Paul Barton-Davis wrote:
> Adding SCHED_FIFO priority and MLOCKALL to lock all current and future
> pages in physical memory makes no visible difference to the program,
> except that it locks the machine more solidly when we get up close to
> breaking point (65-70 oscillators).
> 
> I'll have to try the ALSA driver and see if it can do better.
> 
> --p

Interesting. In my experience (quite a while ago now), it made the
difference betweeen having occasional dropouts due to other processes
running, and (in the SCHED_FIFO case) having dropouts only due to the
kflush deamon. Average throughput probably didn't change that much, but
the key issue was in preventing other processes from kicking in during
the sound-driver calls and not returning in time...


Or perhaps you're talking about the benchmark program results (which
possibly measure only throughput?)
 and not whether you can run Csound without dropouts?

Larry