VHS straight capture, scene-to-scene capture, or what?

[JasonCA]

Hello Viewers,

So, I have a bunch of video tapes that I want to capture; a lot of it is family video and is therefore important at least to me. Essentially, I would like the capture to be good enough to where if the tape was lost or destroyed, the video capture would suffice in its place. The truth is, we can't keep the VHS tapes forever and the VHS tapes do age over time. So...

Do I run each VHS tape through the VCR and capture it on one pass? Or, do I have to do scene-to-scene capture? Or should I possibly be doing a combination of the two? I want archival quality of the the VHS tapes.

Essentially, I want to extract all that analog VHS video data off the VHS tape and capture it in the digital world. So, I tend to think that if I capture at 720x480 in a lossless data format, that I should be able to IN ONE PASS capture all the analog data off the VHS tape in a digital format? You may say that i have to have a TBC, Proc Amp, ...etc. Fine...assume all of that is setup in the chain. Could I leave the settings in the chain exactly the same and capture a VHS tape to the digital world on one pass?

My guess is it may not be that simple. And if so, please convey to me why I would need to sit through each VHS tape (scene by scene correction?) in order to capture all the VHS video that I would otherwise MISS if I simply did it via a one pass approach?

I would also like to think that I could essentially suck all the good VHS data off of it in a digital format, and then years later do post processing on the RAW lossless VHS data? But, this assumes that what I have captured in a digital format is then TRULLY and HIGHLY like what is on the VHS tape. If not, please tell me why it's so difficult to get the full range of signals off the VHS tape and captured in a digital format? Why can't a digital data fully represent the full range of analog signals? Or, maybe digital data can capture everything?

Please tell me where I am wrong?

Thanks,

Jason

[robjv1]

Quote:

So, I have a bunch of video tapes that I want to capture; a lot of it is family video and is therefore important at least to me. Essentially, I would like the capture to be good enough to where if the tape was lost or destroyed, the video capture would suffice in its place. The truth is, we can't keep the VHS tapes forever and the VHS tapes do age over time. So...

VHS tapes certainly do age, but keep in mind that the doomsday proclamations about them rarely hold up to actual real-world situations. I have tapes more than 30 years old that are just fine and show no signs of deterioration. Having said that, I'm definitely not trying to dissuade you from your goal -- it's good to have important footage in more than one format and in more than one space.

Quote:

Do I run each VHS tape through the VCR and capture it on one pass? Or, do I have to do scene-to-scene capture? Or should I possibly be doing a combination of the two? I want archival quality of the the VHS tapes.
Essentially, I want to extract all that analog VHS video data off the VHS tape and capture it in the digital world. So, I tend to think that if I capture at 720x480 in a lossless data format, that I should be able to IN ONE PASS capture all the analog data off the VHS tape in a digital format? You may say that i have to have a TBC, Proc Amp, ...etc. Fine...assume all of that is setup in the chain. Could I leave the settings in the chain exactly the same and capture a VHS tape to the digital world on one pass?

My guess is it may not be that simple. And if so, please convey to me why I would need to sit through each VHS tape (scene by scene correction?) in order to capture all the VHS video that I would otherwise MISS if I simply did it via a one pass approach?

I would also like to think that I could essentially suck all the good VHS data off of it in a digital format, and then years later do post processing on the RAW lossless VHS data? But, this assumes that what I have captured in a digital format is then TRULLY and HIGHLY like what is on the VHS tape. If not, please tell me why it's so difficult to get the full range of signals off the VHS tape and captured in a digital format? Why can't a digital data fully represent the full range of analog signals? Or, maybe digital data can capture everything?

A couple of things you should determine based on what you've said --

You've stated that you want 'archival quality' of the VHS tapes. You should ask yourself, what does that really mean? If it means that you want your VHS stuff to look 'as good as possible' (a subjective measure) then you're really getting more into the topics of restoration than "archival". The point of having a proc-amp, fancy VCR with DNR and TBC, etc is to improve the signal, not simply archive it. It may seem like an unimportant distinction, but it is one that will impact the way you make decisions about how to progress in a project all the way down the line.

To further complicate things, 'archiving' a VHS tape starts with the VCR first of course, and what is 'the best VCR' to use is not so cut and dry, for both technical reasons due to issues with the decks themselves (i.e. they don't play your tapes well) and subjective concerns as well. For example, many people prefer to use the JVC or Panasonic SVHS Line-TBC decks as their primary VCRs, because like many well built decks, these decks tend to reproduce a pretty faithful and neutral VHS image, often without flaws inherent to many consumer decks (white line dropouts, overly sharp and grainy image, poor tracking range). However many of them also have filters to deal with flaws encountered on VHS playback (signal issues, chroma noise, grain, etc) and not everyone is in agreement that they create the most palatable image and you'll see people on both sides of that argument. Personally I think they do a good job (with my tapes) and the plus is that they require less software filtering on the backend, but others won't have it and think they are antiquated and swear by capturing their footage uncompressed and working with everything in software. Just as a side note, be aware that there are some things that cannot currently be corrected/addressed in software though and must handled on the playback device.

The bottom line is that there is no one deck that is right for everyone (nor one that will be guaranteed to do a good job with all of your tapes). There is no one deck that will "do the best job extracting every last piece of good analog data off of the tape" because your opinion of which does the best job will vary from the opinion of others. Other subjective measures like how sharp you like the output to look (one person's perfectly sharp is much to soft to someone else and vice versa) and how much control over each element of the signal you want will influence your decisions and impact your time commitment to the project (and definitely also the amount of money you'll spend).

The VCR is the most important thing you'll buy in the whole chain and will make the most difference in the perceived picture quality, so choose it wisely. Capturing lossless/lossy are both valid ways of approaching this, but they are not as important factors as the VCR itself is, so I can't emphasize that enough. That's something to really think about and really something you should really put into action -- if you have the cash, I would try buying several different VCRs (with return policies) and sit down with a big cross-section sampling of your tapes, spend several hours running tests on them and make two lists -- the ones with the best overall image you and the one that plays your tapes the best. If you can afford to at least own the top one in each column, you'll be in good shape. Most of us have multiple VCRs, some fancy and some consumer decks, you'll often find you need at least one of each if you have a large collection of tapes recorded on multiple VCRs and in multiple tape speeds (SP/LP/EP) over a long period of years or stored in varying conditions.

Buying a bunch of VCRs just to run tests might sound extreme -- but I think a lot of people that do this as a hobby would agree with me, you end up doing it anyway in the long-run. You'll capture a bunch of stuff and think it could never get any better than it is right now, but then you'll get another VCR and you love that output much better, and end up doing it all over again. So take the benefit of our pain and experience and save yourself the trouble! haha

As far as the rest of your question, it seems to be -- will I have to babysit my captures to ensure that they will turn out okay? The answer is, that totally depends on your tapes and also on what your threshold for 'best quality' is. If you actually test your tapes out several VCRs you'll come to learn what their flaws are. This stack of tapes has scratchy hi-fi audio. In this stack of tapes, there are several short segments recorded across the length of the tape, recorded on different VCRs and so the tracking drifts a bit. Maybe all of your tapes are recorded in LP, which can be hard to track in many of the fancier VCRs. On a tape like the second example, you'll probably need to babysit it or at least spot check it.

The real question is, how upset will you be when you find out your tape has an error that is possible to correct, after you've already captured it? If Grandma's birthday party has a tracking line in it momentarily during a key part that was just due to the auto-tracking on a VCR you used, is that going to make you feel the whole capture is garbage? If so, then you're going to be the type of person that watches all of your captures, either as they are happening or after they are done, identifies errors, and goes back and captures the messed up part a second time. I have a VCR that will randomly throw a little bit of noise on the frame at intermittent locations -- maybe 5 to 10 frames total in a two hour tape. It didn't bother me, until I realized it wasn't embedded in the signal, it just occurs randomly on playback, which means I didn't have to just accept it and move on, I could actually fix it. Those are the kind of realizations that cause your time investment in these sorts of projects to go way up, because you start spending 10x the amount of time to correct 5% of the errors. To some people that is crazy talk and to others you can't convince them that there is any other option. You have to figure out what type of person you are.

[kpmedia]

Quote:

Originally Posted by JasonCA

Essentially, I would like the capture to be good enough to where if the tape was lost or destroyed, the video capture would suffice in its place. The truth is, we can't keep the VHS tapes forever and the VHS tapes do age over time. So...

All true and wise.

Quote:

Do I run each VHS tape through the VCR and capture it on one pass? Or, do I have to do scene-to-scene capture? Or should I possibly be doing a combination of the two? I want archival quality of the the VHS tapes. ..... Could I leave the settings in the chain exactly the same and capture a VHS tape to the digital world on one pass?

It depends on the tape. Some tapes have uniform recording quality all the way through, because it's from one event, or from multiple events shot in the same lighting. Sometimes the "video" is a single videotape with multiple recordings, and each one has different color quality, white balance, exposure, etc. You'd want to capture that one segment by segment, for the best quality. (At worst, pick a median value, and correct color in software later on.)

Quote:

My guess is it may not be that simple. And if so, please convey to me why I would need to sit through each VHS tape (scene by scene correction?) in order to capture all the VHS video that I would otherwise MISS if I simply did it via a one pass approach?

- Bad tracking.
- Misaligned tape.
- The tape begins to shed oxide.
- Multiple recording modes.

Quote:

I would also like to think that I could essentially suck all the good VHS data off of it in a digital format, and then years later do post processing on the RAW lossless VHS data?

You can. Sometimes years pass between time I spend working on long-term hobby projects. But I know what is and isn't possible out of a tape, so my transfers are fine. In recent years, software has advanced to a point where I can finally fix certain errors, or go back and redo previously "finished" projects.

Quote:

But, this assumes that what I have captured in a digital format is then TRULLY and HIGHLY like what is on the VHS tape. If not, please tell me why it's so difficult to get the full range of signals off the VHS tape and captured in a digital format? Why can't a digital data fully represent the full range of analog signals? Or, maybe digital data can capture everything?

You're not wrong. It is possible to extract all of the data off the tape. The biggest issue is leaving the errors off of the digital version. It's easy to keep the errors, and more expensive and difficult to beat them off with a stick. That's where the good S-VHS/D-VHS VCRs, TBCs, proc amps, audio mixers and other misc hardware comes in.

"Archiving" is often used to mean "as is", while "restored' is making it all that it can be. I actually disagree with that definition. "Archived" video should be restored, with the best possible quality preserved for the future. Unclean video transfers are simply "duplicated" -- not archived, preserved or restored.

robjv1 gave an excellent reply, and it's hard to expand on that.

[robjv1]

Thanks! I actually really like the distinction you made there too between archiving and duplicating -- what's the point of carefully "archiving" something that looks bad or is a mess?

I'm still a big proponent on having the right hardware first in line as kpmedia stated too. I think you really have to look at it as a two pass approach most of the time. Not only are some errors just not correctable in software, but if you use the right hardware first, it'll save you so much time on the other side. There is no reason to purposely save/encode footage with timebase errors, tracking problems, or chroma issues in attempts to remain 'faithful' to the source. Using the hardware first will tend to make the problems in your footage fall into a more predictable range of issues.

[lordsmurf]

Quote:

Originally Posted by robjv1

timebase errors, tracking problems, or chroma issues in attempts to remain 'faithful' to the source

Actually, I'd say that leaving those error is UNFAITHFUL to the actual source laying underneath the layers of crap created by the format. No sane videographer has ever asked to include timing errors, tracking problems, or chroma noise into a video -- outside of trying to make a video that looks like it came from a crappy abused consumer VHS tape. These are flaws and errors added by the tape, and are neither desirable or germane to the original source content.

I consider it clinically OCD to want to leave all tape-induced errors in a digital archive version of the video.

I've seen the crazy no-filter argument online for years (mostly at amateur-heavy sites like VH) and it just doesn't hold up as valid. The only time you need to keep the mangled/butchered version of a video in an archive is when you're dealing with forensic evidence -- but even then, a copy is restored for actual court proceedings, so people can see/hear what's going on.

Even original paintings and documents get restored. The Declaration of Independence has survived only because of this! In fact, it undergoes regular (daily?) maintenance to ensure the condition is holding up as desired. I've never heard an art critic or art historian scream "You've ruined it!" when a painting was restored to original condition.