Kind of agree, browsers implementators would probably know much better than the app dev how to provide the best quality and adjust layers accordingly, and that the dependency id mechanism is over complicated. However I think it would worthy to be able to give some hints about what you would like to receive, like "send me at fps, fps/2 low res and 1fps high res". Best regards Sergio El 17/9/2017 0:15, "Bernard Aboba" <Bernard.Aboba@microsoft.com> escribió: > Iñaki said: > > "Agreed with this." > > [BA] The dependencyEncodingIds approach potentially provides more > flexibility in the case of spatial scalability, since you could potentially > specify the maximum temporal and spatial layers that can be encoded. > However, AFAIK Ortc Lib doesn't support spatial scalability yet and if > we're only talking about controlling temporal scalability I'm not sure how > much is added by specifying the maximum number of temporal layers > versus saying "do temporal encoding up to the maximum number of layers > supported by the encoder". >
Received on Saturday, 16 September 2017 22:54:09 UTC