[sdiy] [synth-diy] numerically controlled superoscillator without hard sync

[Andrew Simper]

>>
>> And how about other behaviours that makes oscillators have tricky
>> frequencies... like PWM, where the oscillator is still but the
>> percieved pitch varies due to the modulation. How does the harmonic
>> aliasing behave there? Is all still well?
>
> An ideal oscillator with pwm doesn't have "tricky frequencies", it's
> still just a harmonic series.
>
> Hard sync could be difficult.

If you are running at a multiple of the DACs sample rate then true PWM
is not possible (neither is hard sync), since the "modulated"
transition (or sync position) doesn't have to be at a multiple of the
sample rate, it could be at any fraction of a sample. If the
transition is quantized to an integer samples then you will get a
different result to the continuous case, the modulation won't be
smooth but it should still sound ok for most uses, and perhaps even be
quirky enough to  have its own sound that people actually like the
artifacts of.

Andy