Newbie cheap way of VHS to digital questions

[grants225]

hi there,

I am new to this stuff, I am reading forum/guides for few weeks and what I understand so far, that it is so complicated and so many ways to do the same thing, what makes me really confused. I have some family stuff on VHS-C (captured with crapy camcorder) some 10 or more years ago, what I want to digitalize them, edit (cut, trim, titles, maybe some color correction due to wrong white balane set during filming, add background music etc), and make it a little more watchable on bigger tv screen. Editing movie is not a problem, but getting it correct prior inserting into editing software is still no-no for me. I am NOT focusing on best quality, which for now I know needs serious improovment into hardware, cables, pc, capture device and so on and as I am not going to be professional this is not for me.

So after two weeks, I am still in the very beggining (capturing issue), I already made hundreds of tests with various settings. Various guides says sometimes in opposite and blah I need someone advise to speed up my learning curve and show the right way to do it. And bellow is my questions, setups, where various notices or comments is really very appreciated:

First few words on my hardware (I am writing not from the home, so I can not check exact names. If you think it is necessary, I will add it in the next post):
-JVC low end VHS-C camcorder somewhere from ~1998, but I was surrprised that it is still working (battery of course dead)
-Samsung VCR Diamond (get a gift some ~15 years ago, probably peace of crap, no any advanced settings, only one pair of composite out), was surprised, it is still working!
-PC for capture quite old P4 running Win XP. ATI Sapphire Radeon 9000 PRO VIVO (AGP video card with composite and S-Video video in), also no name Bt878 PCI TV/FM capture card (anyway, I get better results with ATI, some comparisson review I read says the same, so stick to ATI), two hard drives, integrated sound card. For video editing, processing is another new PC with i5 win7 64b (ok, not i7but for my needs it is enough for short movies even in 1080p 24fps captured with Nikon D5200). I can not move capture cards to this pc as no AGP slot, or no drivers for win7
-TV LCD samsung 32'' 1080p
-media player WDTV Live, plays quite a lot file formats in 1080p, including .mkv, divx/xvid, wmv, m4p and other popular formats. Connected to TV with HDMI cable. Will be used to play converted VHS-C. I guess it can play interlaced files as well.
- Panasonic HDD DVD-R, not using for capture purposes
- external USB HDD for media storage

Things changes quite fast, TV could become bigger, higher resolutions can come, USB HDD could be replaced with NAS. But as topic says, lets make it cheap and see what we can get the best from it and be prepared to >4k TV era.

Software:
I love VirtualDub, it is free, light, flexible and I dont see reasons to look for alternatives unless you say different. Now the more interesting and more confusing part of (sorry) already quite long post.

VirtualDub settings:
-resolution I set to custom 352x576 (I am PAL guy). I read some articles, where they suggest to run some capture from DVD test source and identify exact capability of the capture card. Is it really necessary? Is resolution is ok? (I dont get better quality with 702x576 and dont understand how it can be)?
-choose data format YUY2
-no video compression (with old PC I have 0 droped frames)
-no audio compression
-PAL-B, composite input, device ATI Radeon
-noise reduction enabled, treshold +20 from the left side
-capture settings 25fps
-capture timing/ selected "Sync audio to video by resampling the audio to a faster or slower rate" NOT sure about this, I have some sync error during capture +/- 5ms, is it normal?

So what you would you advice to change?

Second, I know it will sound stupid, but ~20GB file for 30min seems quite heavy for me. I want it to compress prior editing, I tried using xvid with high bitrate (9000 and above), what makes ~1.5GB more pleasant to the my eye. I know I am loosing some data on this conversion, but visibly I dont see big difference compared to file without this intermediate procedure. I would say that xvid even softens a little noise, what looks better scalled on 32" screen. Comment or suggest on this, please.

And last concern is resolution in editing program and final format to choose. If I caputre 352x576 25fps interlaced, should I resize? Make square pixels, preserve interlacing? Do xvid preserves interlacing? What format would you suggest for final storing edited VHS-C, which would preserve interlacing? Do you make (and how if so) 4:3 to 16:9 transformation? Some guides advises to cut sides of the frame after capture (noise from audio in bottom) and then resize (it is better to do in vdub or to scale in editing programm)?

really thank for everyone in advance for any input

[DB83]

That's is a lot to cover in one topic. But I will try.

Your chosen SAR at 352*576 in your capture scenario is IMO wrong. The advantage of that SAR is for low bitrate Mpeg2 capture when you can go down to 2500 kbps. You are capturing uncompressed so capture at full D1 ie 720*576. The capture size will be the same but the quality will be higher.

There is no sense whatsover in converting to xVID BEFORE you edit - if I read you correctly - you have lost a lot of quality, even with that high bitrate, which you can not get back and made the editing process much more difficult. Better to capture as Lagarith or Huffyuv and edit without any re-encoding.

XVID is a distribtion format with lower bit-rates. You really need not fo above 2000 kbos with that. But H264 will give you better quality that H263(xVID).

Finally, for now, store at the highest quality ie the original capture. You can always come back later and re-edir. When you compress, as I mention above, it will not be so easy to edit.

[sanlyn]

I'd echo DB83. You should not be capturing low-quality analog source into lower-quality lossy compressed formats. Capture to lossless Lagarith or huffyuv AVI at 720x576, which will be your final DVD output size (it wall also do for standard-def BluRay, but that's working ahead of where you are now). IF your sources look anything like the usual tape source, you will definitely need to make your initial processing adjustments in the captured YUV colorspace, before you try working in VirtualDub. There is a proper and an improper way to open YUV captures in VirtualDub RGB without ruining luma and chroma levels. You don't have to become an Avisynth guru, but there are simple and very basic steps you should take before processing video in RGB.

A 90-minute capture in lossless compressed AVI at 720x576 would run about 35 GB or so. If your source is long-play 6-hour stuff, you're in trouble but you can work your way through it a piece at a time.

You will need an external hard drive. Trust us on this. A portable USB drive such as a little Toshiba Canvio would suffice. Don't capture to that drive and don't do cleanup processing on it (they're too slow), but you'll definitely need it for storage. A fan-coooled AC-powered external enclosure with SATA drive would be somewhat better, but costs more and isn't absolutely necessary.

[grants225]

thank you for suggestions, capture results are getting better. One more confuse on color scheme, I set to YUY2 as some guides suggested without understanding much. No I had read what is all this YUV is about at http://www.animemusicvideos.org/guides/avtech/colorspace.html

and not any more so sure about YUY2. If I capture in VirtualDub YUY2 and edit it in Premiere Pro, link above says Premiere works in natively RGB and will make color scheme conversion (what is bad). Should I capture RGB24 in VirtualDuB instead of YUY2? Huffyuv compressor same link above says can handle both YUY2 and RGB.

[themaster1]

you should not and stay in y'uv

[sanlyn]

Conversion from YUV to RGB is handled in different ways by different applications. It might not even be necessary to go to RGB. There are proper and improper ways of doing it. In any case, once you encode your final output, the result will be in YUV. If you capture in RGB24, the new capture size will be at least twice the size (or larger) of a YUY2 capture, and some of the work you need to do will have to be done in YUV anyway.

You appear to be using Premiere as a simple cutter and joiner. That's a very expensive app just for "editing".

Meanwhile, the discussion at this point is moot. We have no idea what your original looks like, so it is nearly impossible to go into more detail about possibilities.

Should I capture RGB24 in VirtualDuB instead of YUY2? Huffyuv compressor same link above says can handle both YUY2 and RGB.

No. Master1 is correct. YUY2.

[sanlyn]

Oops. Sorry, posted twice. I blame the forum's software !!!

[grants225]

You appear to be using Premiere as a simple cutter and joiner. That's a very expensive app just for "editing".

I have Adobe Premiere Pro CS6 trial version, it is cheap . It is our target to be cheap.
I want to use premiere mainly because for Warp stabilizer and Fast color correction filters (white balance problem during shooting indoors). Also I saw bunch of tutorials on youtube and know how to do theese things.

Meanwhile, the discussion at this point is moot. We have no idea what your original looks like, so it is nearly impossible to go into more detail about possibilities.

It is filmed on low end VHS-C camera, and VHS-C cassetes laid down for >10 years untouched. It is not pro / or definitelly not a good source and it needs some minimal post correction. I will include one frame later.


My mentioned link which confused me says:

Programs:
Premiere, and almost all video editing programs, work in RGB because it's easier to deal with mathematically. Premiere demands that all incoming video should be in RGB32 - or 24-bit color with 8-bit alpha channel, specifically, and will convert the YUV footage you give it to that format for processing. Even Premiere Pro which hailed itself as being able to support YUV formats can only support 4:4:4 uncompressed YUV which is hardly any different from RGB. The native DV support is useful but it still doesn't warrant all the hype as very few of the plugins (including Adobe's own sample code) actually use the YUV support at all.

But I do not understand sugesstion s(sorry, I am not english native speaking) : a) you suggest to use captruing in YUY2 -> editing in CS6 (adobe will convert to RGB) -> exporting to final movie in YUV format without problem. Or b) do not use premiere?

[sanlyn]

You should capture in a YUV color space because that is the color space used by your source. Many color and luma level corrections, and noise removal -- especially from low-end analog sources -- can BE REPAIRED ONLY IN THE INITIAL YUV CAPTURE'S UNALTERED COLOR MATRIX.

Meanwhile, the discussion at this point is moot. We have no idea what your original looks like, so it is nearly impossible to go into more detail about possibilities.

It is filmed on low end VHS-C camera, and VHS-C cassetes laid down for >10 years untouched. It is not pro / or definitelly not a good source and it needs some minimal post correction. I will include one frame later.

Interesting. One frame? That would not be useful, neither to us nor to you. Your video was not "filmed", it was recorded onto magnetic analog media. Your statement that it would need "minimal post correction" does not match your description. Low-end tape in that condition is going to be seriously affected by processing in RGB unless it is first prepared for it in YUV. This has already been mentioned, and the more you describe this project the more it appears that you lack an understanding of the media, and that you are ignoring our advice.

To repeat:

Meanwhile, the discussion at this point is moot. We have no idea what your original looks like, so it is nearly impossible to go into more detail about possibilities.