In an analog synth the square wave has some distortion to it: the rise is not instantaneous and there is some overshoot of the peak too. Over the years this was minimized by using high slew rate amp circuits and the like so an old Moog pulse wave is less ideal in shape than a new one.
For a digital square wave you would want to use band limiting to avoid alias artifacts which are quite severe for a square or sawtooth wave. As the frequency goes higher both of those waves will begin to look more like sine waves as the partials are filtered out. Pretty much every commercial digital synth and plugin uses anti-aliased waveforms.
out of interest Chris, do they use wavetables (possible of different lengths for different octaves) or internal oversampling a la Millers example in the doc patches?
best, P
On Mon, Feb 6, 2012 at 8:31 AM, Pierre Massat pimassat@gmail.com wrote:
Dear List,
I need a square wave to use as an LFO in a ring mod patch. Though i could simply use [expr~ $v1>0.5], I m wondering if I could get a smoother square wave. I tried to use a bandlimited square wave, but I don't have enough harmonics and the top of the wave isn't flat enough. Incidentally, what do analog square waves from old synths look like? Anyway, I think i want a square wave with no jumps between 0 and 1.
Cheers!
Pierre.
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list