At 09:09 AM 9/19/2003 -0400, Bert Culpepper wrote:
Hi Alex,

Right, but the behavior your after is not guaranteed by the relevant RFCs (as has been pointed out already). An extra offer-answer exchange is required in the case the answer lists multiple codecs.

Bert is right - if the far end answers with multiple codecs, you can then do a re-INVITE offering just one codec (the one that the far end actually started sending with in the RTP) which will then lock you down to a single codec.


See an example of this in:

http://www.ietf.org/internet-drafts/draft-ietf-mmusic-offer-answer-examples-01.txt

Thanks,
Alan Johnston
MCI

I would hope to see only a single codec in an answer, and see the mechanisms defined in RFC 3388 used in cases where switching codecs is needed during the call, given the practical aspects of building platforms that handle multiple calls at a time.

Regards,
Bert

Alex Zeffertt wrote:

In this case you should not advertise that you can. Use separate "m" lines, each with one codec. This will remove the ambiguity you're having, and be compliant with the relative RFCs.

Hi Bert,
This sounds like a good idea, and it would mean that the answerer
wouldn't dynamically change codecs used on each RTP port - but I think
it give the impression that the offerer can support multiple
_simultaneous_ different codec streams, all on seperate RTP ports.  What
I actually want is the offerer to advertise all the codecs it supports
in such a way that the answerer responds with just one.
Alex
_______________________________________________
Sip-implementors mailing list
[EMAIL PROTECTED]
http://lists.cs.columbia.edu/mailman/listinfo/sip-implementors


_______________________________________________
Sip-implementors mailing list
[EMAIL PROTECTED]
http://lists.cs.columbia.edu/mailman/listinfo/sip-implementors

_______________________________________________ Sip-implementors mailing list [EMAIL PROTECTED] http://lists.cs.columbia.edu/mailman/listinfo/sip-implementors

Reply via email to