[vorbis-dev] RFC draft for Vorbis over RTP

[Jack Moffitt]

Why don't we just say that for transmission of RTP, a single set of
codecs valid for the entire transmission is required?  Here are the
different manners of broadcasting:

1) live.  the codebooks will never change.
2) simulated live. the files streamed will probably all have the same
characteristics.  if not, they could be pre-reencoded to have the same
characteristics, if saving cpu is the concern.  A broadcast app can just
rewrite the granulepos as appropriate.

The only reason to support contantly changing codebooks is essentially
to support playlist style simulated live broadcasting with low cpu
overhead.  Since I see no reason this can't be done without changing the
codebooks, why don't we nail the set of codebooks down at the beginning
and not allow them to change during the life of the stream?

The more I think about this, the format genericity that codebooks buy
isn't really lost with this, since codebooks are still required.  But
the parts of codebooks that don't make sense in streaming are completely
gotten rid of.  And this gets us back to discussing how to send the
original set of codebooks, which SDP/HTTP should be able to handle
without problems.

(Btw, I fully expect someone to respond quoting my own forgotten
arguments against exactly this.)

jack.
--- >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.