Re: [Sc-devel] Amplitude.kr problem

[Julian Rohrhuber <rohrhuber@xxxxxxxxxxxxxx>]

> A while ago Julian reworked the Amplitude UGen slightly to allow the
> coefficients to be modulated. Julian, you left some old calc-func code
> in there, deactivated using some "#if" macros. Is it OK to remove the
> old code? Would make it easier to work on fixing the current code...

I don't know which of the two are to be the better one. Well the 
first is being used, so maybe it is supposed to be the one.


#if 1

void Amplitude_next(Amplitude* unit, int inNumSamples)
{
	float *out = ZOUT(0);
	float *in = ZIN(0);

	float relaxcoef = unit->m_relaxcoef;
	float clampcoef = unit->m_clampcoef;
	float previn = unit->m_previn;

	float val;
	LOOP(inNumSamples,
		val = fabs(ZXP(in));
		if (val < previn) {
			val = val + (previn - val) * relaxcoef;
		} else {
			val = val + (previn - val) * clampcoef;
		}
		ZXP(out) = previn = val;
	);

	unit->m_previn = previn;
}

#else

void Amplitude_next(Amplitude* unit, int inNumSamples)
{
	float *out = ZOUT(0);
	float *in = ZIN(0);

	float maxval = 0.f;
	LOOP(inNumSamples,
		float val = fabs(ZXP(in));
		if (val > maxval) maxval = val;
	);

	float previn = unit->m_previn;
	float coef = maxval < previn ? unit->m_relaxcoef : unit->m_clampcoef ;

	LOOP(inNumSamples, ZXP(out) = previn = maxval + (previn - 
maxval) * coef; );

	unit->m_previn = previn;
}

#endif


> Dan
>
>
> 2007/12/13, Josh Parmenter <josh@xxxxxxxxxxxxxxxxx>:
> >  I think option 1 makes sense. Pitch.kr can take an audio input, so it
> >  makes sense (to me at least) that Amplitude should be able to also.
> >
> >  Josh
> >
> >  On Dec 13, 2007, at 1:50 PM, Dan Stowell wrote:
> >
> >  > Hi all,
> >  >
> >  > Applying Amplitude.kr to an audio-rate signal is allowed, in fact all
> >  > the examples in the helpfile do this. However, I noticed that the UGen
> >  > doesn't really handle this case "correctly" - if run at control-rate
> >  > it simply expects its input to be control rate; the end result on
> >  > audio-rate input is a dodgy downsampling which can lead to incorrect
> >  > results.
> >  >
> >  > For example, these two should really give similar results but they
> >  > don't (give them a few seconds to converge):
> >  >
> >  > x = {(Amplitude.kr(Impulse.ar(44100/64, 0.0),1,1)).poll}.play // 1
> >  > x = {(Amplitude.kr(Impulse.ar(44100/64, 0.02),1,1)).poll}.play // 0
> >  >
> >  > Same goes for these:
> >  >
> >  > x = {(Amplitude.kr(SinOsc.ar(44100/64, 0.0),1,1)).poll}.play // 0.098
> >  > x = {(Amplitude.kr(SinOsc.ar(44100/64, pi/4),1,1)).poll}.play // 0.77
> >  >
> >  > Options for fixing this:
> >  >
> >  > (1) Alter the UGen source, add a new calc func to deal with the a-to-k
> >  > (2) Make Amplitude a subclass of Filter, so it refuses to run k-rate
> >  > on an a-rate input (unhelpful!)
> >  > (3) Alter the Amplitude class so it automatically replaces
> >  > Amplitude.kr(x.ar) with A2K.kr(Amplitude.ar(x.ar))
> >  >
> >  > Any particular thoughts?
> >  >
> >  > Dan
> >  >
> >  > --
> >  > http://www.mcld.co.uk
> >  > _______________________________________________
> >  > Sc-devel mailing list
> >  > Sc-devel@xxxxxxxxxxxxxxx
> >  > http://www.create.ucsb.edu/mailman/listinfo/sc-devel
> >
> >  ******************************************
> >  /* Joshua D. Parmenter
> >  http://www.realizedsound.net/josh/
> >
> >  "Every composer - at all times and in all cases - gives his own
> >  interpretation of how modern society is structured: whether actively
> >  or passively, consciously or unconsciously, he makes choices in this
> >  regard. He may be conservative or he may subject himself to continual
> >  renewal; or he may strive for a revolutionary, historical or social
> >  palingenesis." - Luigi Nono
> >  */
> >
> >
> >  _______________________________________________
> >  Sc-devel mailing list
> >  Sc-devel@xxxxxxxxxxxxxxx
> >  http://www.create.ucsb.edu/mailman/listinfo/sc-devel
>
>
> --
> http://www.mcld.co.uk
>
> _______________________________________________
> Sc-devel mailing list
> Sc-devel@xxxxxxxxxxxxxxx
> http://www.create.ucsb.edu/mailman/listinfo/sc-devel


--





.