Implement freq2cv only via binary op ugens

[capital-G]

If the freq to cv coversion could be done on the server and the language we could completely disguise the eurorack module in supercollider as a normal synthdef.

I think it is feasible but ugly as the current code is also really messy. Biggest obstacle should be the morphing between the cv values in a dynamic array [albeit it is not dynamic at runtime]

[capital-G]

Maybe it is as easy as using a SelectX as we have a linerar composition where we always have the same distance.

[capital-G]

a first sketch

(
SynthDef(\foo, {|out, freq, tuningX=#[200, 300, 400, 500], tuningY=#[0.1, 0.5, 0.6, 0.7]|
	var sig;
	var foo;
	var which;
	
	which = LinLin.kr(
		in: freq, 
		srclo: tuningX[0], 
		srchi: tuningX.reverse[0], 
		dstlo: 0, 
		dsthi: tuningY.size
	).poll(label: "linlin");
	
	foo = Select.kr(freq>tuningX[0], [
		// if lower boundary crossed
		DC.kr(tuningY[0]) * (freq/tuningX[0]),
		Select.kr(which <= tuningY.size, [
			// if tuning matches range
			SelectX.kr(which, tuningY),
			// if upper boundary is crossed
			DC.kr(tuningY.reverse[0]) * (freq/tuningX.reverse[0]),
		]);
	]);
	foo.poll(label: "foo");
	sig = SinOsc.ar;
	Out.ar(out, sig*0.01);
}).add
)

a = Synth(\foo);

a.set(\freq, 200);

a.set(\freq, 500)

a.set(\freq, 200)

[capital-G]

Pro:

• would allow for modulation from the server (e.g. applying a tremolo)

Con:

• Must use Pmono as otherwise we will have overlapping CVs