I was just reading through the Audio Library Development Roadmap, and thought I’d offer my two cents:
Firstly, I agree that frequency modulation of all waveforms would be a great improvement — the sine_fm object is terrific, but could probably be replaced by a waveform object with an audio-rate frequency modulation input.
But the main suggestion I have (and, I suppose, feature request) is an audio-rate phase modulation input to the waveform object. This really broadens the synthesis possibilities, allowing one waveform to “drive” another.
In terms of new objects, the most important first additions to a robust synthesis library (imho) would be audio-rate comparison objects: >, < , ==. If these objects could also report logical transitions at the control rate, this would be ideal.
In terms of math, I’d love to see an audio rate modulo operator. Also (while I know this functionality is duplicated in the mixer) a simple adder would be a design timesaver (alternatively, if node-red allows it, have multiple signals connected to the same inlet be automatically added).
Other things I’d like to put on your radar for signal-rate objects:
* audio-rate accumulator
* sample counter with an audio-rate reset trigger
* a smoother that would create a ramp between values over some number of milliseconds. If this could have the possibility of exp/log curvature, that would be ideal.
oh — one more thought — rather than a WAVEFORM_SAMPLE_HOLD, I’d suggest an actual sample_hold object, with a signal inlet and a trigger inlet.
I suspect you are familiar with these, but, if not: in terms of models for expansion of the ASDT, I’d suggest looking to:
Max/MSP (http://cycling74.com)
Reaktor (https://www.native-instruments.com/en/products/komplete/synths/reaktor-6/)
Also (essentially the same as Max/MSP): Pd (http://puredata.info)
Finally, when and if you do flesh out the ASDT, here’s a community you'll want to get the attention of: http://cdm.link
All the best,
John
Firstly, I agree that frequency modulation of all waveforms would be a great improvement — the sine_fm object is terrific, but could probably be replaced by a waveform object with an audio-rate frequency modulation input.
But the main suggestion I have (and, I suppose, feature request) is an audio-rate phase modulation input to the waveform object. This really broadens the synthesis possibilities, allowing one waveform to “drive” another.
In terms of new objects, the most important first additions to a robust synthesis library (imho) would be audio-rate comparison objects: >, < , ==. If these objects could also report logical transitions at the control rate, this would be ideal.
In terms of math, I’d love to see an audio rate modulo operator. Also (while I know this functionality is duplicated in the mixer) a simple adder would be a design timesaver (alternatively, if node-red allows it, have multiple signals connected to the same inlet be automatically added).
Other things I’d like to put on your radar for signal-rate objects:
* audio-rate accumulator
* sample counter with an audio-rate reset trigger
* a smoother that would create a ramp between values over some number of milliseconds. If this could have the possibility of exp/log curvature, that would be ideal.
oh — one more thought — rather than a WAVEFORM_SAMPLE_HOLD, I’d suggest an actual sample_hold object, with a signal inlet and a trigger inlet.
I suspect you are familiar with these, but, if not: in terms of models for expansion of the ASDT, I’d suggest looking to:
Max/MSP (http://cycling74.com)
Reaktor (https://www.native-instruments.com/en/products/komplete/synths/reaktor-6/)
Also (essentially the same as Max/MSP): Pd (http://puredata.info)
Finally, when and if you do flesh out the ASDT, here’s a community you'll want to get the attention of: http://cdm.link
All the best,
John
