[sdiy] Sequencer interrupt latency?

[Byron G. Jacquot]

> I'm finishing up a PIC based sequencer and now I'm optimizing for
> performance.  My step input feeds the interrupt input.  When a step
> interrupt is serviced, the analog out is set, then the step trigger is
> asserted, and finally the right LED is lit up.  My question is:
> 
> From your experiences, what is the maximum acceptable latency from
> step input to analog out and step input to step trigger?
> 
> By "maximum acceptable latency", I mean that maximum time that it
> becomes noticeable under non-audio rate applications.  Right now, I'm
> getting about 50 usec latency to the step trigger which means I get
> maybe 40 usec to analog out.

The trigger latency usually becomes noticable somewhere around 10 
milliseconds...and that number varies between people, and also 
depending on the situation.  If you're used to acoustic instruments 
with near-instantaneous event-to-sound translation, then you might be 
more attuned to noticing when an electronic instrument takes a little 
while to start playing a note.

But even with acoustic instruments, you're dealing with some acoustic 
delay between the source and your head, of roughly 1 millisecond per 
foot.  I never felt that the acoustic delay from standing 25 feet 
away from a guitar amp was much of a problem...

If you start getting into audio frequencies, then things become a lot 
less tolerant, because you're looking at limiting your Nyquist 
frequency, and comb filtering from even a slight delay is pretty 
audible.

Byron Jacquot