S-Video output on old VHS recordings?

[Brad]

Even though VHS as a format has Y/C separation, is it really a good idea to be outputting S-Video from old VHS tapes? Don't they have the same old-tech 2D separation issues as old Laserdisc players? Particularly recordings from analog CATV or antenna.

What is the advantage versus plugging the composite output into a digital 3D Y/C separating device?

[themaster1]

Depends on the circuitry of the vcr and/or its age, not to mention the cable itself but given my 10 years experience now i can say s-video is best for commercial tapes

[jagabo]

The signal on a laserdisc is stored as composite video. On VHS tapes the video is stored as separate luma and chroma.

[Brad]

The signal on a laserdisc is stored as composite video. On VHS tapes the video is stored as separate luma and chroma.

But you're missing the point... The signal coming over the cable wire into the VCR was also composite which was separated by a machine made in 1990, for instance.

[jagabo]

The signal on a laserdisc is stored as composite video. On VHS tapes the video is stored as separate luma and chroma.

But you're missing the point... The signal coming over the cable wire into the VCR was also composite which was separated by a machine made in 1990, for instance.

The low resolution of VHS will greatly diminish dot crawl artifacts in a composite source. And the recording deck probably has at least a simple dot crawl filter. In any case, adding a second round of dot crawl artifacts isn't going to help.

[misermidas]

Depends on the circuitry of the vcr and/or its age, not to mention the cable itself but given my 10 years experience now i can say s-video is best for commercial tapes

Commercial were (of should have been) made from Y/C (s-VIDEO) sources , it is also so for camcorder files,so S-VIDEO connexion will be better in that case.

With tapes recorded from analog broadcastings, Y/C multiplexing and demultiplexing at recording were not always done properly ,so the tapes may look finer and show less compositing artifacts with composite connexion than with S-VIDEO, unless you use powerful filters when PC restorating your videos.

With S-VIDEO you will get more information from your tapes but some of these may be video noise or artifacts that you will have to remove.

[Brad]

In any case, adding a second round of dot crawl artifacts isn't going to help.

Why would there be additional dot crawl if you use 3/D Y/C? And yes, I don't see dot crawl on my VHS too much but I do see plenty of rainbowing.

There is also the question of whether the playback VCR even maintains separation from the heads straight through to the S-Video output, or if it combines then re-separates itself using old tech.

[jagabo]

In any case, adding a second round of dot crawl artifacts isn't going to help.

Why would there be additional dot crawl if you use 3/D Y/C?

On a plain VHS deck the separate luma and chroma on the tape will be converted back to composite video. Separating that again at the capture device will create dot crawl artifacts. Then the capture device applies the 3D comb filter to remove the last round of dot crawl artifacts. Even a 3d comb filter is not perfect. For any particular tape you should try both methods (S-VHS deck putting out s-video or composite) to find witch works best. But if you don't have an S-VHS deck you can't even try the former. And you won't have a line-TBC to clean up the horizontal jitter.

And yes, I don't see dot crawl on my VHS too much but I do see plenty of rainbowing.

Rainbowing is just another manifestation of incomplete luma/chroma separation.

There is also the question of whether the playback VCR even maintains separation from the heads straight through to the S-Video output, or if it combines then re-separates itself using old tech.

Any good S-VHS deck will keep luma and chroma separate from the tape to the s-video port.

[2Bdecided]

The signal on a laserdisc is stored as composite video. On VHS tapes the video is stored as separate luma and chroma.

But you're missing the point... The signal coming over the cable wire into the VCR was also composite which was separated by a machine made in 1990, for instance.

True, but there is no benefit to re-combining it and separating it again. Now it's separate, keep it separate. Any extra information that may have allowed a better separation is long gone.

[Brad]

I was wondering about that. So if I record a test pattern with the worst hardware around, then play that poorly-separated "component digital" signal out via composite into top-of-the-line hardware, the best it can do is still going to be replicating the rainbowy component signal?

[themaster1]

If you really want to capture composite you may find the comb filter of Dscaler handy, more info below
http://deinterlace.sourceforge.net/Help/TemporalComb.htm

it works for both recordings and preview