Sony Digital 8 S-video connector no good?

[Keith Wiebe]

I just got my new Sony DCR-TR7000 digital 8 camcorder and am impressed by the picture quality but have one problem. When recording from the S-video connector into the camcorder under certain coditions such as a real colorful flower I will loose color information to the right of the flower (extremely big gray image) but when using the composite video in I have no problems! Anyone else experienced this? At least the Sony sync. up to the Rainbow Runner-my other Canon hi-8 machines won't-I get a scrolling picture with them. Picture quality is certainly better than Hi-8. The Rainbow Runner only cuts the quality somewhat in the second highest settings-had to back off from the highest quality because of dropped frames on the colorful red and yellow flowers.

[Keith Wiebe]

I found out it must be the Rainbow Runner causing the concern (or I guess it could be the Mystique 220 video card) because I hooked up my TV to the s-video output and it also had the gray lack of color concern. Composite output is just fine. Has anyone had a problem with the connector?

[Mark F]

I haven't noticed this problem on my DCR-TRV510 (D8 w/LCD)

Mark F.

------------------
OH NO, my retractable cup holder swallowed a CD

[Patrick]

Keith, your second post is a little confusing. You state:

"... I hooked up my TV to the S-video output..."

Are you referring to the S-video output of the RR-S or of the camcorder?

If you connect your camcorder directly to the TV and you do NOT see a problem, it sounds like the trouble is elsewhere.

[Keith Wiebe]

Let me clarify things. When I hook up my TV to the s'video output of the Rainbow Runner I still have a problem with color and gray images like when outputting to the Sony camcorder but not quite as severe but still very noticeable. It must be something in the cable coming out of the Rainbow runner or the runner-video card itself.

[Grigory]

I have a problem similar to yours.

Digital8 camcorder TRV110E (PAL, capable to NTSC playback on PAL or NTSC).
Sony 25' TV, composite input.
Canopus DV Raptor card with analog overlay preview channel, S-video or composite inputs.
Panasonic 14' TV set, composite input.
Avermedia TV phone analog capture card, S-video and composite inputs.

I made red square on white and tried to see what is on screens.

1. Sony 25' composite. Correct chroma/luma delays, at least withing 150 ns standard "gate"
2. Panasonic 14' composite. Chroma components are shifted from luma component
3. Canopus Raptor overlay - both for S-video and composite I have incorrect luma/chroma sync. Same as for Panasonic TV.
4. Avermedia TV phone - correct sync for S-video. Did not try composite connection.
5. DV avi file - correct sync.

The theory says that luma and chroma components are transmitted with different bandwidths. This cause a delay on TV side, which must be compensated by luma component delay line. The precision must be within 120 ns.
In my experiments, Sony TV and cheap BT848 card had correct alignment of luma to chroma components. Panasonic TV and DV raptor card both had misaligned luma and chroma. What is a conclusion - I don't know what is incorrect - D8 output or one pair of devices that I tested.
Cables quality is not important, because I used pass-through connection of Raptor S-video to BT848 card, and it still showed correct timing. I also tried different RCA cables, from good video to cheap audio cables. There is no difference (the cables were short enough to introduce any delay)

The delay mismatch is evident on good quality high chroma contrast images like flowers.

S-video and composite outputs/inputs may have different signal propagation delays because of different filtering. So, some TV's may have aligned pictures only for one type of connection. Note, this effect is not too much visible at normal distance from TV screen, or for smooth images. A good reason to save few cents on proper delay lines?

Being interested in this topic, I made NTSC movie of red square and tried to look at it. Only TV's appeared to be able to show NTSC 4.43 in color. Both had exactly the same behaviour. This is little bit strange, because NTSC has different chroma delay(s) from PAL due to a narrow bandwidth of one chroma component. However, the results were similar. Maybe, delay line changed the delay value in time (6 years).
I have now no RR_G to compare it with other devices.

Grigory