Skip to main content
Topic: VHS ransfer (Read 4513 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

VHS ransfer

No doubt it is more complicated than yes/no but so far I have not seen anything that even approaches the topic.

VHS perceived picture quality varies noticeably depending upon the tape quality and the recording speed, and probably other factors. However, if the goal is to get the best quality digital copy of the tape, whatever is on it, I suspect those aspects don’t matter as far as the question of what ADC to buy  and what computer file format to use. From this theoretical best recording to the computer, other formats, lossy or otherwise, can surely be created after the production work is finished.

Having looked at the ads for a number of capture devices (only VHS to computer hard disk are of any interest here), and a couple of how-to web pages, the most information I have found is that an hour of video will require from 750MB to 3GB of storage space.

That difference could come from selecting different sample rates or bit depths, but I suspect those considerations play a much lesser role than with audio recording. Nowhere were there any specifications suggesting there is any choice for bit depth or sample rate, so my guess is that the file size difference is due to capture format.

That, however, is my query.
1) Do all the video capture devices produce the same input to the computer or is there some hardware specification(s) to look for whose existence I have missed seeing?
2) Is there a generally accepted best capture format that will allow one to make post-recording changes before deciding on a delivery format such as DVD, etc.
3) Are all, or at least most, capture software programs likely to support this format?
4) Are there recommendations based on experience for hardware and/or software? Only fairly minor and simple software editing and fixing are anticipated, not anything approaching movie maker complications.

VHS ransfer

Reply #1
Quote
Nowhere were there any specifications suggesting there is any choice for bit depth or sample rate, so my guess is that the file size difference is due to capture format.
Usually, the basic spec is for resolution (the number of pixels).    i.e. PAL (most of the world) DVDs are 720x576 at 25 frames per second and NTSC (North America & Japan) are 720 x 480 at 29.97 FPS.  The next spec is usually the bitrate (kbps) which is the number of bits in one second of video, and the compression format.    A higher bitrate takes up more space and gives you better quality.

DVDs are MPEG-2.  Commercial DVDs usually have a bitrate of around 5000kbps, which can give you about 2 hours of video (with compressed Dolby audio) on a single-layer (4.7GB) DVD.    With MPEG-4 (and all of the MPEG-4 variations) you can get a smaller file without loosing quality.


Quote
1) Do all the video capture devices produce the same input to the computer or is there some hardware specification(s) to look for whose existence I have missed seeing?
  There are differences, but for VHS capture I don't think you have to go overboard.  The EzCap costs about $35 USD and has a reputation for working well and having reliable drivers, etc.  I wouldn't go any cheaper than that, and you have to watch out for similarly named products like EazyCap, etc.    Hauppauge and ATI have been making video capture cards & devices for a long time.  There are basically two types of devices.  There are devices with built-in "hardware" compression, and devices that use software and the computer's CPU for compression.  A device with video compression built-into the hardware, and devices where the video compression is done in software.    Hardware compression makes it easier on the computer because the computer doesn't have to do any "work" and there is less data to deal with.  So in general you are less likely to get glitches.  I believe all USB capture devices have hardware compression because the USB bus can't handle uncompressed audio/video.


Quote
2) Is there a generally accepted best capture format that will allow one to make post-recording changes before deciding on a delivery format such as DVD, etc.
As a general rule, less compressed files are easier to deal with.    DVI is easier than MPEG, and MPEG-2 is easier than MPEG-4, or the other highly-compressed formats.    For example, people that are editing DVI (usually recorded with a DV camera that uses DV tapes) have very little trouble.  At the other extreme, people who try to edit highly-compressed video captured with a cell phone or with a "still" camera that optionally captures video often end-up with trouble when editing, such as with video "freezing" problems, videos that won't render after editing, or audio & video going out of sync, or other problems & glitches.  (Note that DV is standard definition an is no longer popular.)


And, virtually all of these video formats use lossy compression.    That means you are going through a 2nd lossy compression step if you edit and re-render.    So again, it's generally better to capture to DV or high-bitrate MPEG-2, even if you want a smaller more compressed file as your final format.

Quote
3) Are all, or at least most, capture software programs likely to support this format?
The capture software generally comes with the capture hardware, so it's compatible with whatever you buy.    Video editing software generally works with most formats, but there are many-many variations of every format, and like I said, some of the highly-compressed formats are more likely to be troublesome (even if the editing software is supposed to support the format).


Quote
4) Are there recommendations based on experience for hardware and/or software? Only fairly minor and simple software editing and fixing are anticipated, not anything approaching movie maker complications.
I had a Hauppauge PCI card and I captured most of my VHS tapes with it, but the one I have is now obsolete and does not run on Windows 7.  I bought an EzCap, and I might have used it once.   

For editing I have Corel VideoStudio (about $70 USD) and another less popular MPEG video editor, a special-purpose DVD authoring application, and a handful of other (mostly freeware) special-purpose audio/video tools.   

If you want to make DVDs & Blu-Rays Video Studio can do that too.    DVD & Blu-Ray authoring is different from video editing...  You can't simply "burn" the video files onto a disc, they have to be in a particular structure/format (and you'll usually want a menu).  Although, there is a good chance your Blu-Ray player can play a non-standardized disc.

VideoHelp.com has tons of information about video capture. editing, making DVDs, and all kinds of information about digital video related hardware & software.

VHS ransfer

Reply #2
What about no compression or lossless compression? That would unquestionably be my choice if I had the choice. That is the basic thing I most wanted to find out, although I tried not to limit the scope of my questions. Maybe lossless could not be accomplished with a USB connection but is it otherwise available? (I would guess what I want at least loosely corresponds to "raw' from a digital still camera)

Whether any of the material will ever go onto DVD or any other optical disk, rather than just remaining on hard drive, is open to question but can I start off without it being limited?

VHS ransfer

Reply #3
[quote author=AndyH-ha link=msg=879147 date=1414822681]... Whether any of the material will ever go onto DVD or any other optical disk, rather than just remaining on hard drive, is open to question but can I start off without it being limited?[/quote]

Do the math. Assume a capture device that captures 8 bits per colour per pixel, therefore 24 bits / 3 bytes per pixel. (Some do 10 or 12 bits per pixel.)
For NTSC, that's 3 bytes x 720 pixels x 480 rows x 30 frames per second = about 30 megabytes per second. 1.8 GB / minute, 108 GB per hour. Less than 10 hours on a 1 TB disk.
So it's practical to capture "raw" with modern hard disks and processors, though you'd need to use a PCI card capture device - I don't know of any USB ones that capture raw, and it would probably have to be USB 3.

Assuming you're capturing from standard VHS tapes, you don't need to capture at such a high resolution.  VHS tape is (very) roughly equivalent to 333x480 digital pixels. You can capture at 360 x  480, which halves your data rate and disk space requirements.
Regards,
   Don Hills
"People hear what they see." - Doris Day

VHS ransfer

Reply #4
Ok, the amount of data is a real constraint effecting storage space, work space, and processing time. Is the result of this lossy video compression markedly different than lossy compression of audio. What about the audio that frequently goes with the video?

I have transferred audio from many cassettes. Even though the production on those cassettes was a finished product that I don't need to modify, I typically applied five to seven data operations across the entire file to improve various aspects, sometimes several more file wide operations, plus "local" manipulation of problem spots. For very good reasons I would never consider working in mp3, even though that is often the final format I produce after everything else is done.

With such large files for video, they surely must be written to disk again and again if one does much work on them. If it is of necessity written in a lossy compressed form, will there not be some significant data deterioration as there would be with stand-alone audio? (not to mention the audio part of the AV project that must be kept in synch with the video) Is this loss a compromise that has to be made with processing, just as a matter of course?

VHS ransfer

Reply #5
some thoughts;

With standard-definition video you can capture/render in uncompressed modes on a modest modern pc, however there are some lossless (ut-video seems to be popular on win platform) and lossy video formats that you can use for the purpose of intermediates (I use prores a lot on mac, there is also cineform, ect).

Video editing app can render effects/color corrections and such in lossy format for playback/preview purposes, but will re-render uncompressed when you are ready for final (If this is non-raid system, perhaps one would want to put that "cache" location to a fast disk, different than the one with captured material).

A lot of video editing apps also have the ability to compress directly into final formats, like mpeg2 for dvd.

On windows there are various frame-server plugins for various editing apps, that you can use as zero-size intermeditate for serving to your specific external mpeg2 encoder (I used to use http://www.debugmode.com/frameserver/ ). It is possible to build some crazy paths like : premiere > debugmode > virtualdub(filters) > avisynth(filters) > encoder

I can get more "thoughts" if you need ....

p.s. Not sure what you mean with "standalone audio", but basically when you are done with the lenght related video/audio editing, you can extract the audio part and sweeten/mix in audio specific application (possibly with proxy-quality video preview), I mean, audio is audio (my usual workflow is to throw one audio.omf and one video.dv at the audio guy).
PANIC: CPU 1: Cache Error (unrecoverable - dcache data) Eframe = 0x90000000208cf3b8
NOTICE - cpu 0 didn't dump TLB, may be hung

VHS ransfer

Reply #6
A clearer expression could have been found, but by stand-alone audio I mean an audio only recording, the only type I've worked on, which is unlikely to come out optimally if one tries to do very much using a lossy compressed format from beginning to end.

Obviously I will have much to learn if I need to/decide to pursue video very far. Now at least I know that there are, possibly affordable, means on the PC platform.  I 'm thinking that the best course, however, is to get started with some tools of the type DVDdoug wrote about and find out when and where they become unsatisfactory.

VHS ransfer

Reply #7
some thoughts, part 2;

[quote author=AndyH-ha link=msg=879337 date=1414964813]if one tries to do very much using a lossy compressed format from beginning to end.[/quote]
Audio part should be always uncompressed throught the chain, even if you capture to DV that means unc. audio (16-bit audio at 44.1 kHz).

p.s. Master backups, what I do to regain disk space over time (Talking about HD here);
a. 1st three months:
material: hq 422 prores master (if that was a capture from tape) or whatever the original format was
video master: hq 422 prores master
audio master: omf & pcm mix.

b. after some time:
material: x264 8bit (Unless it was small allready, like AVC material) or delete.
video master: x264 8bit (roughly around DV bitrates)
audio master: omf & alac/flac mix < stays the same basically

(projects are automagically rsynced to 2nd drive on the network, captures are usually not)

(At certain time I had my own 40 tera fiberchannel drive and I kept everything in stage a., so it depends ...)
PANIC: CPU 1: Cache Error (unrecoverable - dcache data) Eframe = 0x90000000208cf3b8
NOTICE - cpu 0 didn't dump TLB, may be hung

VHS ransfer

Reply #8
You are asking in the wrong place Andy. Visit videohelp.com forums and DigitalFAQ forums.

Unless these are valuable home movies, or other priceless exclusive footage, don't bother.

If you're still reading, you need:
  • An expensive working VCR with a line-TBC, or a cheaper working VCR + a device (like certain DVD recorders and certain camcorders) that acts as a line-TBC when you loop a video signal through it's line in > line out.
  • A capture device that lets you adjust the capture levels (you don't want to clip video signals any more than you want to clip audio signals - you lose the brightest + darkest parts)
  • Capture to full SD resolution. VHS is about half horizontal resolution, but there's no great benefit (and a little more pain + pitfalls) to capturing half rez, so don't.
  • Lossless or near-enough lossless capture (I use DV. It's not perfect, but it's near enough for most people).
  • Don't capture straight to low bitrate MPEG unless you intend to do no processing with the capture. It's a low quality source to process from, and too lossy.
  • If you have hours to burn, some cleanup (there's a world of AVIsynth scripts).
  • Decent encoding to whatever.


You will almost certainly end up compromising on some of these, and some are more serious than others. Getting the tape to play properly, and not messing up the capture, are most important. Lossy vs lossless is really a secondary issue compared to that.


Decent VCRs are expensive and increasingly hard to find in good working conditions.

VHS pictures aren't stable: they always wobble. How much depends on the tape and the VCR. A TBC (Time Base Corrector) attempts to stabilise them. Frame TBCs are for removing copy protection and avoiding the capture dying when there's missing/bad frames. They don't improve quality (quite the opposite, some of them!). A Line-TBC is what you need, somewhere, unless you are lucky and/or don't care.

Raw VHS captures are noisy and despite being soft (=low resolution) eat bitrate (lossy or lossless). Low bitrate lossy goes blocky and smudgy.

AVIsynth doesn't typically require intermediate files. It's a processing script that sits between your raw captures and the final encoder. Hence no intermediate lossy stages. If you do need intermediate stages (for speed, simplicity, or procrastination), you might as well use lossless files to store the results from each stage.




If you haven't played it recently, you will have forgotten how awful VHS looks. All these measures make it look slightly better. It's never going to be a pro DVD, never mind BluRay.

The most important thing is spotting any specific problems you might have, and correcting them early on - rather than capturing every single tape you have, and then finding out the captures are no good. Problems include... Interference. Tracking problems. Dodgy cables. Dying VCR. Dropped / corrupted frames. Audio de-sync after capture/edit. Doing some stupid processing that messes up the interlacing. Over processing the result. Dying before you've finished the job. Copying more than anyone is ever going to watch and never editing it.

Cheers,
David.

VHS ransfer

Reply #9
Doing some stupid processing that messes up the interlacing. Over processing the result. Dying before you've finished the job. Copying more than anyone is ever going to watch and never editing it.

nice lineup 
PANIC: CPU 1: Cache Error (unrecoverable - dcache data) Eframe = 0x90000000208cf3b8
NOTICE - cpu 0 didn't dump TLB, may be hung

VHS ransfer

Reply #10
Obviously there is much to learn on the front end. It probably comes together better once one spends some time and effort with it.

Is it the case that ALL VCRs output only half the number of lines per frame of a full NTSC television? If so, that should logically mean that the recording devices only create that smaller number of lines per frame, otherwise someone would have produced a means of playing back what is there. Maybe some of the cameras capture in higher resolution but require a different playback device to display it? What was used to create, and play, the material that was broadcast, back when television was analogue? Surely that used the full capabilities of the system?

What if one is trying to deal with home video of the earlier digital age? Is it as complicated to get a good on-computer capture or is the material already in a reasonable format to allow simple transfer to a hard drive?

VHS ransfer

Reply #11
[quote author=AndyH-ha link=msg=879428 date=1415044834]... Is it the case that ALL VCRs output only half the number of lines per frame of a full NTSC television? ...[/quote]

No. All VCRs output the "full frame" as was recorded. NTSC broadcast (and VCR recorded) video is interlaced. Each frame is split into two fields. Take a piece of paper. Draw a series of parallel lines on it, spaced slightly apart. That's one field. Now draw a second set of lines, each line in between two of the first lines. That's the second field. 60 fields per second, 30 frames (pairs of fields) per second.

The head drum in a VCR has two heads. Each lays down one complete field as the tape moves past. so that the two fields are adjacent to each other on the tape. When you press "pause", both fields are scanned from the same field on tape, which may be where the myth came from.

Interlaced scanning reduced hardware and bandwidth requirements. Modern non-broadcast systems such as DVD can record and playback in "progressive" mode, where all of the lines in a frame are output one after the other instead of being interleaved.

[quote author=AndyH-ha link=msg=879428 date=1415044834]What if one is trying to deal with home video of the earlier digital age? Is it as complicated to get a good on-computer capture or is the material already in a reasonable format to allow simple transfer to a hard drive?[/quote]

It depends... there were several competing formats in the early days. You may need a "swiss army knife" type tool such as ffmpeg to convert them into a format suitable for editing.
Regards,
   Don Hills
"People hear what they see." - Doris Day

VHS ransfer

Reply #12
http://en.wikipedia.org/wiki/Videotape and something about analog video tape horizontal "resolution" and bandwidth (yes, vertical res must be full, otherwise interlacing could not be possible).

Quote
home video of the earlier digital age?

Was there anything else than DV and similar?
PANIC: CPU 1: Cache Error (unrecoverable - dcache data) Eframe = 0x90000000208cf3b8
NOTICE - cpu 0 didn't dump TLB, may be hung

 
SimplePortal 1.0.0 RC1 © 2008-2020