You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Once frame marking is implemented (#298) there are two pending subjects to improve/fix in H264::PayloadDescriptorHandler::Process():
We should check incremental "picture id" somehow and let the packet pass if its picture id is older than highest seen. However no idea how to check it in H264 frame-marking.
NOTE: Better don't do it. We cannot check tl0picidx since many packets with different tid would have the same value.
Eventually (and this happens in all simulcast streams) there are no H264 packets with the framingMarking->base (Base Layer Sync) set, so we cannot wait for it in order to let higher temporal layers pass.
- VP8: Check that the payload has pictureId and tl0PictureIndex before using them.
- H264: Check tl0picidx is greater than highest seen (related to #306).
* VP8: Check that the payload has pictureId and tl0PictureIndex before using them.
* PayloadDescriptorHandler: rename Encode() to Process().
* Codecs&Simulcast: move current and target temporal layer stuff to the codec handler.
* Upgrade Catch to 2.8.0.
Confirmed that this is fixed in Chrome 78.0.3904.97 (OSX), so eventually we should uncomment this block. However, we must be sure that it would not affect older libwebrtc based devices.
Once frame marking is implemented (#298) there are two pending subjects to improve/fix in
H264::PayloadDescriptorHandler::Process()
:tl0picidx
since many packets with differenttid
would have the same value.framingMarking->base
(Base Layer Sync) set, so we cannot wait for it in order to let higher temporal layers pass.The text was updated successfully, but these errors were encountered: