Hi Rob, I hope you're going well ! Thanks for noticing that. Indeed the parameter is never read thus stored with its default value of 0. But this has no influence on the actual jitter buffer of the video stream, because this parameter is not used anywhere in the code. Instead, when creating the video stream, the same jittcomp value as for the audio stream is used.
The assumption made by linphone is that audio & video stream arrive synchronized. If you are observing that video arrives earlier, it can be either: - the sender that is sending audio "too late", because of sound card latency or whatever - or of course the receiver that plays the audio with latency, also for soundcard, echo canceller or whatever reason. Best regards, Simon Le lundi 24 janvier 2011 à 08:36 +0100, Maris a écrit : > I have been experimenting with video calls, and find that in general video > arrives earlier than audio on the far end. Therefore I considered setting > the parameter video_jitt_comp to a value unequal to 0, in order to have > both parts of the "stream" synchronized. However, when I set this > parameter, it is automatically forced to 0 again in .linphonerc after > closing the application (V 3.3.2, Linux). (?) > > -Rob > > _______________________________________________ > Linphone-users mailing list > [email protected] > http://lists.nongnu.org/mailman/listinfo/linphone-users _______________________________________________ Linphone-users mailing list [email protected] http://lists.nongnu.org/mailman/listinfo/linphone-users
