Re: [webrtc-pc] sendEncodings in "create an RTPSender" should reflect platform capabilities

#1862 certainly makes a difference, but it doesn't allow the user to change from unicast to simulcast after creating the sender, or to figure out whether the sender can support more layers than initially configured.

Tangential: There's a problem with the #1862-added text in the case where one wishes to start off with one layer of simulcast and activate new layers later - in that step 14 in "create an RTCRtpSender" removes the RID, and since RID is not modifiable via setParameters, we can't set it afterwards.
That means that if the sender wants to use simulcast, he MUST start off with the max # of layers he ever wants to support, and just set "active" = false on the layers he doesn't want in the beginning.


-- 
GitHub Notification of comment by alvestrand
Please view or discuss this issue at https://github.com/w3c/webrtc-pc/issues/1872#issuecomment-391108336 using your GitHub account

Received on Tuesday, 22 May 2018 19:14:24 UTC