[Asterisk-video] Re: [asterisk-dev] Video packetization proposal

[Mihai Balea]

On Jun 2, 2007, at 10:35 PM, Jeffrey C. Ollie wrote:

> On Fri, 2007-06-01 at 18:32 -0400, Mihai Balea wrote:
>>
>> Theora does not support packetization
>
> Not true:
>
> http://svn.xiph.org/trunk/theora/doc/draft-ietf-avt-rtp-theora-00.txt

Well, actually it is true.  The codec itself does not have any built  
in support for arbitrary size chunks.  When you use Theora, all you  
can get is complete encoded frames that usually are too big to be  
sent in one network frame.  The xiph draft arbitrarily slices up the  
full frames into usable, network sized chunks. The downside is that  
if one or more such chunks are missing or are miss-ordered, nasty  
things happen leading to the entire frame being useless.  We use a  
similar technique in iaxclient and we decided that it is better, from  
a user experience point of view, to drop an entire frame rather than  
display an incomplete one.

Compare to H.264 which does have built in packetization and is  
resistant to lost or miss-ordered packets.

>
>>  and it is, unfortunately, the
>> only free (as in beer and speech) video codec.
>
> Untrue as well:
>
> http://dirac.sourceforge.net/
> http://schrodinger.sourceforge.net/

Last time I checked, about 6 months ago, Dirac wasn't ready for  
realtime encoding/decoding use.  The algorithms appear promising, but  
they really need to optimize their code.

Mihai