I am a bit wiser now after trying out several things. As I am keen to use the best possible quality in my capture application that is still to be written, I would very much like to understand what is responsible for the quality improvement, so that I can implement this. At least these simple tools don't complain about the missing chrominance channel input. tried here AmCap, GraphEdit with a basic filter chain I have selected, sourceforge capture tools). Non of the other tools I used could reproduce this signiificant quality improvement. Even though I select S-video, they show similar quality to composite (e.g.
Someone at Hamamatsu has recommended this to me as the luminance channel has a significantly higher bandwidth over the composite input which is indeed resulting in crisper videos with more undistorted detail! The puzzling thing however is that this increase in quality can currently ONLY be realised if I use the Pinnacle Studio software to capture the avi (Pinnacle studio is first throwing an error message saying that it cannot detect a vaild input signal but does that job anyway and well if you click the message away, here you go). However I made another interesting observation: I am connecting my analog Hamamatsu video processor to the Pinnacle Dazzle via the S-video input even though the Hamamatsu device is delivering a black and white signal only - consequently I am using the luminance channel only. It revealed under the "add filter" option quite a few Pinnacle proprietary DirectShow filters, which are interesting to explore. Chris, all I am currently playing around and prototyping a bit with the MS WDK tool GraphEdit.