> > To get really accurate timing in all cases, this thread should even have > > higher priority than the one created by jackd (*), otherwise all events > > that occur while a jack_process() is running will be delayed until the > > end of all processing for the current cycle. > This won't help, because the process routine will only be called once by jack during one jack period. If its the first in the graph then your routine may have been processed before an event has actually arrived.
The only reasonable ways to achive lower latencies is to sync jack to an interrupt source with higher interrupt frequency than the soundcard or to change jack so it subdivides its period into smaller periods and busy wait to equally spread the subperiods over the jack period. Ralf
