Hi,

I am trying to create some abstractions that would create the basic waveforms in the control domain. I figured this would be cheaper than using signal objects with snapshot to alter control values, when for example setting the frequency of an [osc~] object.

If this really is better I do not know.

I have been fooling around with [sin], [line], [metro] and [delay] to create sine, triangle, saw, square and pulse width control messages, but I would be happy if I could get an example on how to do it.

Or maybe it would be better to do it in tables?

Thanks!

@cfry you can use a metro with a counter and [wrap] to make a control-rate phase accumulator going from 0-1. Then for each waveform create some kind of transfer function, you could also use [tabread4] for functions in tables, of course.
see [c_lfo] (which uses the message rate phase accumulator [c_mphasor]) in rjlib https://github.com/rjdj/rjlib for an example of some simple waveforms

as an example a simple square wave could be accomplished with a wrapped phase-accumulator going into a [> 0.5], then to control the duty cycle change the 0.5 by putting different values from 0-1 in the right inlet.

@seb-harmonik.ar that is a nice library, so many useful objects. To bad the objects are depending on so many of the other objects. I try to stay vanilla. I could not find it on deken btw.

In my own fuzzy experiments with [sin] and [metro] it became way heavier for the cpu than [osc~] and [snapshot~].

@cfry here is a quick hack of my [curve~] abstraction for control rate, it will not make a perfect sine, but it gives a good deal of control. PW/PWM can be implemented as above with [>].

curve.pd

The original is over on the abstraction page, has some info and shows a couple other features (end out and hold) which I did not bother with in this quick hack.

Edit: Should have shown the end out, but it is just the unused output (right) of [pd fifo], use this to retrigger the list to get LFO opperations.

@oid Thanks, I will try it out in my next project!

@cfry If you set the grain for [line] in @oid 's [curve] the graph will have greater resolution......
David.

@whale-av Good catch, missed that in my quick hack.

@oid Worth catching....... it's a good hack.....
David.