[theora-dev] MMX and extended-MMX acceleration patch for encoding

[Rodolphe Ortalo]

On Fri, 9 May 2003, Dan Miller wrote:

> just as a point of reference, the original VP3 package was able to
> encode 320x240, 30 fps material in real-time on a 1 ghz processor.
> 

Do you think there were any regression from this package to the C code
currently in Xiph's CVS?

Apparently the pure C implementation of VP3 is already pretty well
optimized, at least there is no single performance point. (That's why the
MMX benefits are not very significant unless done in several places). At
some points, IIUC, the C code was prepared for MMX-like additions too, and
that's also why I gave it a try, based on these hints.

Just by curiosity, does later versions of VP (like VP5 and VP6) bring much
improvement wrt VP3 performance?

Rodolphe

<p>--- >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 'theora-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.