[sdiy] They aren't sawtooths, they're ramps

[David G. Dixon]

> snip >
> > "Ramp" refers to the ramping part of the sawtooth wave, but could also
> > refer to waveforms where the ramp is an important part of a waveform
> > that's not really a sawtooth wave.  For example; a wave that starts
> > low, ramps up, holds there for a bit, and then resets.
> >
> > Relaxation oscillators are what they are, they don't define the
> > sawtooth wave.

I tend to call it a sawtooth when it ramps downward, and a ramp when it
ramps upward.  However, it seems to me that we should reserve the term
"sawtooth" for either upward or downward ramps which are linear; i.e., those
which are generated by a carefully designed VCO feeding a fixed current to
an integrator with one end of the cap pinned to virtual ground, and which
can be counted upon to contain both even and odd harmonics with amplitudes
inverse to their harmonic number.  At least, in synth circles, I think we
should make this distinction.

Waveforms derived from relaxation oscillators definitely do not fit the
bill, and therefore should not be called "sawtooth", any more than some
random "rounded" waveform should be called "sine".