Hi,
Can you please explain how RTCP feedback forwarding works?
http://lynckia.com/licode/rtcp-feedback.html
Do you follow a specific spec?
If I have one peer publishing audio and 10 peers receiving this audio
stream from the server, I’ll have one SR packet and 10 RR packets.
Does the server collect RR packets from all receivers and send an average
of results or just forward them as is?
Does the server modify the RR packet header (SSRC of sender)?
Thanks
Hi Pablo,
At this point we forward all the RR packets, we might include some control mechanism if we find the performance degrades with many users. Some of our recent commits are dedicated to improve our RTCP/RTP support and it would be quite easy to implement something like this.
We do, however modify the SSRCs to match the ones established in the SDP negotiation.
Cheers!–
Pedro Rodriguez
On 27 Mar 2014 at 16:46:10, pablo (pablo.platt@gmail.com) wrote:
Hi,
Can you please explain how RTCP feedback forwarding works?
http://lynckia.com/licode/rtcp-feedback.html
Do you follow a specific spec?
If I have one peer publishing audio and 10 peers receiving this audio stream from the server, I’ll have one SR packet and 10 RR packets.
Does the server collect RR packets from all receivers and send an average of results or just forward them as is?
Does the server modify the RR packet header (SSRC of sender)?
You received this message because you are subscribed to the Google Groups “lynckia” group.
To unsubscribe from this group and stop receiving emails from it, send an email to lynckia+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.