> > sorry for this newbie question, but i'm having serious trouble using > > the alsa lib 0.9.0beta8. After updating the code of my sequencer software > > from lib 0.5.8 api to 0.9.0 ,i simply can't start the queue.
> You need here snd_seq_drain_output(seq) to trigger the start event. Thanks very much! I solved the problem one hour ago: it really was the missing snd_seq_drain_output. I failed to replace snd_seq_flush_output by snd_seq_drain_output. Sorry. Now that 0.9.0 works, i have to test the rtc timer to get better midi timing. But there are still two remaining problems (both in 0.5.11 and 0.9.0): snd-card-mtpav: I use the Motu Midi Express XT 8 port interface. The only multiport interface, which is supported by alsa. If i play midi songs to more than one output port at the same time, i sometimes get hanging or dropped notes. When this happens, some or all output LEDs of totally unused output ports are flashing. Seems as if the data got routed to a wrong port - or was even broadcasted to all ports. The more midi traffic, the more often this error occurs. And if i add some more midi traffic, the sequencer queue slows down. I couldn't believe this when i heard this slowdown first, but i measured it with gettimeofday and snd_seq_get_queue_status and with a manual beat counter. For example, playing 5 identical techno songs (only note events) to 5 different interface ports results in a drop from 140bpm to 129bpm, depending on the current midi traffic load in the song. _Very_ strange! And, yes, i queue the events with their absolute tick times. Is it possible that a bug in the hw driver can slow down the queue? I mean, playing does not pause for short times, and then goes on at 140bpm. The speed slowly goes down from 140 to 129 as the songs get more complex. Any suggestions? Thanks! Stef(an). _______________________________________________ Alsa-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/alsa-devel
