@shensley wrote:
Blockquote
The Oscillator class still uses the math.hsinfunction instead of some fast-approximation which could be pretty limiting depending on how-many you’re using and what waveform. You could sanity check that by switching them all to the naive square waveform and seeing if you’re still close. And you could run the LFOs at once per callback instead of once per sample, but then you’ll still have to filter or interpolate it to get smooth changes on most parameters.
Wouldn’t it make more sense to do SIN/COS oscillators “DSP-style” like this:
I suspect that these might need to be implemented with integer math in order for their amplitude to be stable over time, and then converted to a float return value for use. Run once per sample as an audio-rate oscillator or once per callback as an LFO. This should be super-fast on this processor, right?