Re: varying a synthdef with respect to input frequency

Just to follow up with the solution I've arrived at:

SynthDef("flute", {arg freq = 55, sig = 65;
	var freq1 = In.kr(freq);
	var cutoff_freq1 = 300;
	var wave = Mix.fill(5,[
		SinOsc.ar(freq1)*(-1),
		SinOsc.ar(freq1*2)*0.5,
		{Select.ar(freq1 >= cutoff_freq1, [SinOsc.ar(freq1*3) * (-0.125), SinOsc.ar(freq1*3) * (-0.05)])},
		{Select.ar(freq1 >= cutoff_freq1, [SinOsc.ar(freq1*4) * 0.02, SinOsc.ar(freq1*4) * 0.08])},
		SinOsc.ar(freq1*5) * (-0.333)]);
 	Out.ar(sig, Mix.ar(wave/20));
}).writeDefFile(~dir);

So, I've gotten rid of the loop since it was mainly the first 5 harmonics and just put them into a mix manually. I'm not completely happy with the frequency curve yet, but that's just a matter of tweaking the parameters. I'll probably end up using SelectX instead of Select since I'm pretty sure there will be more cutoffs as I refine it. Thanks for the help!

Regards,
Bill

On Wed, Mar 24, 2021, at 10:17 PM, [email protected] wrote:
> Thanks for the link and the explanation! I can see I've got more to learn.
> 
> Regards,
> Bill
> 
> On Wed, Mar 24, 2021, at 7:36 PM, [email protected] wrote:
> > On Thu, Mar 25, 2021 at 5:36 AM <[email protected]> wrote:
> > > And given the values of freq and cutoff, it prints 200. What I want is to be able to do the same kind of comparison, but on a control signal, but this doesn't compile (if that's the right term). The only difference is that freq and cutoff are now control signals.
> > >
> > (
> > var freq = DC.kr(300);
> > var cutoff = DC.kr(310);
> > z = if (freq > cutoff, 100, 200);
> > z.postln;
> > )
> > 
> > Pressed for time at the moment, so, short answer: read this --
> > https://supercollider.github.io/tutorials/If-statements-in-a-SynthDef.html
> > 
> > "The only difference is that freq and cutoff are now control signals."
> > That isn't the only difference. It means that 300 and 310 now exist in
> > the server only. The true and false functions are *language*
> > constructs. They don't know the server values. Also, postln is for
> > language side values.
> > 
> > If you use the UGen-if-interpolation style, you'd get:
> > 
> > (
> > var freq = DC.kr(300);
> > var cutoff = DC.kr(310);
> > z = if (freq > cutoff, 100, 200);
> > z.postln;
> > )
> > 
> > -> a BinaryOpUGen <<== this is telling you it's a server thing
> > 
> > (
> > {
> >     var trig = Impulse.kr(0);
> >     var freq = DC.kr(300);
> >     var cutoff = DC.kr(310);
> >     var z = if (freq > cutoff, 100, 200);
> >     // z.postln; nope
> >     Poll.kr(trig, z);
> >     Silent.ar(1)
> > }.play;
> > )
> > 
> > UGen(BinaryOpUGen): 200
> > 
> > *But* when you have a lot of these, I think my suggestion will be more
> > efficient.
> > 
> > hjh
> > 
> > _______________________________________________
> > sc-users mailing list
> > 
> > info (subscription, etc.): 
> > http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
> > archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
> > search: http://www.listarc.bham.ac.uk/lists/sc-users/search/
> >
> 
> _______________________________________________
> sc-users mailing list
> 
> info (subscription, etc.): 
> http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
> archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
> search: http://www.listarc.bham.ac.uk/lists/sc-users/search/
>

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/