[MPlayer-dev-eng] Re: [PATCH] theora support

[David Kuehling]

Jonathan Rogers wrote:
> That's great! I'll try it immediately. I've been waiting for Theora
> support to be added to a better player, as the example one seemed to
> take too much CPU time to be practical on my machine (for decently
> sized movies). Ever since I heard of Theora, I've been excited to see
> how it'll compare to MPEG4 video.

I had the same problems with the example player.  But the current
implementation of my theora driver might not yet be 100% efficient: it
copies the YUV-Buffer from Theora into mplayer's MPI structure.  I'm not
sure, but it might be possible to make the MPI image just point to
Theora's YUV data.

>From my tests it seems to compare well to MPEG4 (I was suprised, I read
lots of negative comments on the net).  There are bad artifacts in
"extreme" scenes like sharp red text on black background, but on the
other hand high-motion and high-complexity scenes encode almost without
noticable artifacts at quite low bitrates (at least when compared with
typical low-bitrate MPEG4).

> By the way, did you notice that the video in the example oggs in
> <URL:http://www.theora.org/test/test.zip> doesn't play correctly with
> the current example player?

No, I used encoder_example in combination with mplayer to create various
samples.  I don't know any location to get up-to-date samples from... 

David Kühling
-- 
GnuPG public key: http://user.cs.tu-berlin.de/~dvdkhlng/dk.gpg
Fingerprint: B17A DC95 D293 657B 4205  D016 7DEF 5323 C174 7D40