Log in

View Full Version : Betamax capture to DVD - need some help!


skaven
4th January 2005, 06:08
Ok - so I've spent a few hours reading through the analog capture guide but I'm still a bit confused as to what would be the best resolution to capture at. Let me give you guys a general rundown of the source and PC hardware:

Source: Betamax home videos, old Sanyo player with component output and mono audio.

PC: 1.6ghz, 512mb ram, 120gb ATA133, 30gb ATA133, Sony DVDRU510
Capute card: Avermedia DVD EZmaker PCI (Philips 7130 chipset?)

I would like to capture without losing quality (and hopefully being able to clean up the videos) but I would also like to maximize the amount of video per DVD. If I won't gain any quality by going with a higher resolution capture, is it worth doing?

The more video I can fit per DVD the better as I'm giving this to various members of the family and there are many hundred hours of source video.

My plan was to use Virtual VCR to capture and then VirtualDub to edit/clean.

I also own Studio DV 8.0 - not sure if its even worth using though. Ideally I would like to have menus (with thumbnails of the scene and perhaps a description) and chapters to be able to move around.

Any tips would be greatly appreciated!

Arachnotron
4th January 2005, 17:56
I would say:

- use an component to Y/C adapter and cap Y/C

- cap at 704x576/480 (are you in NTSC or PAL land?)
- process
- encode to 704x576/480 MPEG2 interlaced
- also resize down and encode to 352x576/480 MPEG-2 interlaced
- play back on TV, compare and see if you can live with the results :)

There have been many heated debates about the ideal resolution for tape captures on this and other forums. Theoretically, for optimum results with betamax you should be somewhere halfway between 312 and 624x576 for optimum results. (Many people will disagree with this though :))

But for an older tape you may get by at a lower resolution too, and with a final target of DVD you don't have many resolutions to choose from anyway :)

So, simply try.

Keeping the stuff interlaced simplifies filtering. Don't use 352x288 MPG1. There you are really starting to loose vertical detail.

mic
10th January 2005, 19:51
FWIW, personal opinions and all...

If the final target is mpg2, try and do mpg2 capture, all I frame preffered, at the highest size/bit rate you can manage. Don't have to go through color conversions etc., and this boosts quality a bit. Use DGIndex to work with the captured mpg2.

For final format, if you're trying to maximize space, might consider SVCD format on DVD, though all players won't handle it. As far as capture bitrate/resolution impacts quality, the more data you have, the more any filtering will have to work with. If you can capture to mpg2, the resulting files will be small, so there shouldn't be any real tradeoffs using highest bitrate/size settings.

Trivia: mpg2 (especially all I frame), can do a quite decent job as a capture format, roughly equal to DV at the same bit rates.

Inc
13th January 2005, 19:39
Originally posted by mic
[B]FWIW, personal opinions and all...

If the final target is mpg2, try and do mpg2 capture, all I frame preffered, at the highest size/bit rate you can manage.

In that case he can just use a software mjpeg codec like Picvideo at Q=19 which is almost the same as "I-Frame-only-mpeg2" capturing at high Bitrates AND the goal beside that is a resulting YUY2 4:2:2 capture, best if processing is wanted afterwards like using avisynth.
Youre right, in your recommendation no colorspace conversion will occur but when processing the video afterwards its better to have the full vertival chroma informations :)

Sergei_Esenin
13th January 2005, 23:08
In that case he can just use a software mjpeg codec like Picvideo at Q=19

Agreed. Picvideo @ quality 19 is almost always going to be better than or equal to any high bitrate MPEG-2 capture, offer better chroma preservation, and be easier to edit with frame-accuracy. Lossless codecs like huffyuv are obviously even better, but more marginally so.

As for Betamax capture--I'm lucky enough to have a Sony EDW-30F high-end ED Beta deck, and from experience the same filters recommended for VHS work well. For high-quality commercial Beta tapes lighter settings should be used though, because they have a bit less chroma noise and sometimes better resolution. There's a great thread in the Avisynth Usage forum right now about VHS filtering, that's applicable here.

ADLANCAS
16th January 2005, 02:31
@Arachnotron
- use an component to Y/C adapter and cap Y/C
Could you please post a link about such adapter ?

Thanks,

Alexandre

jggimi
16th January 2005, 16:24
search google with the following keywords:

component s-video adapter

Arachnotron
16th January 2005, 18:37
Could you please post a link about such adapter ? I'm in PAL country, so I don't have one, I only know they exist.

One thing to look out for though: In the most simple (cheap)form you combine the Pb and Pr wires to produce C. This is probably not a good thing to do since it messes up the impedance and can lead to quality loss.

You will need some extra electronics, maybe some extra resistors and capacities, maybe more. As jggimi said, google is your friend in this.

ADLANCAS
16th January 2005, 23:50
Thanks

mustardman
17th January 2005, 11:55
@skaven : Reading what you say, if you do indeed have "many hundred hours", then you may be better off purchasing a DVD recorder and going direct to that (unless you want to do some editing).

I am going through the same process, but putting all my VHS material to DV (files, not to DV tape, yet...)

I have about 200 hours, and doing it properly, it has taken me about a year to reach about 100 hours. No editing, just capturing, repairing and writing to DV files (archived on DVD discs).

Also, don't bother with Studio 8, it really is a pile of steaming, well, you know, the brown stuff. I use DVDlab for authoring (great once learned - took me about 4 hours) and MainConcept for encoding (can't fault it yet).


MM.