Hey Gary,
Cool. 10ms bodes well. I'm sure RTcmix has more overhead than STK due
to the scheduler. Another issue,
however, was that I was using code compiled with -g ... and that's it.
No optimizations whatsoever. I've
fleshed out the bugs preventing me from using -O and -m486. I'll bet
they get me some reduced latency. Time for another test after ICMC.
There's apparently some work afoot to improve this in 2.4 kernels.
Let's hope.
I also bet a higher end audio card like the Midiman or RME (we have one
of those) might be helping things out. Maybe not. ALSA might also have
some advantages. Probably more likely. I haven't gotten into the
ALSA API yet.
DT
--
Technical Director
Virginia Center for Computer Music
http://www.people.virginia.edu/~djt7p
(804) 924-7355
