[asterisk-video] How do you send large frames?

[Duane Storey]

Hi,

Video frames are typically split into smaller chunks via fragmentation of
the video data itself.  For example, H.263 has two packetization modes (RFC
2190 and RFC 2429).  In these RFCs, it describes how you can split one frame
into multiple UDP packets.  H.264 is a little nicer since it was written
with networks in mind -- each H.264 "chunk" is called a NAL (network
abstraction layer) unit.   Encoders can output small NALs directly and keep
the packet less than MTU.

Duane

On 5/16/06, Olle E Johansson <oej at edvina.net> wrote:
>
> (retransmitting this for the mail archives)
>
> Friends,
>
> I am still a student in the video area... One question I have is how
> you send large video frames over UDP/RTP?
> UDP fragmentation? Or do you use the multiple RTP frame function -
> where you use the same seq number?
> Or any other trick?
>
> I got an error report from a friend that we somehow cancelled or cut
> large frame updates.
>
> Also, the format drivers for video seems to make an illegal
> assumption about the size of two frames.
>
> /O
>
>
> _______________________________________________
> --Bandwidth and Colocation provided by Easynews.com --
>
> asterisk-video mailing list
> To UNSUBSCRIBE or update options visit:
>    http://lists.digium.com/mailman/listinfo/asterisk-video
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.digium.com/pipermail/asterisk-video/attachments/20060516/b9245239/attachment.htm