lchiu7
23rd November 2008, 03:19
I capture DVB-T streams using a Hauppauge Nova-T 500 (these are usually H.264/AAC or H.264/AC-3). I have an external UHF antenna.
When viewing these streams using PowerDVD8 I can occasionally see video glitches (dropouts, momentary freezes) but the AV stays in sync. The sync remains even if I edit the streams using TS Packed Editor. However if I demux the video and audio and try to mux a different audio streams back, at the point of the glitches, the AV sync goes out.
The glitches play havoc occasionally with programs that process the video stream (dgavcindex will generate errors or stop prematurely, H264cutter will abort) but I guess they were never written to handle these kinds of error situations.
I am assuming (this is not an area of expertise of mine) that the original streams contain time code information in the packets that allow the playback devices/software to maintain sync. But once the audio is say removed and converted and muxed back (I am using txmuxer) those timecodes no longer exist and so should there be drops in the signal, there are no codes for the software to lock back onto?
If so then at least I know what I am dealing with.
Thanks
When viewing these streams using PowerDVD8 I can occasionally see video glitches (dropouts, momentary freezes) but the AV stays in sync. The sync remains even if I edit the streams using TS Packed Editor. However if I demux the video and audio and try to mux a different audio streams back, at the point of the glitches, the AV sync goes out.
The glitches play havoc occasionally with programs that process the video stream (dgavcindex will generate errors or stop prematurely, H264cutter will abort) but I guess they were never written to handle these kinds of error situations.
I am assuming (this is not an area of expertise of mine) that the original streams contain time code information in the packets that allow the playback devices/software to maintain sync. But once the audio is say removed and converted and muxed back (I am using txmuxer) those timecodes no longer exist and so should there be drops in the signal, there are no codes for the software to lock back onto?
If so then at least I know what I am dealing with.
Thanks