Re: Simulcast V1

On 15/08/15 18:56, Bernard Aboba wrote:
> On Aug 14, 2015, at 22:51, Stefan Håkansson LK
> <stefan.lk.hakansson@ericsson.com> wrote:
>>
>> Trying to get my head around it: would it be fair to say that the
>> main difference between Adam's proposal and track cloning relates
>> to the SDP produced by the browser? On the media plane it's the
>> same (given Bundle is used): the different simulcast streams travel
>> on the same ports with the same encodings etc.
>
> [BA] The media would have similar transport characteristics and could
> be signaled on the wire in a similar way, but there is still the
> question of how to control characteristics of the simulcast streams
> being sent. In track cloning, constraints are used for this, in which
> characteristics of each track are individually specified.  In ORTC
> RtpEncodingParameters are used, which allows the relationship between
> the streams to be explicitly described (e.g. Low-res is 1/4 the
> resolution of hi-res).  Implementations of these control mechanisms
> could produce the same result with similar performance, but they
> might not.

I understand, but the proposal from Adam had nothing like that IIUC, it 
was just a simulcastCount on the RtpSender which in turn would lead to a 
SDP O/A cycle where the remote end would decide the relationships via 
the SDP answer. Quite different to the ORTC API where the sender (JS) 
controls.