Bill Unruh wrote: > Just to clarify, the crystal oscillators on the different cards to not run > at the same rate. Thus say one runs at 44100 and the other at 44050. That > means every second, the one card will bring in 50 fewer samples than the > other one. If you remove enough samples to flush the second buffer, the > first will have 50 left in it at the end of the first second, 100 the > second, etc. If you try to remove enough samples to flush the first, the > second buffer will not have enough samples in it to deliver to you. Is there no mechanism in ALSA for correcting drift, either between devices, or against a global clock? Video software has to correct for drift between the video and audio outputs, naturally. So such algorithms exist. It would be possible for a program to output to two ALSA devices, keeping the outputs somewhat synchronised by adjusting samples in one or both outputs. It might be useful, I'm not sure, to have an ALSA module do that, so programs can assume the sample rates of different devices are locked, or lock a rate to a global clock (e.g. the NTP synchronised system clock). The module might have better access to timing measurements for more accurate synchronisation. -- Jamie ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ Alsa-user mailing list Alsa-user@xxxxxxxxxxxxxxxxxxxxx https://lists.sourceforge.net/lists/listinfo/alsa-user
