[vorbis] Icecast2 in xiph CVS, PATCH!

[Monty]

On Fri, Sep 28, 2001 at 12:16:42AM +0200, Gian-Carlo Pascutto wrote:
> As I understand it, the problem is that the peeler would have
> to know about the psychoacoustics to know what to peel. So
> the encoder must structure the ogg in a fit way so that the
> peeler can deduce what to remove or not.

No. The problem is that the rc2 encoder is not good about putting
vital audio information first.  The rc3 modes most certainly will be.

> Do I got this right? Is it possible to make a 'smart' peeler
> that can handle oggs that aren't structured in that way?

The encoder can do things to make peeling much easier; what rc2 is
missing is doing the iterative cascading passes in an intelligent
order.  One *could* write a peeler to restructure the cascading passes
(without reencoding) to do a better job, but it would be rc2 specific.

The unintelligent structuring of the rc2 cascading was, frankly, a
fuckup on my part.  I won't make the same mistake in rc3.

Monty

--- >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.