On Thu, Jul 23, 2009 at 1:39 AM, Bjoern
Rasmussen<bjoern_rasmussen at hotmail.com> wrote:
> I'm trying to stream a Theora encoded video over the internet using
UDP. Since Theora typically produce an encoded frame of 10 KBytes in my scenario
I need to split up this frame in smaller chunk which are then sent to clients.
This, however, makes my application very vulnerable to packet loss, so I'm
wondering if there is some way to make the Theora encoder produce many small
frames instead of one big frame? (They have to be less than 1400 bytes).
Note that there is an RTP payload format draft. You might look at that
for ideas.
?http://svn.xiph.org/trunk/theora/doc/draft-ietf-avt-rtp-theora-00.txt
Generally speaking though, theora doesn't have good support for packet
loss concealment. We have some ideas for how an encoder could do a
better job with this, but I'm not aware of any specific code.
FWIW,
?-r