Configuration Controls B Index Mnemonic bit 7 bit 6 0x3B TRACKB ALT656 VBI_PH AGC_LKG AGC_LPG VNOISCT PH_CTRL VBI_FR VBI_PH ALT656 bit 5 bit 4 bit 3 VBI_FR PH_CTRL VNOISCT bit 2 bit 1 bit 0 AGC_LPG[1:0] AGC_LKG AGC gain tracking loop time constant for initial tracking mode. 0 Same as steady state time constant.* 1 2X faster than selected steady state time constant. AGC gain steady state tracking loop time constant. 0 Fastest.* 1 Fast. 2 Slow. 3 Slowest. Vertical sync noise control enable. 0 Vertical sync adjusts with all sync phase changes.* 1 Vertical sync large phase errors must occur for 4 lines to activate a phase change. Controls phase detector response. 0 Syncs after the “0” point reference have priority.* 1 Syncs prior to “0” point reference have priority. Disables frequency compensation for VCR head switch lines only. 0 Frequency tracking independent of this control.* 1 Frequency tracking disabled for VCR head switch lines. Enables phase compensation for VCR head switch lines only. 0 Phase tracking independent of this control.* 1 Phase tracking enabled for VCR head switch lines only. Alternate 656 Vertical blank location for 50 Hz video. 0 Vertical blank size per the ITU 656 Specification (ends at 656 digital line 23).* 1 Vertical blank size same as 60 Hz (ends at 656 digital line (50 Hz) 6). Modified on May/04/2000 ELECTRONICS PAGE 82 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO RTC Genlock output signal control Index Mnemonic 0x3C RTC TDMOD RTC_PID RTC_DTO bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 0 TDMOD 0 0 0 0 RTC_DTO RTC_PID Test bit for chroma demodulation mode. 0 Normal operation.* 1 Test mode. Polarity control for PAL ID transferred within the RTC data stream. 0 Same polarity as default PID pin.* 1 Inverted polarity. Enables a DTO reset inside the S5D0127X01 and sends a DTO reset within the RTC data stream. Function is activated on the rising edge of RTC_DTO. 0 Function disabled.* 1 Function enabled one time when set to 1. Modified on May/04/2000 ELECTRONICS PAGE 83 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Command Register E Index Mnemonic bit 7 bit 6 0x3D CMDE ODFST VSALG CHIPREVID HCORE VSALG ODFST bit 5 bit 4 bit 3 HCORE[1:0] bit 2 bit 1 bit 0 CHIPREVID Four additional bits for determination of current Revision and differentiation from the S5D0124D01. 0 S5D0124D01. 9 S5D0127X01 Revision A.* Luma path horizontal coring. Noise limiter for high frequency portion of luma. 0 Coring function is disabled.* 1 1 bit of coring. 2 2 bits of coring. 3 4 bits of coring. Vertical scaling line dropping algorithm. 0 Vertical scaling drops the same lines in the Odd and Even fields -- good for fast motion video.* 1 Vertical scaling drops lines based on the final de-interlaced video. This is a better vertical scaling but may be sensitive to fast motion video. Alternate the first scaling line between Odd and Even fields. 0 Even field is the first scaled field.* 1 Odd field is the first scaled field. Modified on May/04/2000 ELECTRONICS PAGE 84 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO VS Delay Control Index Mnemonic bit 7 bit 6 0x3E VSDEL TR_MS NOVIDC bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 VSDEL[5:0] VSDEL[5:0] When the chip is programmed for digital video input operation, this register provides an offset for the internal line counter to align with input video (VS can be either from the VS pin or from embedded timing code). The register content is unsigned. NOVIDC Allows NOVID bit to be output to PORTB (pin 24). TR_MS 0 Normal operation.* 1 The NOVID bit is output to PORTB if DATAB[2:0]=1 and DIRB=1. Enable alternative initial tracking mode state machine. 0 Normal operation - Horizontal tracking mode is controlled by the HFSEL[1:0] bits.* 1 Variable tracking modes during locking time. Modified on May/04/2000 ELECTRONICS PAGE 85 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Command Register F Index Mnemonic 0x3F CMDF CBWI bit 7 bit 6 CTRAPFSC VIPMODE bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 EVAVY UVDLEN UVDLSL REGUD TASKB CBWI Chroma bandwidth increase. This function should be used for digital video input mode only. TASKB 0 Normal chroma bandwidth.* 1 Increased chroma bandwidth. Select between task A and B as described in “VIP Specification V. 1.0”. REGUD 0 Select CCIR 656 timing codes (T-bit is always 1).* 1 Select between task A and B when VBI data is output. If active video is output, T-bit is set to 1(task A). If VBI data is output, T-bit is set to 0 (task B). Control register update control. UVDLSL 0 Registers are updated immediately after being written to.* 1 The following registers and register bits are updated only during the start of vertical sync after they are written to: Index 0x02, indices 0x17 through 0x1D, bit 0 of index 0x04, bits [2:0] and [6:4] of index 0x0E. U or V delay control when UVDLEN is set to 1. UVDLEN 0 V is delayed by 1 CK period.* 1 U is delayed by 1 CK period. Enable the function of UVDLSL. EVAVY 0 UVDLSL is disabled.* 1 UVDLSL is enabled. Control the output of INVALY, INVALU, and INVALV codes when EVAV is inactive. VIPMODE 0 Output of these codes are not affected by EVAV.* 1 These codes are output when EVAV is inactive (line is being dropped by the vertical scaler). Allows transfer of hardware sliced VBI data as ancillary data during the following line’s horizontal blanking period. CTRAPFSC 0 Standard S5D0124D01 original sliced VBI data transfer.* 1 Optional ancillary sliced VBI data transfer. Enable chroma trap location based on Fsc frequency instead of field rate. 0 Chroma trap based on field rate (same as S5D0124D01).* 1 Chroma trap based on detected Fsc frequency. Modified on May/04/2000 ELECTRONICS PAGE 86 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Gamma Base Index Mnemonic 0x40 GAMMA0 GAMMA0[7:0] 0x41 GAMMA1 GAMMA1[7:0] : : : : : : 0x5F GAMMA31 GAMMA31[7:0] GAMMA0 -GAMMA31 bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 Gamma correction base. The desired output for 8*N, where N = 0, .., 31, is programmed into GAMMAN. Note that data written into these addresses are simultaneously written into addresses 0xC0 through 0xDF. Gamma Correction Delta Index Mnemonic bit 7 bit 6 0x60 GAMMAD0 - - GAMMAD0[5:0] 0x61 GAMMAD1 - - GAMMAD1[5:0] : : : : : : : : : : 0x7F GAMMAD31 - - GAMMAD31[5:0] GAMMAD0 ..GAMMAD31 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 The Nth location of the 32 locations is programmed with a 6-bit unsigned number which represents the gamma correction delta for the gamma bases N and N + 1. The last location will contain the gamma correction delta for gamma base 31 and presumed base 32 which has the value of 256. Note that data written into these addresses are simultaneously written into addresses 0xE0 through 0xFF. Modified on May/04/2000 ELECTRONICS PAGE 87 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO U/V Gamma Base Index Mnemonic 0xC0 GAMUV0 GAMUV0[7:0] 0xC1 GAMUV1 GAMUV1[7:0] : : : : : : 0xDF GAMUV31 GAMUV31[7:0] GAMUV0 -GAMUV31 bit 7 bit 6 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 U and V gamma correction base. The desired output for 8*N, where N = 0, .., 31, is programmed into GAMUVN. U/V Gamma Correction Delta Index Mnemonic bit 7 bit 6 0xE0 GAMUVD0 - - GAMUVD0[5:0] 0xE1 GAMUVD1 - - GAMUVD1[5:0] : : : : : : : : : : 0xFF GAMUVD31 - - GAMUVD31[5:0] GAMUVD0 ..GAMUVD31 bit 5 bit 4 bit 3 bit 2 bit 1 bit 0 U and V gamma correction delta. The Nth location of the 32 locations is programmed with a 6-bit unsigned number which represents the gamma correction delta for the gamma bases N and N + 1. The last location will contain the gamma correction delta for gamma base 31 and presumed base 32 which has the value of 256. Modified on May/04/2000 ELECTRONICS PAGE 88 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO ABSOLUTE MAXIMUM RATINGS Characteristics Symbol Value Units 5-V supply voltage (measured to VSS) VDD -0.5 to + 7.0 V 3.3-V supply voltage (measured to VSS) VDD3 -0.5 to + 4.5 V Voltage on any digital pin VPIN -0.5 to (VDD+0.5) V Ambient operating temperature (case) TA -35 to + 100 °C Storage temperature TS -65 to + 150 °C Junction temperature TJ 150 °C Tvsol 220 °C Vapor phase soldering (1 min.) Notes: 1.Absolute maximum ratings are limiting values applied individually while all other parameters are within specified operating conditions. 2.Functional operation under any of these conditions is not implied. 3.Applied voltage must be current limited to a specified range. OPERATING CONDITIONS Characteristics Symbol Min Typ Max Units 5-V supply voltage (measured to VSS) VDD5 4.75 5.0 5.25 V 3.3-V supply voltage (measured to VSS) VDD3 3.0 3.3 3.6 V TA -20 70 °C Ambient operating temperature, still air Modified on May/04/2000 ELECTRONICS PAGE 89 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO ELECTRICAL CHARACTERISTICS Characteristics Symbol Min Typ Max Units Supply +5V (VDD, VDDA, VDDA1), normal operation IDD5 150 mA +3.3V (VDD3), normal operation IDD3 141 mA +5V (VDD, VDDA, VDDA1), power down mode IDD5 90 mA +3.3V (VDD3), power down mode 5 mA IDD3 Analog Characteristics Integral linearity error (AGC/ADC only) EI-ADC 1.0 3.0 lsb Differential linearity error (AGC/ADC only) ED-ADC 0.5 1.5 lsb Total harmonic distortion (4 MHz full scale) THD 54 dB Signal to noise ratio (4 MHz full scale) SNR 42 dB Analog bandwidth (50 IRE to 3 dB point) BW Input voltage range (peak-peak) 100 IRE input Input resistance AY0-AY2,AC0-AC2 VI(PP) RIN Input capacitance for analog video inputs CIN 10 pF Charge current for offset control IOFF ±4 µA Cross talk between analog inputs a -42 FLUMA 1.5 dB Differential gain - complete chip (Modulated 40 DG IRE ramp) 1.5 % Differential phase - complete chip (Modulated 40 IRE ramp) 1.0 degree 800/500 kHz 4 MHz 0.5 1.5 200 Vpp kΩ -50 dB Video Performance Luminance frequency response (maximum variation to 4.2 MHz - multi burst) DP Chrominance frequency response (3 dB point) - FCHROMA CBWR=0/1 Chroma nonlinear gain distortion (NTC-7 Combination) CNGD 1 % Chroma nonlinear phase distortion (NTC-7 Combination) CNPD 1.25 degree Chroma to luma intermodulation (NTC-7 Combination) CLI 1 IRE Chroma luma gain equality (NTC-7 Composite) DELCL ±20 ns Chroma luma delay equality (NTC-7 Composite) AMPCL 98-101 % Noise level for unified weighting 10 kHz-5 MHz NLUMA (100 IRE unmodulated ramp) -58 dB Chroma AM noise (red field) NCAM -60 dB Chroma PM noise (red field) NCPM -54 dB Modified on May/04/2000 ELECTRONICS PAGE 90 OF 96 S5D0127X01 Data Sheet Characteristics MULTIMEDIA VIDEO Symbol Min Typ Max Units Digital I/O Characteristics Input low voltage (other digital I/O) VIL VSS-0.5 0.8 V Input high voltage (other digital I/O) VIH 2.0 VDD+0.5 V Input low voltage (SCLK,SDAT,RST) VILI2C VSS-0.5 0.3VDD V Input high voltage (SCLK,SDAT,RST) VIHI2C 0.7VDD VDD+0.5 V Input low current (VIN = 0.4 V) IIL -1 µA Input high current(VIN=2.4) IIH -1 µA Digital output low voltage (IOL=3.2mA) VOL 0.4 V Digital output high voltage (IOH=400µA) VOH Digital three-state current IOZ 50 µA Digital output capacitance COUT 7 pF Maximum capacitance load for digital data pins CL-DATA 30 pF Maximum capacitance load for CK and CK2 outputs 60 pF 2.4 V CL-CK Timing Characteristics - Digital Inputs XTALI input pulse width low tpwlX 15 20 ns XTALI input pulse width high tpwhX 15 20 ns 120 CK Clock and Data Timing Analog video input to digital video output delay tdCHIP Pulse width high for CK (KS0112 operates at frequencies from 24.5 MHz to 29 MHz) tpwhCK 15 18.5 22 ns Pulse width high for CK2 tpwhCK2 30 37 44 ns Delay from rising edge of CK to CK2 tCK2 4 Delay from rising edge CK to data change (including pins Y0-Y7, C0-C7, HAV, VAV, EHAV, EVAV, HS1, HS2, VS, ODD, PID, SCH) tdD 16 23 ns 14 21 ns ns (CK is output) tdD (CK is input) Minimum hold time from rising edge of CK for data output) thD 7 ns Delay from falling edge of OEN to data bits in 3-state tzD 20 ns Delay from rising edge OEN to data bits enabled tenD 18 ns 400 kHz 400 pF Timing Characteristics -IIC Host Interface SCLK clock frequency rSCLK 0 Capacitive load for each bus line Cb Hold time for START condition thSTA 0.6 µs Setup time for STOP condition tsSTO 0.6 µs Rise and fall times for SCLK and SDAT tR, tF 20 Modified on May/04/2000 ELECTRONICS 300 ns PAGE 91 OF 96 S5D0127X01 Data Sheet Characteristics MULTIMEDIA VIDEO Symbol Min Typ Max Units SCLK minimum pulse width low tpwlSCLK 1.3 µs SCLK minimum pulse width high tpwhSCLK 0.6 µs SDAT setup time to rising edge of SCLK tsSDAT thSDAT 100 ns 0 ns SDAT hold time from rising edge of SCLK Note: AC/DC characteristics provided are per design specifications. Modified on May/04/2000 ELECTRONICS PAGE 92 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO tpwhCK CK tpwhCK2 CK2 tCK2 OEN thD Y,C,AV VS,HS,ODD SCH,PID tdD tzD Figure 41. tenD Data Output Analog Video Input Digital Video Output Blank Active video Active video tdCHIP Figure 42. Analog Video Input to Digital Video Output Delay SDAT tBUF thSDAT tsSTO tR tpwhSCLK SCLK thSDA tpwlSCLK Figure 43. tsSDAT tF IIC Host Interface Detailed Timing Modified on May/04/2000 ELECTRONICS PAGE 93 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO VCC FE. BEAD 67 90 AC0 92 AC1 94 AC2 0.1 µF 75 77 VRT S5D0127X01 78 VRB 96 TEST 97 COMP2 0.1 µF 8 24.576 MHz 7 XTALI XTALO 22 pF SCLK SDAT AEX0 AEX1 44 39 38 37 36 35 34 33 HS1 HS2 VS HAV EHAV VAV EVAV ODD PID SCH OEN CK CK2 26 76 23 25 5 3 4 22 17 24 15 18 21 VSS 57 HS2 20 k 95 VSS 91 VSS VSS 87 83 VSS VSS 65 64 VSS VSS 60 VSS 41 40 VSS VSS 19 14 VSS VSS Modified on May/04/2000 ELECTRONICS C7 C6 C5 C4 C3 C2 C1 C0 TESTEN 6 75 72 69 70 13 Host Interface 22 pF 56 55 54 53 48 47 46 45 To Frame Buffer VDD3 66 VDD VDD3 59 43 VDD3 42 VDD VDD3 20 12 VDD3 VDD3 11 98 VDDA 93 VDDA 88 AY2 Y7 Y6 Y5 Y4 Y3 Y2 Y1 Y0 To Frame Buffer Controller 86 AY1 VDDA 84 AY0 Analog Video Input VDDA VDDA1 9 85 10 µF 89 0.1 µF 3.3V PAGE 94 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO Package Dimension 100-QFP-1420D 23.90±0.30 0~ 8× 20.00±0.20 0.15 +0.1 0 14.00±0.20 0.10MAX (0.83) 17.90±0.30 -0.05 #100 #1 0.65 0.30±0.10 (0.58) 0.10MAX 0.00MIN 2.65±0.10 3.00MAX 0.80±0.20 Dimensions are in Millimeters Modified on May/04/2000 ELECTRONICS PAGE 95 OF 96 S5D0127X01 Data Sheet MULTIMEDIA VIDEO SAMSUNG SEMICONDUCTOR WORLDWIDE OFFICES HEAD OFFICE SAMSUNG ELECTRONICS JAPAN CO., LTD. 8/11FL., SAMUNG MAIN BLDG. 250, 2-KA, TAEPYUNG-RO, CHUNG-KU, SEOUL, KOREA TEL: 2-727-7114 FAX: 2-753-0967 HAMACHO CENTER BLDG. 31-1, NIHONBASHI-HAMACHO, 2-CHOME CHUO-KU, TOKYO 103, JAPAN TEL: 3-5641-9850 FAX: 3-5641-9851 SEMICONDUCTOR BUSINESS SALES & MARKETING DIVISION SAMSUNG ELECTRONICS HONG KONG CO., LTD. 15/16FL., SEVERANCE BLDG. 84-11, 5-KA, NAMDAEMOON-RO CHUNG-KU, SEOUL, KOREA TEL: 2-259-1114 FAX: 2-259-2468 65TH FL., CENTRAL PLAZA 18 HARBOUR ROAD WANCHAI, HONG KONG TEL: 852-2862-6900 FAX: 852-2866-1343 SAMSUNG ELECTRONICS TAIWAN CO., LTD. SAMSUNG SEMICONDUCTOR INC. 3655 NORTH FIRST STREET SAN JOSE, CA 95134, U.S.A. TEL: 408-954-7000 FAX: 408-954-7873 30FL., NO.333 KEELUNG RD. SEC 1, TAIPEI, TAIWAN, R.O.C TEL: 886-2-757-9292 FAX: 886-2-757-7311 SAMSUNG ASIA PRT., LTD. SAMSUNG SEMICONDUCTOR EUROPE GMBH SAMSUNG HOUSE AM KRONBERGER HANG 6 65824, SCHWALBACH/TS TEL: 49-6196-663300 FAX: 49-6196-663311 SAMSUNG SEMICONDUCTOR EUROPE LTD. GREAT WEST HOUSE GREAT WEST ROAD, BRENTFORD MIDDLESEX TW8 9DQ TEL: 181-380-7132 FAX: 181-380-7220 80 ROBINSON RD., #20-01 SINGAPORE 068898 TEL: 65-535-2808 FAX: 65-227-2792 SAMSUNG ELECTRONICS CO., LTD. SHANGHAI OFFICE 9F, SHANGHAI INTERNATIONAL TRADE CENTRE NO.2200 YANAN(W) RD. SHANGHAI, P.R.C. 200335 TEL: 8621-6270-4168 FAX: 8621-6275-2975 SAMSUNG ELECTRONICS CO., LTD. SEMICONDUCTOR BUSINESS BEIJING OFFICE 15FL., BRIGHT CHINA CHANG AN BLDG., NO.7, JIANGUOMEN, NEI AVENUE BEIJING, CHINA 100005 TEL: 8610-6510-1234(0) . Circuit diagrams utilizing SAMSUNG and/or SAMSUNG ELECTRONICS products are included as a means of illustrating typical semiconductor applications; consequently, complete information sufficient for construction purposes is not necessarily given. The information has been carefully checked and is believed to be entirely reliable. However, no responsibility is assumed for inaccuracies. Furthermore, such information does not convey to the purchaser of the semiconductor devices described herein any license under the patent rights of SAMSUNG and/or SAMSUNG ELECTRONICS, or others. SAMSUNG and/or SAMSUNG ELECTRONICS, reserve the right to change device specifications. LIFE SUPPORT APPLICATIONS SAMSUNG and/or SAMSUNG ELECTRONICS products are not designed for use in life support applications, devices, or systems where malfunction of a SAMSUNG product can reasonably be expected to result in a personal injury. SAMSUNG and/or SAMSUNG ELECTRONICS’customers using or selling SAMSUNG and/or SAMSUNG ELECTRONICS products for use in such applications do so at their own risk and agree to fully indemnify SAMSUNG and/or SAMSUNG ELECTRONICS for any damages resulting from such improper use or sale. Modified on May/04/2000 ELECTRONICS PAGE 96 OF 96