> The reason I was doing this is the comment in the following line in > DeviceSource.cpp: > > gettimeofday(&fPresentationTime, NULL); // If you have a more accurate time - > e.g., from an encoder - then use that instead. > > With respect to that comment: what is the recommended way of doing this? Use > gettimeofday to get the timestamp for the frame, feed them into the encoder > and use the associated output timestamps of the respective NAL units for > sending them? Would that be the way it is intended to be used?
Yes, I think so (if I understand you correctly). Using presentation times generated by the encoder is generally better (more accurate) than just calling “gettimeofday()” in your “doGetNextFrame()” subclass. However, the presentation times must still be aligned with ‘wall clock’ time - i.e., the same as “gettimeofday()”. You can’t just start the presentation times at 0; that’s what was tripping you up before. Ross Finlayson Live Networks, Inc. http://www.live555.com/ _______________________________________________ live-devel mailing list live-devel@lists.live555.com http://lists.live555.com/mailman/listinfo/live-devel
