well, i tried it with a phasor, but still got the same results.  the
really low frequencies (such as .001 Hz) tend to stretch the period
out.  i might try it with linseg instead of phasor since i can have
better control over the period.
i would love to know why this happens, and if the code could be fixed to
accomodate very low freqs.

:P

Richard Karpen wrote:
> 
> Pete,
> 
> Your right and probalby this should be documented somewhere in the
> csound manual. around 8-9 years ago I was using oscili for envlopes
> over very long stretches of time such as you're doing and it took
> me many long and frustrating days to figure out why I was not getting
> the results I expected to get. No time right now to explain why this
> happenes.
> 
> My quick advice: Use a phasor/table(i) pair instead of oscil(i) and you'll
> get exactly what you want.
> 
> Richard
> 
> On Sun, 25 Jul 1999, pete moss wrote:
> 
> > anyone know why oscil and phasor wont work with very low frequency
> > oscillation?  i tried generating signals with periods of 100 to 600
> > seconds, but at those low freqs, the period gets stretched.  anything
> > below about .1 Hz starts to go wrong.  is this problem caused by the
> > single vs. double issue raised earlier, or is it a different problem.
> > anyone know a different way to generate a very low freq signal with
> > correct period length?
> >
> > :P
> >
> > i am using consound 3.56 if it matters