Capture whole tape with one set levels, or different levels per scene?
 

I am starting my capture process and I am wondering if it works ok to set Brightness and Contrast levels during preview that satisfy the whole tape or if I really should readjust levels for each scene? I'm asking because I don't know how much capture data is needed to do proper filtering after capture. Setting levels to prevent out-of-bound conditions for the whole tape is the easiest approach but will mean the histogram range for some scenes will be narrow. Separate levels for each scene means multiple files but each scene will have the widest permissible histogram range thus more data.

I think I know the answer (separate levels for each scene), but I thought I would find out what other users think.

I have started adjusting levels for each scene. Now I'm facing the issue that sometimes the Mitsubishi HD2000U or the VC500 is outputting different levels for a given scene. I adjust the levels for a scene and when I pause and restart or rewind and play the scene again something has changed the levels. My preset levels are the same but the incoming signal is different. This would mess up a capture if the level has changed during capture from what I saw during preview. Has anyone seen this?

Thanks,
Steve

Few users who have 4-hour and 6-hour tapes would capture an entire tape at one time, although a 2-hour or less tape is usually capped entirely. And no one expects perfect captures. What they do is set up levels for a worst-case scenario, also keeping in mind that some occasional overflow is acceptable and can be corrected later. It's far easier and faster to make specific corrections after capture -- and you'll have to doctor certain segments anyway, no matter what you do. Start-stop-rewind-restart short captures create terrific wear on the tape and the player.

Thanks for the info! I thought the answer would be the opposite.

My tapes are two hours long. I logged levels and time information for 32 scenes on the first tape. It was a tedious process and I was not looking forward to setting up and capturing up to 32 separate scenes. My biggest concern was the big difference between indoor and outdoor shots.

I'll capture the whole tape using the worst-case settings and see what happens. I guess if a few scenes need more data I could recapture them. When it's time to filter the video, would it be best to chop up the video into separate video segments so different filters and filter settings can be applied more accurately to each type of lighting condition?

Again thanks!
Steve

As for myself and most users I'm familiar with, corrections are made on a segment-by-segment basis, as the need arises. You might find some stretches of video that need very little work or none at all, then a few that are stubborn challenges, and many that are a breeze. That's the nature of analog video -- from order to chaos, minute to minute. I've had digital recordings off cable that were cleaned up, edited, and on a new DVD inside of an hour. And I've had 90-minute tapes that took 4 weeks. Analog is a different world from digital, but the idea is to turn that disorder into consistent quality as best as one can within reasonable limits.

I always set a base level for the tape, and capture.

If something important (ie, something that won't be discarded, or just not prominent in the final piece) is just way off, then it's re-captured with the proc amp. When simply archiving, I'm more apt to just let it ride at the set level, with only minimal re-captures.

Because when you fix every segment, you're not talking a "tape" anymore, but a clip. Many, many, many clips. And that takes time. If personal, and I don't really care, it's not corrected. If for pay, but I'm not getting compensation commensurate to the work load, it's not then either.

Video can be a rabbit hole, a trap, a time vampire. You can restore and tweak for years, if you get OCD about it. Obviously you need to capture well, and pay attention to error correction (restoration). But at some point, you can go overboard. One of the real arts behind video is learning where that cutoff is. Hacks and amateurs quit early, and make crap. Newbies and the OCD often overdo it (George Lucas, anybody?). Seasoned pros tend to know when enough is enough (or not enough is not enough).

As sanlyn also mentions, you rarely capture more than 2-3 hours of video at once. The main reason has nothing to do with time, and everything to do with the tape. Most 6+ hour tapes don't have perfect tracking throughout, especially if from multiple cameras/VCRs using the same tape. You'll often capture several segments in 6 hours for that reason alone. The VCR overheating can also cause some issues. Stuff gets hot after several hours! Never leave a capture unattended, check what's happening at least once every 5-10 minutes.

If it truly is a case of indoor vs outdoor, two basic exposures, then capture the tape once tweaked to each. Discard all bad exposed footage with a post-capture edit (and lossless re-save AVI or MPEG, do not re-encode it).

It's funny you mention rabbit holes. I seem to fall into multiple holes every day on video, regaining on-line privacy (yea, right), Microsoft issues, ad nauseam. Since I don't know what filters to use for different video situations, I expect to really get dirty falling in and climbing out of holes. I also get trapped underground because each description of different items on sites such as Avisyth's wiki site takes me off in different directions and other sites to where it's sometimes hard to remember where I started.

I like your idea of capturing the whole tape twice at different levels and tossing out the bad sections as part of picking the best image for each scene.

Thanks very much for the help,
Steve