[vorbis-dev] rtp implementation
kmarks at apple.com
Mon Apr 9 10:58:23 PDT 2001
At 2:56 pm -0700 8/4/01, Jack Moffitt wrote:
> > I would like to implement the rtp specification.
>> Is anyone else interested in helping with the design/code?
>I would be glad to help, since I thought I would have to do this myself
>I think probably the best way to do it is just to start with a
>packetizer. The output is just a collection of RTP packets,
>appropriately stamped, etc.
>Icecast does not currently have an RTP streaming layer, and that will
>have to be implemented at some point as well. As long as we keep the
>two independent, we should end up with a flexible architecture.
>> Jack: i'm interested in your opinion on the difficulty of this since you
>> worked on the rfc spec.
>Before you begin, there are a few warnings and suggestions:
>- this is a draft. it will change. whatever code you write will need
>modification as the draft changes. keep this in mind :)
>- i have several edits to the draft that are minor trivial things, but I
>would like to put them in before you start. i just want to incorporate
>the feedback i got and flesh out a few sections.
>- i would suggest that you only consider unicast rtp streams for
>starters. the multicast method of reliably transmitting the codebooks
>has still not been determined. and the ietf seems to think that
>reliable multicast might be the best solution, which afaict doesn't
I suggested a couple of approaches on the RTP-list, but you didn't respond:
Use a byte in each RTP packet to define the codebook.
Pre-define several codebooks (like JPEG packing does) and use
byte<128 to be one of the pre-defined ones and byte>128 to be a
dynamic one. As a lot of vorbis files use the default codebooks, this
is a big simplification.
Send non-default codebook(s) in the sdp description (Which is sent reliably).
Use a scheme like QT packing uses for SampleDescriptions;
--- >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 'vorbis-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.
More information about the Vorbis-dev