S-VIDEO significantly grainer than A/V when digitalizing Hi8.

[BiggK]

I'm using a digital8 handicam and have recorded the same tape(virtualdub with huffyuv) using both S-VIDEO and A/V.

The copy recorded using A/V is significantly smoother. Faces and solid objects look clean and smooth, while the copy recorded using s-video has a grainy noise permeating through the video.

Does anyone know what could be causing this? Surely S-VIDEO should be providing a cleaner picture.

Could it be to do with A/V including less detail and thus giving a softer image that looks smoother to my eye? It doesn't seem that way to me and regardless, the A/V recording is much more aesthetically pleasing.

[Cornucopia]

If you are using digital 8, why aren't you using firewire for a straight digital out without conversion? That way would give you the highest quality (assuming you aren't doing passthrough, or analog playback).


Scott

[Brad]

^ The topic title indicates these are Hi8 tapes.

Could it be to do with A/V including less detail and thus giving a softer image that looks smoother to my eye?

Probably. You didn't say which capture device you're using, but many devices sort of smear composite input. Without attaching samples for us to look at, we can only guess.

[Richard_G]

Digital8 camcorder will play back Hi8 tapes (and D8) and output as DV. Most definitely the best way to do it, a straight digital transfer.

[dellsam34]

Digital8 camcorder will play back Hi8 tapes (and D8) and output as DV. Most definitely the best way to do it, a straight digital transfer.

The reason why some people use analog outputs for V8 and Hi8 analog tapes even when using a D8 camcorder is to avoid the lossy DV codec, So that way they can capture lossless AVI instead of lossy AVI.

[nicholasserra]

^ And to use a frame TBC in the signal chain.

For analog 8mm cassettes I always do at least two passes, one lossless via s-video, and one DV via firewire. They have different qualities, so I store both.

[jagabo]

Composite video includes the chroma on a high frequency carrier on the composite signal. The chroma and luma need to be separated at the receiving end. The chroma carrier can be removed from the luma with a low pass filter. Hence the nose reduction.

[mitchp10]

I'm having a similar issue I think, I'm trying to convert Hi8 tapes to digital but I'm playing it back on a Hi8 camcorder, a Sony CCD-TRV68. I had just cleaned the heads before playing back this tape, so it's not dirty heads. S-Video looks much crisper, much more detailed (as it should) but it has much more grain than composite. Attached is a screenshot from each. Is there any way to fix this, like maybe it's a bad cable? I don't have another S-Video to test it on, so that could be it.

S-Video:

[Attachment 62531 - Click to enlarge]

Composite:

[Attachment 62532 - Click to enlarge]

[dellsam34]

Composite is just washing out the details therefore hiding grains, It is not better than S-Video by any stretch. The grains are due to low light conditions from a low resolution imaging sensor, it has nothing to do with the format itself, Take a clean digital video and record it to a Hi8 camcorder equiped with VCR mode and see how it looks.

[jagabo]

The s-video cap is closer to what's on the tape. If your capture device has a sharpness setting turn it down. All it does is increase the noise with low res consumer recordings. Same for the camera. Your images show a lot of DCT ringing and blocking artifacts. Don't use lossy compression codecs while capturing. Use lossless instead. You can reduce noise and compress later.