That's because ATI 7500 will be fed by S-Video directly from the LD player, whereas Conexant will be fed by Component video from Panasonic DVD Recorder middleman.

And by the way chroma resolution is better on Conexant than on ATI AW. It doesn't really matter because the source is LD but still. If you were to capture from DVDs, you'd notice the difference.

I don't understand what you mean by "capture from DVD". Who would capture a DVD? What format are you capturing to?

Well, the DVD Recorder is a middleman between the LD player and Conexant capture device. So technically I capture from the DVD Recorder's output which is fed the signal from the LD player. Everything is lossless. No DVDs involved.

Update for AIW 7500. I decided to have Sanlyn's settings for contrast, i.e. 106, but increased brightness to 166. It gives a lot more rooms for dark images while not clipping whites at the very bright scene.

Sorry for the off-topic, but you know Waifu2x?

No.

The settings posted in this thread are usually suggestions based on several cards and input sources. In this case the settings you refer to were made when the card that was connected to my PC was an AIW 9600XT AGP. Contrast and brightness settings are customized for specific capture devices and specific input sources. They will differ with changing situations.

I have captured Video Essentials LaserDisc using both the Conexant based USB card and AIW 7500.

Attached are the screenshots from the AIW AVI. I was told by the LaserDisc expert that I need to reduce brightness to the point where the blackest bar is invisible on black background. Should I really be doing this? The put white arrows to the edge of the blackest bar. What should I do with grayer bar on the Color Bar pictures (I indicated it with yellow arrows). If I need to adjust the video level, should I be adjusting the video output of the LD player or should I be adjusting Brightness of capture cards?

Of course.

Nothing.

Adjust the output of the player first. The only problem with this is that if your monitor isn't properly calibrated it can't assure viewing accuracy. You should be adjusting first for the view you have of those test patches and follow your expert's advice or test disk instructions as best you can. Don't forget that once you have an input program or movie to capture you will likely have to check levels again with a capture-time histogram.

I know it's probably a little late to chime in here, but I was reading through this thread and noticed some misinformation...

metaleonid, sanlyn is correct here. The majority of LD players have only composite output, with the exception being some of the higher-end players. Don't be misled by the presumption that an s-video connection will inherently have better output though, in this case.

The analog video signal stored on LDs is in the composite domain.

As you mentioned, these higher-end players may have a simple 2D/comb filter applied to their s-video output. You will almost certainly be able to achieve better results by sending the direct composite output through a high quality 3D filter.

Hi Kooz,

Yes, the data stored on LD as composite. A lot of LD players though have both S-Video and Composite outputs. I have 2 LD players with Composite and S-Video outputs. They are Pioneer CLD-D703 and Pioneer Elite LD-S2. The latter one does Y/C separation internally and then recombines it for Composite output.

I capture both S-Video and Composite though. I feed composite to either capture or middleman devices with 3D comb filters. While 3D comb filters remove dot crawls and rainbows in static test patterns, I actually found that simple S-Video captures from LD-S2 often yield superior results in real (dynamic) scenes (non test-pattern captures). 3D comb filter may even introduce dot crawl at scenes where S-Video captures perform fine. I also found that S-Video captures are generally sharper. However, there's a talk that there exists the best 3D comb filter found on ADV7842 board. I haven't tried that one yet.

--Leonid

Interesting, thanks for the follow-up. I was always under the impression that the s-video outputs on the later/higher-end players were just introducing an unnecessary and, in some cases, harmful conversion. I'm not really sure how you could end up with a sharper capture unless it's undergoing some additional processing. But to be fair, I do not have LD players of my own to make such comparisons. I've just done a lot of reading about LDs as I find the format fascinating.

I plan to post 12 1 second or so samples of the same footage using various hardware, players and connections. So you'll be able to determine how it is handled in real life.

The additional processing is introduced by 3D comb filters not by 2D comb filters. Kurtis Bahr indeed warned me a while back that motion adaptive 3D comb filters soften the image.

Cool. I look forward to checking those out.

I maybe ready to provide real time captures as soon as today or tomorrow. I also wanted to provide the moving zone test pattern using both S-Video and Composite with motion adaptive 3D comb filter for you to see what is really going on.

Should I start a separate thread for these samples?

If the captures illustrate what's been discussed in this thread, post them here.

I'm pretty new here, but for what it's worth, I agree.

Not really. This thread is about ATI. But I was going to post samples of various combinations of players and capture cards.