Let us say A makes an SDP offer which looks like

m=audio 49120 RTP <http://www.voip-telephony.org/rfc/avt> /AVP 97
a=rtpmap:97 EVRC/8000
a=fmtp:97 maxinterleave=2
m=video 49170/2 RTP/AVP 98
a=rtpmap:98 MP4V-ES/90000
a=fmtp:98 profile-level-id=1;config=XYZ

My understanding is that these fmtp parameters are specific to the UA
and not to the session; however I have a colleague who claims that these
are properties of the session i.e. for a sendrecv channel both UAs must
use the same fmtp values so for e.g. B which receives this offer must
also use maxinterleave of 2, etc. But, I think maxinterleave is a
property of the UA and need not be the same for both A and B. Same thing
with the MP4 vol header which is an encoder property and hence need not
be the same for both end points in a one-to-one session. So I believe
this is a valid SDP answer to the above offer

m=audio 49120 RTP <http://www.voip-telephony.org/rfc/avt> /AVP 97
a=rtpmap:97 EVRC/8000
a=fmtp:97 maxinterleave=5
m=video 49170/2 RTP/AVP 98
a=rtpmap:98 MP4V-ES/90000
a=fmtp:98 profile-level-id=1;config=ABC

Per my colleague, the above is not a valid SDP answer to the offer. The
RFCs seem very vague about the interpretation of these fmtp parameters.
Is this not a source of interop nightmares where vendor implementations
depend on their interpretation of these ambiguous specs or am I missing
something fundamental here? Your input is most appreciated.

Thanks,
-Ravi.



_______________________________________________
Sip-implementors mailing list
[email protected]
https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors

Reply via email to