DVD to Raw AVI

[.Kal-El]

I was wondering if there's a Post Production software out there for me to be able to Convert a DVD to raw or pure AVI format. Basically not compressing the file anymore. I wouldn't need to have the output to be a certain file size, I just didn't want to lose any quality once coverting it to AVI. Also if it could also keep the resolution and aspect ratio intact also.

[redwudz]

VirtualDub Mod. Just load the vobs in there and select 'Uncompressed AVI' for the output. You do realize the resulting file will be huge?

[.Kal-El]

Thanx redwudz for the fast reply. I was prepared for a big file. I needed it to edit the video if need be and can convert to H.264, .mov, MP4, etc...

[guns1inger]

You could also use a lossless compression, such as huffyuv or MSU, which should save you some space but still preserve the quality.

[redwudz]

What guns1inger said. I guess I should mention what I mean by 'huge'.

A 1GB VOB will inflate to about 37GB using 'Uncompressed RGB'. Figure a average commercial DVD is about 7GB average. 7 X 37GB would give you a filesize of about 259GB. I did say Huge.

Anyway, converting to uncompressed AVI is not really a viable option when you are talking about DVDs. I would recommend what guns1inger mentioned, using HuffyUV. It will give you a much more usable size to work with.

[.Kal-El]

How does this software covert the audio though? Or does it convert it at all?

[.Kal-El]

Tried VirtualDub Mod, it's okay for the average user. But, I need something more with kick, a software that won't lose the quality of the video, it doesn't have to be something cheap. At this point were willing to spend the extra $$$$$$.

[guns1inger]

VirtualdubMod doesn't really affect the video quality. It is the codecs you choose. If you use uncompressed, or huffyuv lossless, then you get the same quality as the original. If the quality is changing, it is becuase of something you are doing in your process.

[Cornucopia]

VdubMod doesn't "lose" the quality of the video, cheap or no. Just very direct. Another fast and direct (and free) way would be DVD2AVI (or it's new incarnation) and "Save AVI" -- Uncompressed.

Scott

[manono]

Tried VirtualDub Mod, it's okay for the average user.

You're joking, right? In this case free is best. As mentioned, install a lossless codec (I use Lagarith, less than 20 GB per hour of video), and encode away. You might be better off frameserving the Vobs with an AviSynth script file. If you consider yourself an advanced user, it might be more up your alley.

[celtic_druid]

Converting to raw in VDubMod would actually mean quality loss since you would end up with raw RGB24. What you want is YV12 or as suggested a lossless codec that supports YV12.

Raw YV12 will give you half the filesize of raw RGB24 anyway. If you have the helix YV12 codec installed then you can output YV12. Otherwise new versions of regular VirtualDub let you change the output colourspace.

[vhelp]

Hi everyone.. this is ment to confuse you all

I think the true method here, with respect to the direction of
original quality, is to use the method that takes the .VOB's
actual MPEG-2 and decodes the MPEG part, BUT KEEPS the YUV
components, or YV12 format, and preserve in that way.

My understanding of MPEG (still at this moment) is limited.
I am in a study/research to learn it, but at a very slow pace.
(I'm hoping to someday develop an MPEG encoder - nothing fancy)

Anyways..

After the video is converted to YUV color space, and then to YV12
sub-sampling (or format) this data is further processed through
the series machanism's of the MPEG compression phase, giving us
our final small size.. assuming everything went well.

This final phase is what gives MPEG its size, among other things.

( for instance, on commercial DVD VOB's, you have 1 gig size vob's,
and each vob file holds approx 20 minutes of MPEG audio/video )

Now, when we turn the knob in the other direction, (to decode) it
first has to decode the MPEG, and put send this decompressed video
to a YUV color space container, with the 'original' video data
untouched and in it's orignal YV12 sub-sampling format -- a
sub-division color space.

This YV12 video data is the final format that we want to maintain
and/or work with. This, IMHO is the closest to source's original
RAW video. Now we have the RAW video data.

The next step is to find a codec that will compress this YV12 video
to some degree, but in a lossless manor, to further reduce the size
of this YV12 video.

( because once we decompress the MPEG-2, we will have a very large file.
So we naturally need to compress the YV12 video, if we can. )

But, if we can't comrpess the YV12 video, then at least we should be
able to archive it in it's *native* YV12 video, untouched.

Now, to my knowledge with working with AVIsynth, I do not believe it
does this process above. AVIsynth depends on the codec to do the
necessary work to decode/decompress video data archived in a container
who's format is destined to be delt with by the external subroutine.

For intance, AVIsynth has built-in functions to open video files.. some
common commands as: AVIsource(fname); DirectShow(); mpeg2source("c:\source.d2v")
for instance.

(syntax/spelling of some keywords may not be accurate, but it does serve
to basically illustrate how avisynth handles video files)

But all these files get decoded "externally" by the calling codec.

I'm sorry. The point that I'm trying to make here, is that AVIsynth
is not decoding the source files directly. And, as such, you shouldn't
assume that it does. So, if we want to take out YV12 video data and
work with it, its up to the codec what we choose, to decode our YV12
video, and do any further process to it.

But, we could use AVIsynth to work with this same YV12 video in its
native YV12 color space, and gain some points while at it.

Ok. so we have a our original VOB file, decoded to YV12 video, and
we have it archived in a YV12 codec container.

At this point, we can choose to save it, HUGE'ly so. Or, we could
fine-tune it with further filtering (via AVIsynth) or whatever, or
re-compress it to another video format.. ie, divX; XviD; MPEG again,
etc.

Ok. My other point was about my belief that there is no decoder available
to take an MPEG-2 and decode it, and *KEEP* it in it's native YV12 format.
I honestly don't know or never heard of any such machnism or method
for this. My only assume here, is that whatever codec that are available
to decode the MPEG-2, does so, and transfer the video contents through
a RGB conversion. Course, I could be stood, corrected

Then, there is another direction to work with..

The MPEG encoder of choice, when the time comes to re-encode, assuming
that the decision is made to go back to MPEG.. rather than AVI.

If you go through the process again, from VOB to YV12, and you decide
to re-encode to MPEG again, your main concirn now, is to choose the
proper MPEG encoder to re-encode without further <conversions> of
the YV12 source.

For instance. Say you want to use CCE encoder. This enocder, to my
knowledge, requires the source to be in YUY2 color space. Ok. So
what's the deal ?? The deal here, is that now, you have to either
perform the color space conversion yourself (through other means, such
as AVIsynth frameserve method, and do color space conversion there,
ie, YV12->YUY2, and then frameserve to CCE) or other conversion means.
(I can't think of everything, you know)

Or, and 2nd instance.. Say you want to use the TMPGenc encoder. This
encoder works in RGB color space. Thus, you would have to do to a
color space conversion.. YV12->RGB and proceed to encode w/ TMPGend,
or, (through other means, such as AVIsynth frameserve method, and do color
space conversion there, ie, YV12->RGB, and then frameserve to TMPGenc) or,
you could let TMPGenc do the conversion built-in, (though it will use
the codec for the YV12 conversion, and it will convert to RGB, YV12->RGB
as it encodes the source)

And, last.. there is even one more instance.. Say your want to use an
encoder that accepts YV12 color space. HCenc is an example. I have
not used this encoder, but I have heard that it works with video sources
in YV12 color space. And this is a new (or latest) MPEG encoder that
is currently floating around. HCenc (or HC) might be your next-to-best
alternative, if your final archival plans are for back to MPEG again.

Ok, I'll stop here

-vhelp 3786

[manono]

celtic_druid suggested helix YV12. I suggested Lagarith. Both are capable of keeping it in YV12. I also suggested frameserving using AviSynth. It changes nothing unless you tell it to. I usually enjoy reading your posts vhelp, but that one made no sense at all.

[.Kal-El]

Thanx for the assistance everyone. I'll try these steps and see what kind of results I would get from it.

[vhelp]

>> I usually enjoy reading your posts vhelp, but that one made no sense at all.

Ok. Dissregard what I wrote earlier then, since I could not explain
my views properly regarding YV12 and other items. My applogies.

What I was trying to say was, that if AVIsynth was going to be the
tool to do further work with, then we would like to have:

** MPEG-2[decode] -> RAW[ YUV[YV12] ] -> AVIsynth[ RAW[YV12] ]

Anyway. I was going to continue on, but I cut it short here, cause I
feel that it would sway from the subject. Thanks for taking the time
to read my response, fwiwnw

-vhelp 3789

[manono]

I'm glad you answered, as I was feeling guilty for being harsh. I actually read it twice. After the first time I went "huh?" to myself, but then felt that perhaps I didn't get it, and that maybe there were nuggets of wisdom hidden in the incoherent rambling. So I read it again and came away with the feeling that I had wasted 10 minutes of my life. But obviously some thought and quite a bit of time went into the writing, so I apologize for the earlier remarks. I'm no AviSynth expert, and certainly no colorspace expert. Although old, maybe the Doom9 AviSynth YV12 FAQ will help:

http://forum.doom9.org/showthread.php?t=37276

[vhelp]

Ok. Regarding your last comment..

Don't sweat it. I must had used enough faul language at your last
remark throughout the day, enough already. No kidding around, I do
appreciate your comments afterwards - thank you

Not to spend much time on this.. I did have a few reference bookmarks
regarding a few discussions on YUV and color space, etc. I have read them
throughout, though they left me with more un-answered questions. Thus, the
majority of what I've learned regarding YUV and its color space branches
has been from my own attempts through self excercises, thanks mainly to my
long-time gifts in programming, and my thriving energy to just do it.

Thank you for the link. I recall reading it, but did not have it in
my bookmarks for reading. Hopefuly, it will have that answer that I am
searching for -- "does mpeg2dec.dll hand a RGB or YV12 image to calling program"

If I went off T here, my appoligies to all

-vhelp 3790

[celtic_druid]

mpeg2dec was originally YUY2, but then support for YV12 was added for AVISynth 2.x and became the default. I would suggest using dgdecode though, which also uses YV12 as default. Used to be I420 from recollection, but it was changed to YV12 because of some issues with mplayer/mencoder's avs support.

I420 is just YV12 with the chroma planes swapped.

[manono]

Hi-

does mpeg2dec.dll hand a RGB or YV12 image to calling program

That's an easy one to answer. First, it's DGDecode.dll now. You should upgrade from DVD2AVI/MPEG2Dec.dll or whatever you're using, to a current version of DGIndex/DGDecode.dll. And then you might read the included DGDecodeManual, where, among other things, you'll find:

Additional features include: YV12, I420, and YUY2 colorspace output (and RGB24 via DGVfapi),

And this:

upConv: 0 to 2 (default: 0)

Upsample from 4:2:0 to YUY2 (4:2:2) or RGB24.
- 0: Do not upsample
- 1: Upsample to YUY2 (ignored if input is already 4:2:2)
- 2: Upsample to RGB24

And this:

To do plain YV12 decoding:

MPEG2Source("[PATH\]project.d2v")

You'll find more info in the DGMPGDec FAQ here:

http://forum.doom9.org/showthread.php?t=87809

[Gi-Ma-Mi]

how about a DVD (DVD-9?) with multiple episodes for a series (like those at www.yesasia.com)
what if you want to back up those episode seperately into difference avi files, like if the dvd contains 4 eps, and if they divided each eps into small chapters (ask this additional question just in case they really did that :P )i want to get 4 seperated avi file in result
how you do that?
i saw someone ripped their DVDs like that, into seperated files, and they usually use 700mb file as their standard, can i make it bigger than that? and if i do, how many mb should i make it?

[.Kal-El]

Is there maybe an all in one software that my company could purchase? At this point i'm even considering to just convert the VOB files to MPEG 2. By the way thanx for everyone's input again. I really appreciate it.

[celtic_druid]

The VOB's would already be MPEG2. You can just demux.

The only all in one app that comes to mind is mencoder. It can rip from a DVD directly to an avi containing raw video in numerous different colourspaces. Can also remux the audio or convert to PCM.

[whitejremiah]

This is particularly geared at Gi-Ma-Mi......another option would be to go like smartripper>vob>DGIndex>VFapi>whatever output format you want.... Also make sure in DGIndex to change to force film, if your source requires it.......

[celtic_druid]

I'd suggest AVISynth instead of VFAPI.