What test patterns are most useful in testing analog conversion gear?
I’m getting into the testing phase of a variety of capture cards, TBCs, VHS players, upscalers, and deinterlacing methods. My question is, how would you best “objectively” test them?
Certain test patterns should expose some of the pros and cons of various hardware and possibly even identify dropped frames if those digital patterns are recorded onto VHS and then played back for capture. It would also give a reference as to what the original looks like compared to different capture methods.
I also don’t have particularly bad tapes that already have significant timebase errors, so are there any good methods for creating timebase errors during the recording?
A few patterns that I think will be useful are attached - but they may not be ideal, hence reaching out to those that know more on the topic or have ideas on what to try.
The AudioVideoSyncFrameDropDVD test can be played on a DVD player and recorded to VHS until the tape is full. Audio sync and frame drops and duplications should be obvious as the tape progresses due to the incremental change between frames. It is a looped version of this YouTube video here:
https://www.youtube.com/watch?v=ucZl6vQ_8Uo in 6 minute increments and the DVD is set up to loop that chapter continuously. The idea is that it should be able to get around needing to see if you are having frame drops or duplications.
To use the sync test yourself, you’d just burn it to a DVD using any program that can burn ISOs and then record the DVD player’s output to a VHS or other analog format, then capture it to see if audio goes out of sync or if you see frames repeat or that are missing.
There other patterns I haven’t made into DVDs yet (they could all be put onto a single DVD), but wondering if there are some critical patterns that show the most information in terms of automatic gain control, sharpness, blending of frames, video artifacts, deinterlacing issues etc.
The one Interlaced pattern has the “I” in odd fields and the “II” in even fields, so you can tell if your capture method is just discarding one of the fields as one of them won’t show up in the capture. It’s also kind of interesting to play it in VLC and see how the different deinterlacers in there handle it.
I believe the test I have labeled as AGC is meant to stress the automatic gain control and determine if the capture card fails at capturing dim details when there are very bright areas on screen, but feel free to correct me if that is not the purpose of this test pattern.
I also plan to use the Snell and Wilcox Zone Plate test, I'll attach that in the next post since it is too large for a single post with the other files.
Any thoughts or hardware testing is appreciated and hopefully the Audio/Video Sync test helps someone test their own gear!
-- merged --
Attached is the Snell and Wilcox test pattern that shows a variety of things including how diagonal lines, high contrast, and resolution are handled. It contains two versions of the pattern that switches half way through. This came off of Internetarchive.
