hi,
we're currently writing a piece of software that does some statistical analysis and interpoliation, and then triggers pd via a tcp socket, the problem is that both sleep() and nanosleep() syscalls give us an approx. jitter of 30msec which is way too much, and we're somehow stuck ...
any hints anyone ?
thanks in advance,
x
Hi Chris,
select() may work better (or else it might not... depends on OS and other things.) Also, if the problem is just to reduce time jitter in the messages Pd receives, you can time-stamp the messages (gettimeofday(), usually accurate to a millisecond or better) and have the Pd patch de-jitter them using the "pipe" object...
cheers Miller
On Thu, Aug 09, 2001 at 08:28:54PM +0200, CK wrote:
hi,
we're currently writing a piece of software that does some statistical analysis and interpoliation, and then triggers pd via a tcp socket, the problem is that both sleep() and nanosleep() syscalls give us an approx. jitter of 30msec which is way too much, and we're somehow stuck ...
any hints anyone ?
thanks in advance,
x
-- chris@lo-res.org Postmodernism is german romanticism with better http://pilot.fm/ special effects. (Jeff Keuss / via ctheory.com)