[vorbis] Video codec

[Sean Wieland]

Ralph Giles wrote:
> 
> On Fri, 8 Sep 2000, Sean Wieland wrote:
> 
> > Well, I was thinking more along the lines of reasonable storage
> > considerations.  At least 5GB per DVD is a bit harsh.  Whereas .5GB per
> > OGG-Video means plenty of room on one of those new 80GB hard drives for
> > a good-sized movie collection in addition to a good-sized CD collection,
> > all of the regular apps, games, etc.
> 
> Alright, I see what you mean. I just wanted to be clear that DVD-video is
> already lossily compressed; what you were asking for was not "Oh look, I
> can fit my CD collection in 1/10th the space using Vorbis" but "Oh look, I
> can fit my mp3 collection in 1/10th the space using Vorbis *and*
> downsampling to 11kHz mono".

Well sorta.  I understand very well that DVDs merely use MPEG-2 video
and AC3 audio, which are then multiplexed into Video OBjects (VOBs); and
that both MPEG-2 and AC3 are lossy compression methods.  But what I'm
describing is the same perceptual quality, but at 1/10th the storage
space.  So "downsampling to 11kHz mono" would be unaccpetable for that
purpose.  But using Vorbis at 128kbps to encode the 6 channels of audio
instead of 386kbps (or whatever crazy high bitrate AC3 on DVDs use) AC3.

> > I'm talking about 352x240 resolution VHS-quality streaming video at
> > about 100kbps.
> > I agree that DVD level resolution and quality without broadband would be
> > extremely difficult (as in near impossible).  Although DVD resolution
> > and quality at about 1Mbps for archive on CD would be nice.
> 
> I still think this is overly optimistic. None of the current codecs come
> close to VHS quality at 100kbps. But we'll see. ;-)

Exactly why the Ogg Multimedia Project should write one!  This tarkin /
DWT method sounds really promising, and if some source code and
documentation were provided I just might be able to get a few friends to
help develop it code-wise.  Besides, it'd be nice to say Ogg Multimedia
did something better than anyone else (and possibly bring some Free as
in Speech and as in Beer standardization to the Real/ASF/QuickTime
proprietary wars), as opposed to trying to fight the MP3 momentum with
"look, Vorbis is just as good now and technically quite a bit better". 
(Which is not to say that Vorbis isn't a greatly needed technology, just
that it's a harder political sell.)

-Sean


-- 
 +-----=[export-a-crypto-system-sig RSA-3-lines-PERL]=-----+
 |#!/bin/perl -sp0777i<X+d*lMLa^*lN%0]dsXx++lMlN/dsM0<j]dsj|
 |$/=unpack('H*',$_);$_=`echo 16dio\U$k"SK$/SM$n\EsN0p[lN*1|
 |lK[d2%Sa2/d0$^Ixp"|dc`;s/\W//g;$_=pack('H*',/((..)*)$/)  |
 +--------=[http://www.cypherspace.org/~adam/rsa/]=--------+
"Most people would rather die than think, and most people do." 
                                       -- Bertrand Russell

--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to 'vorbis-request at xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is needed.
Unsubscribe messages sent to the list will be ignored/filtered.