SDA 9489X ’PIP IV Advanced’ and SDA 9589X
’SOPHISTICUS’ belong to a new generation of Picturein-Picture (PiP) processors that combine high-quality
digital PIP signal processing, digital multistandard
color decoding and AD/DA conversion on a single chip.
Both devices are equipped with CVBS and Y/C input
interfaces. In addition the SDA SDA 9589X is also able
P-DSO28-1
to process YUV input signals for displaying high quality
video signals e.g. coming from a DVD source.
Figure 0-1Picture-In-Picture
The integrated digital color decoder is able to decode all analog TV standards (PAL,
NTSC and SECAM) and detects the standard automatically. Therefore the IC is suited
for world-wide use.
A picture reduction from 1/4 to 1/81 of original size selectable in fine steps is possible.
The transfer functions of the decimation filters are optimally matched to the selected
picture size reduction and can furthermore be adjusted to the viewer’s requirements by
a selectable peaking. A maximum of 324 luminance and 2x81 chrominance pixels per
line are stored in the memory. The PiP supports split-screen applications as well as
multi-PiP display.
• Single chip solution:
– AD-conversion for CVBS or Y/C or YUV
1)
, multistandard color decoding, PLL for
synchronization of inset channel, decimation filtering, embedded memory, RGBmatrix, DA-conversion, RGB/YUV switch, data-slicer and clock generation
integrated on chip
• Analog inputs:
– 3x CVBS or 1x CVBS and 1x Y/C or 1xYUV
1)
alternatively
– Clamping of each input
– All ADCs with 8 bit amplitude resolution
– Automatic Gain Control (AGC) for Y and CVBS
• Inset Synchronization:
– Multiple time constants for reliable synchronization
– Automatic recognition of 625 lines / 525 lines standard
• Color Decoder:
– PAL-B/G, PAL-M, PAL-N(Argentina), PAL60, NTSC-M, NTSC4.4 and SECAM
– Adjustable color saturation
– Hue control for NTSC
– Automatic Chroma Control (-24 dB ... +6 dB)
– Automatic recognition of chroma standards: different search strategies selectable
– Single crystal for all standards
– IF-characteristic compensation filter
• Decimation:
– PIP sizes between 1/81 and 1/4 adjustable with steps of 2 lines and 4 pixel
– Resolution up to 324 luminance and 2x81 chrominance pixels per inset line
– Horizontal and vertical filtering dependent on picture size
– Automatic zoom in/out possible with three speeds
• Display Features:
– 7 bit per pixel stored in memory
– Field and joint-line free frame mode display (even at 100/120 Hz AABB with picture
sizes<=1/9)
– Two ’split-screen’ modes with horizontal decimation of 2 and vertical of 1.5 or 1.0
–POP display
– Up to 12 pictures of 1/36th size (11 still and 1 moving)
– Up to 6 pictures of 1/16th size (5 still and 1 moving)
– Up to 3 pictures of 1/9th size (2 still and 1 moving)
– Display on VGA and SVGA screen (f
limited to 40kHz)
H
– 8 different read frequencies for 16:9 compatibility
– Line doubling mode for progressive scan applications
1)
SDA 9589X only
Micronas1-5
SDA 9489X
SDA 9589X
Preliminary Data Sheet
– Freeze picture
– Coarse positioning at 4 corners of the parent picture
– Fine positioning at steps of 4 pixels and 2 lines
– Wipe in / out programmable with 3 time periods
• Output signal processing:
–7 Bit DAC
– RGB or YUV switch: insertion of an external source without PIP processing
– Digital interpolation for anti-imaging
– Adjustable transient improvement for luma (peaking)
– Contrast, Brightness and Pedestal Level adjustable
– Analog outputs: Y, +(B-Y), +(R-Y), or Y, -(B-Y), -(R-Y) or RGB
– Three RGB matrices available: NTSC(Japan), NTSC(USA) or EBU
– 64 different background colors and 4096 different frame colors
– Plain or 3D frame with variable width and height
• Data Slicing:
– Slicing of closed-caption (CC) or wide-screen-signaling (WSS) data
– Violence blocking capability (V-chip)
– Several filter for XDS data extraction
• On-screen display:
– 64 characters programmable
– 5 characters displayed in every PIP picture or 3 rows of 20 characters each
– 4 different character luminance values or frame color
– 4 background luminance values or (semi-) transparent mode
2
•I
C-Bus control (400 kHz)
• High stability clock generation
• PDSO 28-1 package (SMD)
• Full SDA 9488X and SDA 9588X backward compatibility
• SDA 9388X / SDA 9389X pinout compatibility
• 3.3V supply voltage (5V input capable)
Features
Micronas 1-6
SDA 9489X
SDA 9589X
Preliminary Data Sheet
2Pin Configuration
XIN
XQ
HSP
VSP
SDA
SCL
VDD
VSS
I2C
INT
IN1
IN2
IN3
FSW
Pin Configuration
1
2
3
4
5
6
7
8
9
10
11
12
13
14
PDSO 28 -1
28
27
26
25
24
23
22
21
20
19
18
17
16
15
CVBS1
VREFM
CVBS2
VREFL
CVBS3
VSSA1
VDDA1
VREFH
VSSA2
VDDA2
OUT1
OUT2
OUT3
SEL
Figure 2-1Pinning
Figure 2-2Package Outlines
Micronas2-7
SDA 9489X
SDA 9589X
Preliminary Data Sheet
Numb
er
1XINIcrystal oscillator (input) or external clock input
2XQOcrystal oscillator (output)
3HSPI/TTLhorizontal sync for parent channel
4VSPI/TTLvertical sync for parent channel
5SDAI/OI
6SCL II
7VDDSdigital supply voltage
8VSSSdigital ground
9I2C II
10INTO/TTLinterrupt
11IN1I/anaV/R input for external YUV/RGB source
NameTypeDescription
2
C-bus data
2
C-bus clock
2
C Address
Pin Configuration
12IN2I/anaY/G input for external YUV/RGB source
13IN3I/anaU/B input for external YUV/RGB source
14FSWIfast switch input for YUV/RGB switch
15SELOfast blanking output for PIP
16OUT3O/anaanalog output: chrominance signal +(B-Y) or -(B-Y) or B
17OUT2O/anaanalog output: luminance signal Y or G
18OUT1O/anaanalog output: chrominance signal +(R-Y) or -(R-Y) or R
19VDDA2Sanalog supply voltage for DAC
20VSSA2Sanalog ground for DAC
21VREFHI/anauppper reference voltage for ADC and DAC
22VDDA1Sanalog supply voltage for ADC
23VSSA1Sanalog ground for ADC
24CVBS3I/anaCVBS3 or V (SDA 9589X) or C Input
25VREFLI/Olower reference voltage for ADC
26CVBS2I/anaCVBS2 or U (SDA 9589X) or Y (from Y/C) Input
27VREFMI/Omid-level reference voltage for ADC
28CVBS1I/anaCVBS1 or Y (from YUV, SDA 9589X) Input
An analog inset CVBS signal can be fed to the inputs CVBS1-3 of SDA 9589X/SDA
9489X. Each of these sources is selectable via I
2
C bus (CVBSEL). CVBS2 and CVBS3
can be used as separate Y/C inputs. At SDA 9589X YUV sources can be connected to
CVBS1, CVBS2 and CVBS3 provided YUV operation being enabled (YUVSEL). Using
an external switch SDA 9589X can operate in applications with both YUV and CVBS
signals.
CVBSELYUVSELInputremark
D1D0
CVBS1CVBS2CVBS3
000CVBS
010CVBS
100Y (VBS)CY/C mode
110CVBS
XX1Y (VBS)U (CB)V (CR)YUV mode
(only SDA 9589X)
Table 4-1Input selection
4.1.2AD-Conversion
All signal are clamped and AD-converted with an amplitude resolution of 8bit. CVBS and
Y signals are clamped to the sync bottom whereas U/V and C signals are clamped to
their mid-level during blanking.
Inset
Video
HD
CLMPIST
CLAMPI
CLMPID
Figure 4-1Clamping timing
Micronas 4-10
SDA 9489X
SDA 9589X
Preliminary Data Sheet
System Description
The clamping pulse can be shifted in position (CLMPIST) and length (CLMPID) to adjust
to the specific application. The ADCs are driven by a 20.25 MHz free running crystal
clock which is not related to the incoming CVBS signal.
To avoid aliasing by subsampling the CVBS signal and the Y/C signals should be bandlimited to 10MHz. In the same manner the U/V signal frequency spectrum should not
exceed 5 MHz. The digital filtering suppresses all frequencies above the usable
spectrum.
4.1.3Automatic Gain Control
To accommodate to different CVBS input voltages an automatic gain control has been
implemented. The chip works correctly for input voltages in the range from 0.5 to 1.5V
pp
For best signal-to-noise ratio, the maximum CVBS amplitude is recommended if
available. The AGC behavior can be chosen out of four possibilities (AGCMDE).
The sync height serves as reference for the gain control in the typical application. When
using overflow detection only, the gain is set to maximum and is reduced whenever an
overflow occurs. This procedure will be executed again when a channel change is
detected or the gain control is manually reset by AGCRES.
.
2
1.5
1
Input Voltage [V]
0.5
0
0246810121416
Automatic Gain Control Characteristic
AGCVAL
Figure 4-2AGC characteristic
4.1.4Signal Magnitudes
The nominal CVBS signal with 75% color has a magnitude of 1 V
is left to permit signals with 100% color resulting in 1.23 V
pp
. The upper headroom
pp
. The Y signal must always
contain the sync part. Its levels correspond to the CVBS levels except for the missing
color and burst. After A/D conversion the video part is clamped to its black value and is
amplified to 224 digital steps. The nominal signal levels ensure correct brightness and
saturation. The YUV signal levels conform to the ITU 601 recommendation.
Micronas4-11
SDA 9489X
SDA 9589X
Preliminary Data Sheet
CRYC = 1.2 Vpp
255
224
128
32
burst
0
255
217
68
upper headroom
white
black
burst
4
0
lower headroom
SRY = 1 Vpp
Figure 4-3CVBS/Y and chroma ADC input signal range
255
240
212
255
240
212
System Description
upper headroom
75% chroma
lower headroom
upper headroom upper headroom
100% chroma
SRC = 0.89 Vpp
CRYC = 1.2 Vpp
75% U
128
SRUV = 0.7 Vpp
44
16
0
lower headroom
Figure 4-4UV input signal range
AGCVALConversion
D3D2D1D0
Range
CRYC
0000 0.5Vpp0.42V
.........
1000 1.2V
pp
.........
CRUV = 0.8 Vpp
Signal
Range
SRY
1.0V
pp
128
44
16
pp
75% V
CRUV = 0.8 Vpp
SRUV = 0.7 Vpp
0
Signal
Range
SRC
0.89V
lower headroom
pp
Conversion
Range
CRUV
0.8V
pp
Signal
Range
SRUV
0.7V
pp
1111 1.5V
pp
1.25V
pp
Table 4-2ADC conversion range and required input signal voltage
Micronas 4-12
SDA 9489X
SDA 9589X
Preliminary Data Sheet
System Description
4.2Inset Synchronization
Horizontal and vertical sync pulses are separated after elimination of the high frequency
components of the CVBS signal by a low pass filter. Horizontal sync pulses are
generated by a digital phase-locked-loop (DPLL). Its time constant is adjustable between
fast and slow behavior in four steps (PLLITC) to consider different input sources (e.g.
VCR). Noisy input signals become more stable when a noise-reduction is enabled
(NSRED). Additionally weak input signals from a satellite dish (’fishes’) become more
stable when SATNR is enabled. Both should be enabled to have best available
performance. When NOSIGB is enabled, a colored background is shown instead of the
picture when PIP is out of synchronization. The detected line standard is indicated by
SYNCSTAT.
4.3Chroma Decoding And Standard Identification
The system is able to decode NTSC and PAL signals with a subcarrier of 3.58MHz and
4.43MHz (PAL B/M/N/60, NTSC M/4.4) as well as SECAM signals with 4.05/4.2MHz
subcarrier. The system may be forced to a certain standard, or an automatic standard
detection can be used (CSTAND). For automatic standard detection, some standards
which are not likely to be received can be ignored to improve the detection process.
Depending on the detected line standard (525 or 625 lines) the color standard detection
circuit searches for 60 Hz signals (NTSC-M / PAL-M / PAL 60 / NTSC44) or 50 Hz signals
(PAL-B / SECAM / PAL-N) respectively. Within each line standard, the standard is
detected by consequently switching from one to another. This standard detection
process can be set to medium or fast behavior (LOCKSP). In medium behavior 30 fields
(in fast 20) are used to detect the standard. If not being successful within this time period
the system tries to detect another one. For SECAM detection, a choice between two
recognition levels is possible (SCMIDL) and the evaluated burst position is selectable
(BGPOS).
.
CSTANDEXNTSC-
D1D0
M
PAL60PAL-NPAL-MPAL-BSECAMNTSC
44
00
0.
1
10
11
Table 4-3Considered color standards for automatic standard detection
For getting the chrominance information the digitized video signal is multiplied with the
regenerated color subcarrier once in-phase and once phase-shifted by 90°. After
lowpass filtering digital UV is available for PAL and NTSC. The subcarrier is regenerated
Micronas4-13
SDA 9489X
SDA 9589X
Preliminary Data Sheet
System Description
by a digital PLL. At SECAM operation the PLL runs free and generates the line-wise
alternating subcarriers. A CORDIC structure demodulates the frequency-modulated UV
signals. The following SECAM de-emphasis filter characteristic is adjustable (DEEMP).
The chroma signal can be filtered before demodulation by means of a selectable IFprefilter (IFCOMP).
0
5
10
gain [dB]
15
DEEMP = ’00’
DEEMP = ’01’
DEEMP = ’10’
DEEMP = ’11’
5
2.5
IFCOMP = ’00’
0
IFCOMP = ’01’
2.5
gain [dB]
IFCOMP = ’10’
5
7.5
3.584.4
20
00.511.522.5
frequency [MHz]
10
23456
frequency [MHz]
Figure 4-5SECAM de-emphasis filter characteristic and IF-compensation filter
characteristic
The Hue Control (HUE) influences the phase of the demodulation subcarrier between
-44.8° and 43.4° in steps of 1.4°. This is provided for NTSC only and adjustment is
ineffective for PAL and SECAM signals.
The reference for the subcarrier generation is a crystal stable clock of 20.25000 MHz. In
order to avoid color standard detection problems, the maximum deviation of this
frequency should not exceed 100ppm. For a good PLL locking behavior a maximum
deviation of 40ppm is recommended. A small frequency adjustment (-150 ... +310 ppm)
is possible for using a crystal with small frequency deviations (SCADJ). For test
purposes, CPLL allows to open the loop of the chroma PLL.
For deviations in the chroma signal up to 30dB, a stable output amplitude after chroma
decoding is achieved due to the ACC (Automatic Chroma Control). If the chroma signal
(color burst) is below a selectable threshold (CKILL), the color will be switched off.
Alternatively the color-killer can be bypassed and the color can be switched on or off
under all conditions (COLON). By setting ACCFIX, the automatic chroma control is
disabled and set to a default value.
Micronas 4-14
SDA 9489X
SDA 9589X
Preliminary Data Sheet
System Description
CKILLCOLONcolor killed at damping of
D1D0
00030 dB
01018 dB
11024 dB
110color always off
XX1color always on
Table 4-4Color-killer adjustment
The bandwidth of the chroma filter is adjustable via CHRBW. The bandwidth depends
on whether the decoder is in SECAM operation or not. A change in CHRBW does not
result in a chrominance position shift on the screen.
CKSTAT can be read out and gives information whether the color is switched on or off.
STDET indicates the detected color standard. Additionally PALID signals whether a PAL
signal or a NTSC signal is applied.
4.4Comb Filtering
Depending on the selected picture size and color standard, a comb filtering is performed
for luminance and chrominance. A comb filter uses the spectral interleaving of the
encoded luminance and chrominance to separate both without cross artifacts. Thus
cross-color and cross-luminance are suppressed effectively. For NTSC sources, a comb
filtering is performed for all picture sizes. Due to reduced bandwidth in horizontal and
vertical direction a strong reduction of cross artifacts can be achieved for PAL signals.
The same applies for the luminance signal of SECAM signals.
4.5Luminance Processing
The A/D-converted CVBS (or Y) signal is digitally clamped to back porch. Depending on
the transmitted standard and operational area, an offset between black- and blanking
level can be found in the incoming signal (’7.5 IRE’). As for some applications a black
offset is not desired, controlling may be done using LMOFST. The positive or negative
offset is added to the Y signal before scaling.
Micronas4-15
SDA 9489X
SDA 9589X
Preliminary Data Sheet
Received signalProcessed signal
BLANK value
LMOFST
BLANK value
LMOFST
BLACK value
='00' (no additional offset)
BLACK value
='00' (no additional offset)
BLANK value
LMOFST
BLANK value
LMOFST
BLACK value
='10' (reduction of 16 LSB)
BLACK value
='01' (addition of 16 LSB)
Figure 4-6Black level correction of luminance signal
System Description
M standard signals
B/G/H/I/N standard signals
The color carrier is removed out of a CVBS signal by means of a notch filter. It is set to
the corresponding color carrier (3.58 or 4.4 MHz) only if the standard is detected
permanently. This prevents the luminance sharpness of being changed within the
standard search process. For Y signals the notch is disabled.
For a fine adjustment of delaycompensation between luminance and chrominance,
YCDEL allows a luminance shifting in 16 steps of 50ns.
4.6Decimation
4.6.1Single PIP Mode
Luminance and chrominance signals are filtered in horizontal and vertical direction. The
coarse horizontal and vertical picture size (1/2, 1/3, 1/4, 1/6) is independently
programmable with SIZEHOR and SIZEVER. A fine adjustment in steps of 4 pixel and 2
lines is possible by HSHRINK and VSHRINK, which allows correct aspect ratio for
multistandard applications (50/60 Hz mixed mode, (S)VGA).
For main decimation factors, the stored number of pixel and lines are listed in the
following tables.
Micronas 4-16
SDA 9489X
SDA 9589X
Preliminary Data Sheet
SIZEHORhorizontal
scaling
D1D0
PIP Pixel per line
Y(B-Y)(R-Y)
System Description
002:13248181
013:12165454
104:11604040
116:11082727
Table 4-5Number of stored pixel per line dependent on SIZEHOR
SIZEVERvertical scalingPIP lines
D1D0
625 lines source525 lines source
002:1132108
013:18872
104:16654
116:14436
Table 4-6Number of stored lines per field
4.6.2Continuos Zoom
The continuos zoom feature changes the picture size rapidly in an animated manner. It
is available in single-PIP mode for picture sizes smaller or equal 1/4 of the undecimated
picture.
There are three possibilities of using the zoom feature:
• The PIP is zoomed via HSHRINK and VSHRINK manually. This requires an I
2
C
protocol each time the picture size should change. CZMEN should be used to
synchronize the update of HSHRNK/VSHRNK with SIZEHOR/SIZEVER.
• A different way is to make usage of the automatic zooming. The zoom speed can be
controlled by CZMSPD. When switching PIP on or off by using PIPON, the PIP zooms
automatically to the selected picture size or disappears at size of 1/81.
• A zooming between two picture sizes can be performed by changing the HSHRINK,VSHRINK, SIZEHOR, SIZEVER values when CZMEN is enabled. Then the new
picture size is obtained by zooming and not taken immediately.
Automatic zooming is only possible in frame mode. Being in field mode, the picture size
remains stable until frame mode occurs or until the internal counter reaches the desired
Micronas4-17
SDA 9489X
SDA 9589X
Preliminary Data Sheet
System Description
picture size. Then the size changes immediately. Equal to the wipe process, the zooming
direction depends on the coarse position (CPOS).
Table 4-8Number of stored pixel per line dependent on HSHRNK
4.6.3Horizontal And Vertical Fine Positioning
All picture sizes are pre-centered inside the frame. In addition, if necessary the vertical
and horizontal acquisition area can be shifted by VFP for vertical and HFP for horizontal
direction.
Micronas4-19
SDA 9489X
SDA 9589X
Preliminary Data Sheet
System Description
4.6.4Multi Display Mode
SDA 9589X and SDA 9489X offer the feature to display a sub-picture more than once.
The picture size and arrangement depends on the display mode (DISPMOD) and not on
SIZEHOR or SIZEVER. Hence variable scaling is not possible in these modes.
Display
Mode
100SIZEHOR/
DISPMODSizePicture
D1D0625525
configuration
single PIP mode324
SIZEVER
HSRHNK/
PixelLines
132
-
60
-
24
108
-
20
VSHRNK
2013 X1/9one upon another
216264216
(same content)
3104 X 1/16one upon another
156264216
(same content)
Table 4-9Multi-display modes
The display modes are shown in the appendix. The sizes of the partial pictures are listed
in table 4-11 .
4.6.5Split Screen
For split screen applications two selectable ’double window’ modes in which one half of
the picture is generated by the ’Sophisticus’/’PIP IV Advanced’ can be used. The split
screen mode can be selected by two possible combinations of DISPMOD.
The D1.5 mode is suited for displaying split screen on 16:9 tubes keeping the aspect
ratio. The DW1 format covers the full height of the screen.
4.6.6Multi-PiP Mode
There is a great variety of multi-pip modes available. Up to 11 different still pictures and
one moving picture can be shown. This is useful to give an overview over broadcasted
programmes (e.g. tuner-scan) or for supervising purposes. For multi-PiP modes only
three fixed picture sizes are available (1/9, 1/16 or 1/36). The picture size and
arrangement depends on the display mode (DISPMOD) and not on SIZEHOR or
SIZEVER. Variable scaling is thus not possible in these modes. Because of limited
memory capacity, the number of frozen multi-pictures is limited dependent on picture
size to the number shown in the table below:
picture sizesmaximum number of pictures
(including one live picture)
1/93
1/166
1/3612
Table 4-10Maximum number of pictures in multi-PIP mode
The partial picture that is written is addressed via WRPOS. With INFRM, a frame for
separation of every PiP can be selected. This is adjustable to single or dual PIP mode
(INFRMOD). The current updated picture can be highlighted with PIPHLT. To avoid
garbage pictures after switching from one mode to another the selected picture can be
blanked with PIPBLK. MPIPBG defines wether the picture will be blanked with black or
with the adjusted background color.
For compatibility reasons to other devices, the DISPMOD register is split into two
segments. If a display mode is chosen that is not implemented, the PIP insertion is
switched off automatically (PIPON = ’0’). The sizes of the partial pictures correspond to
the sizes of the inset pictures of the single PIP modes.
Micronas4-21
SDA 9489X
SDA 9589X
Preliminary Data Sheet
Display
Mode
DISPMODSizePicture
D6 D5 D4 D3 D2625 525
configuration
System Description
PixelLines
4000012 X 1/9,one upon another216176 144
5000102 X 1/9,side by side 4328872
6000113 X 1/9,side by side6488872
7001003 X 1/9one upon another 216264 216
8001014 X 1/16side by side6246654
9001106 X 1/16inverted U shaped624132 108
10001116 X 1/16U shaped624132 108
11010004 X 1/162 rows of 2 pictures312132108
12010014 X 1/16one upon another156264216
130101012 X 1/36 6 rows of 2 pictures216264 216
140101112 X 1/36 2 rows of 6 pictures6488872
15011009 X 1/363 rows of 3 pictures324132108
160110112 X 1/36 3 rows of 4 pictures432132 108
170111011 X 1/36 angular of 11
648264 216
pictures
18011119 X 1/36angular of 9 pictures540220 180
19100001X1/3Double Window
324176 144
(V=1.5)
20100011X1/2Double Window
324264 216
(V=1)
21 10010OSD only
all otherPIP off (PIPON=0)
Table 4-11Display Modes
4.7Display Control
The on-chip memory capacity is 768 kbits. Provided that the same standard (50 or 60
Hz)
video sources are applied to inset and parent channel, joint-line free frame mode
display is possible. This means that every incoming field is processed and displayed by
the SDA 9589X/SDA 9489X processors. The result is a high vertical and time resolution.
For this purpose the standard is analyzed internally and frame mode display is blocked
Micronas 4-22
SDA 9489X
SDA 9589X
Preliminary Data Sheet
System Description
automatically, if the described restrictions are not fulfilled. Then only every second
incoming field is shown (field mode). Field mode normally shows joint-lines. This is
caused by an update of the memory during read out. The result is that one part of the
picture contains new picture information and the other part contains one earlier written
field. The switching from or to frame mode is free of artifacts.
Activation of frame-mode display is blocked automatically if at least one of the following
conditions is not fulfilled:
• Inset and parent channel have the same field repetition frequency. This means that
frame mode is possible only for 50Hz inset and parent sources or 60Hz inset and
parent sources.
• Interlace signal is detected for inset and parent channel. For progressive scan or
(S)VGA display therefore only field mode is possible. For some VCRs in trick mode,
often no interlace is detected also.
• The number of lines is within a predefined range for inset (FMACTI) or parent
(FMACTP) channel (assuming standard signals according to ITU)
FMACTPparent
standard
number of
lines per field
FMACTIinset
standard
number of
lines per field
050 Hz310...315050 Hz310...315
150 Hz290...325150 Hz290...325
060 Hz260...265060 Hz260...265
160 Hz250...275160 Hz250...275
Table 4-12Required number of lines for frame mode display
The system may be forced to field mode by means of FIESEL. Either first or second field
is selectable. ’One of both’ takes every second field independent of the field number.
This is meant for sources generating only one field (e.g. video-games).
For progressive scan conversion systems and HDTV / (S)VGA displays a line doubling
mode is available (PROGEN). Every line of the inset picture is read twice.
Memory writing is stopped by FREEZE bit. The field stored in the memory is then
continuously read. As the picture decimation takes place before storing, the picture size
of a frozen picture can not be changed.
Synchronization of memory reading with the parent channel is achieved by processing
the parent horizontal and vertical synchronization signals connected to the pin HSP for
horizontal synchronization and pin VSP for vertical synchronization. HSPINV or VSPINV
respectively allow an inversion of the expected signal polarity.
Micronas4-23
SDA 9489X
SDA 9589X
Preliminary Data Sheet
System Description
HSP
VSP
VSPDELVSPDEL
VSPD
(internal)
values in brackets () apply for 100Hz systems
field 0 windowfield 1 window
tH/2 = 32 (16)
←
s
tH = 64 (32) ←s
=151 (75) ←s
max
Figure 4-8Field detection and phase adjustment of vertical pulse (VSP)
Depending on the phase between inset and parent signals a correction of the display
raster for the read out data is performed. As the external VSP and HSP signals may
come from different devices with different delay paths, the phase between V-sync and
H-sync is adjustable (VSPDEL). An incorrect setting of VSPDEL may result in wrong or
unreliable field detection of parent channel.
Normally a noise reduction of the incoming parent vertical pulse is performed. With this
function missing vertical pulses are compensated. The circuit works for 50/60 Hz
applications as well as progressive and 100/120Hz application. (S)VGA signals are
supposed to be very stable and therefore not supported by the noise suppression. By
means of VSPNSRQ, vertical noise suppression is switched off.
A great variety of combinations of inset and parent frequencies are possible. The
following table shows some constellations.
valid for some parent frequencies. Please refer to Chapter 4.7.2
2)
3)
3)
Table 4-13Available Features with varying inset and parent standards
4.7.1100 Hz Frame Mode
If the picture size is smaller or equal than 1/9 PIP a true frame mode display for 100Hz
parent standard with a double field repetition rate is possible (display raster ∼∼ϒϒ only).
The picture size is indicated by the horizontal and vertical decimation factors that must
be equal or below 1/3 of undecimated picture size in both directions. This guarantees
enough memory for a joint-line free picture with full vertical resolution. For bigger pictures
only field mode is supported. The 100 Hz frame mode is activated if READD=’1’ for the
above mentioned picture sizes. For an acceptable quality without line flicker or motion
artifacts only the mode ∼∼ϒϒ is supported for HSP and VSP. If the sequence ∼ϒ∼ϒ is
detected, the field mode will be activated again. Continous switching between these
modes is possible, resulting in continous switching between field- and frame mode.
Micronas4-25
SDA 9489X
SDA 9589X
Preliminary Data Sheet
4.7.2Mixed Standard Applications And (S)VGA Support
remark
(N
apel
X N
aline
@ fV)
720X576@50Hz
(TV)
702X488@60Hz
(TV)
720X576@100Hz
(TV 100 Hz)
702X488@120Hz
(TV 120 Hz)
720X576@50Hz
(TV progressive)
fH
(kHz)
(
TH
←s)
T
Hact
(←s)
15.664.052.0625/
15.763.652.7525/
31.232.026.0625/
31.231.826.4525/
31.232.026.0625/
lines/
active
576
488
576
488
576
f
dot
(MHz)
13.5interlace
13.5interlace
27interlace
27interlace
27prog-
System Description
scancorrect
aspect
ratio
ressive
702X488@60Hz
(TV progressive)
640X480@60Hz
(VGA)
640X480@72Hz
(VGA)
640X480@75Hz
(VGA)
800X600@56Hz
(SVGA)
800X600@60Hz
(SVGA)
800X600@72Hz
(SVGA)
800X600@75Hz
(SVGA)
31.231.826.4525/
488
31.531.825.4525/
480
37.926.420.3520/
480
37.526.720.3500/
480
35.228.422.2625/
600
37.926.420.0625/
600
48.120.816.0666/
600
46.921.316.2625/
600
27prog-
ressive
25.2progressive
31.5progressive
31.5progressive
36.0progressive
40.0progressive
50.0progressive
49.5progressive
800X600@85Hz
(SVGA)
1024X768@43Hz
(SVGA)
53.718.614.2631/
600
35.528.222.8817/
768
56.3progressive
44.9interlace
Table 4-14Examples of supported parent signals
Micronas 4-26
SDA 9489X
SDA 9589X
Preliminary Data Sheet
System Description
SDA 9589X and SDA 9489X allow multiple scan rates for the use in desktop video
applications, VGA compatible or 100Hz TV sets. All features are provided in ’normal’
operating modes at auto detected 50Hz and 60 Hz parent and inset standards. 2f
modes (100/120Hz and progressive) are supported by line frequency- and pixel clock
doubling and are not detected automatically. Even on a 16:9 picture tube correct aspect
ratio can be displayed by selecting the suitable parent clock. The video synthesizer
generates also a special pixel clock for VGA display (see chapter 5.5.9 for details). As
(S)VGA consists of a variety of scan rates the correct aspect ratio is not adjustable for
all modes with the parent clock (HZOOM) because of the limited count of frequencies.
For single PIP only, correct aspect ratio is maintained by the vertical and horizontal
scaler (HSHRINK and VSHRINK).
It is possible to display (S)VGA sources for parent display, as long as the horizontal
frequency is lower than 40 kHz and the signal does not contain more than 1023 lines.
For progressive scan mode, PROGEN must be set. Additionally field-mode should be
forced to prevent unallowed frame-mode displaying (FIESEL). As the (S)VGA normally
does not fit to the display raster generated in the vertical noise suppression, VSPNSRQ
should be disabled. (S)VGA signals for inset channel are not supported.
H
PROGENREADDExpected input signal
0050 or 60 Hz signal interlace
01100 or 120 Hz signals interlace
10(reserved)
1150 or 60 Hz or (S)VGA signal progressive
Table 4-15Selection of display field repetition
4.7.3Display standard
For a single-PiP, the number of displayed lines depends on the selected picture size and
on the signal standard. For multi picture display, the number of displayed lines depends
on the selected picture size and on the signal standard of the parent signal. Additionally,
a standard can be forced by DISPSTD.
Micronas4-27
SDA 9489X
SDA 9589X
Preliminary Data Sheet
System Description
DISPSTDDISPMODDisplay Standard
D1D0
000PIP depends on detected inset standard (single pip)
00>0PIP depends on detected parent standard (multi display)
01xPIP display is always in 625 lines mode
10xPIP display is always in 525 lines mode
11xfreeze last detected display standard and size
Table 4-16Display standard selection
If a 625 lines picture is shown with a 525 lines parent signal, some lines are missing on
top and bottom of picture. If a 525 lines picture is shown with a 625 lines display
standard, missing lines at top and bottom are filled with background color or black
depending on MPIPBG.
625 lines / 50 Hz
525 lines / 60 Hz
Figure 4-950 and 60 Hz Multi PiP display on 50 Hz and 60 Hz display
4.7.4Picture Positioning
The display position of the inset picture is programmable to the 4 corners of the parent
picture (CPOS). From there PIP can be moved to the middle of the TV Picture withPOSHOR and POSVER. The corner positions can be centered coarsely on the screen
with POSOFH and POSOFV. Depending on coarse position, one PIP corner remains
stable when changing the picture size.
Micronas 4-28
SDA 9489X
SDA 9589X
Preliminary Data Sheet
CPOSCoarse
D1D0
Position
Reference
corner of PiP
increasing
POSVER
System Description
increasing
POSHOR
00upper leftupper leftdown right
01upper rightupper rightdownleft
10lower leftlower leftupright
11lower rightlower rightupleft
Table 4-17Coarse Positioning
There are 256 horizontal locations (4 pixel increments) and 256 vertical locations (2 line
increments). The pixel width on the screen depends on the selected HZOOM factor.
Even POP-positions (Picture Outside Picture) in 16:9 applications are possible.
CPOS='00'
POSVER
CPOS='10'
Figure 4-10Coarse Positioning
4.7.5Wipe In / Wipe Out
POSHOR
CPOS='01'
POSVER
CPOS='11'
POSHOR
With the wipe in / wipe out function it is possible to let appear or disappear the complete
inset picture starting or ending at the corner of the inset picture position defined by
CPOS. Thereby the size of the visible picture-part is continuously increased and
decreased respectively. During this procedure the frame is shown with its chosen widths.
3 different wipe in / out time periods or ’no wipe’ are programmable via WIPESPD. The
wipe algorithm always works in horizontal and vertical direction.
Micronas4-29
SDA 9489X
SDA 9589X
Preliminary Data Sheet
CPOS='00'
CPOS='10'
CPOS='01'
CPOS='11'
wipe out
CPOS='00'
CPOS='10'
wipe in
CPOS='01'
CPOS='11'
System Description
CPOS='00'
CPOS='10'
CPOS='01'
CPOS='11'
Figure 4-11Wipe display
If WIPESPD is set accordingly, PIPON controls the wipe operation. When PIPON
changes the wipe operation starts. During this period, the readable PIPSTAT indicates
the ongoing wipe-process. A transition of PIPON from ’0’ to ’1’ triggers the wipe-in. The
wipe-in process stops when the picture reaches its programmed size. When PIPON
changes from ’1’ to ’0’ the wipe-out starts. The wipe-out is finished when the PiP picture
vanishes. Even for multi-picture display wipe operation is possible. A change of PIPON
or WIPESPD during wipe operation has only an effect after the wipe operation has been
finished.
4.8Output Signal Processing
4.8.1Luminance Peaking
To improve picture sharpness, a peaking filter which amplifies higher frequencies of the
input signal is implemented. The amount of peaking can be varied in seven steps by
YPEAK. The setting ’000’ switches off the peaking. The value ’001’ is recommended as
This value provides a good compromise between sharpness impression and annoying
aliasing. The characteristic for all possible settings is shown in fig. (4-12). The
emphasized frequency depends on the adjusted decimation. The gain maximum is
always located before the band-limit ensuring optimal picture impression.
Micronas 4-30
Loading...
+ 72 hidden pages
You need points to download manuals.
1 point = 1 manual.
You can buy points or you can get point for every manual you upload.