i have an interesting question.
i m a film editor and has worked on feature films with my old nice betacam VTR and mojo analog to do offline editing. really had to do with editing no finishing, just a nice EDL or cutlist.
some music video work with dps velocity and quality was nice while capturing, editing, Color correcting and outputting to tape.
recently i wanted to upgrade my setup and wanted to do HD film. while my avid was little bit old i opted for premiere pro 2 with XXXXXXXXX card. but problem ....quality was bad in all those 4 steps. video was captured through component connections.
so concluded that it was capture card's fault because video was captured in 8 bit YUV uncompressed so no compression artifacts
now i have an interesting test to do but don't have full equipment so hope you guys will help me.
first need to create some file (i would have created some test patterns some graphics and some video) and render it to YUV 8 bit uncompressed.
save it as source file.
play this file in premiere pro and capture on the other end with premiere pro and label as target file.
1. two machines should use SDI connection to connect to each other (as people say SDI is just bit by bit transfer nothing less nothing more)
2. test with component connections can be also done but first test is necessary.
now take source and target file and put some frame of them in photoshop or afterfx and apply difference blending mode to see if they are really bit by bit same. or how much loss in second test.
hope some guy with two decklink machines will perform it and clear my doubt and reassure faith in capture card so i could convince my clients that its not capture card problem or premiere pro problem.
(also this fact adds to my doubt that some video captured through DVS card in lustre looks better than premiere pro, if SDI is no loss connection, is there something in DVS centaurus that improves the quality of video.)