[vorbis-dev] lossless ogg encoding with good compression rate

[Beni Cherniavsky]

Carsten Haese wrote on 2003-06-26:

> Yes, the specs specify the correct way to decode, but I have yet to see
> a computer that performs infinite precision math.

:-)

> The difference comes down to different rounding strategies between
> different platforms. Since Vorbis is lossy, this is not a problem in
> practice, because those errors are drowned out by the quantization
> error that the encoder produces. It only becomes a problem when you
> try to make a lossless process from an inherently lossy process.
>
This is a slight problem also when you start to discuss lossless
transformations of Vorbis files.  E.g. the discussed peelabilizer, may
it come soon: how do you define it's being lossless except by
comparing the decoded output bit-by-bit on a reference decoder?  OK,
that is not perfect either, a truly lossless Vorbis transform should
produce an infinitely-equal mathematically ideal decoding.  It is
probably not even important that an imprecise might decoder might
produce a differnt approximation to this same ideal decoding.  All I
want is a decoder of which I can be sure that it always matches the
ideal decoding in the top N bits.  What is N for libvorbis (on a given
machine/compiler)?


-- 
Beni Cherniavsky <cben at tx.technion.ac.il>
--- >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-dev-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.