about: Poly-Ana

[sambean]

Oh wait, I got it now... needed to remove that line. Also I had made some changes that I forgot to put back.

Seems to be working!! Testing and evaluating...

Good to hear, I'm not sure what optimization will do to the function, but if you have any warnings when compiling about the float/long reinterpritation, you can perform the same using a float/long union.

Sam

[mistertoast]

Hmmm, I'll give it a try then, but it seems like it'll be stronger than what I expect. Maybe that's just because, again, I only deal with the low end of the curve. (Obviously I'm programming from the gut here. But it got me this far! )

I think I must be programming by my gut, too.

And yes, a more optimal approximation would be nice. Also would be nice if this could be built into the pitch to frequency calc! i.e. in a special case version just for the filter.

I don't think it's too awfully bad to use the correct form if you have a sin table.

Curious, what form of frequency DO you have for your current note?

I just have the stride distance as a float for moving through my waveforms. But that could change soon, anyhow. I'm still in the early stages.

I think you've always had the filter cutoff tuning problem even at 8x, but at that rate all your audible frequencies were smashed down near zero, where sin(x) ~= x.

[AdmiralQuality]

I think you've always had the filter cutoff tuning problem even at 8x, but at that rate all your audible frequencies were smashed down near zero, where sin(x) ~= x.

Yes, could very well be. Will test. Unfortunately I'm chasing a bug at the moment and can't get to it right now. Will ASAP though.

[mistertoast]

Mine is fixed now. At least with Q (resonance) at zero. Current problem is that resonance on the high notes is much "bigger" than resonance in the low notes.

I tell ya, you solve one problem, another one pops up. I'll try a simple scale first. It'll probably be another irritating curve to solve the problem again.

[AdmiralQuality]

[mistertoast]

AQ,

It's startling, isn't it? The number of trade offs when you write a synth. I think aciddose has mentioned it--there are so many details to make a judgment on that it's no wonder two synths don't sound the same.

When I started, I expected the process of coding a synth to be much more cut & dried than it is.

[AdmiralQuality]

Agree! Though I can't say it was a surprise. The devil is in the details!

[liqih]

for pitch to freq factor this is my function,
6th order approx

FF = pitch*0.028881;
FF = ((((FF*0.166667+0.5)*FF)+1)*FF)+1; // you may expand this if you wish
FF = FF*FF;

[AdmiralQuality]

for pitch to freq factor this is my function,
6th order approx

FF = pitch*0.028881;
FF = ((((FF*0.166667+0.5)*FF)+1)*FF)+1; // you may expand this if you wish
FF = FF*FF;

Changed first line to FF = pitch*0.346572; as I'm feeding it semitones.

Seems to work OK at the high end, at the low end it goes all out of whack.

[camsr]

I know in FFT processing the lower frequencies are less defined compared to the higher ones, because it is a linear relationship. Think about that maybe?

Think about it this way AQ, you're trying to represent infinite amounts of frequencies with a finite number of knob positions (you know what I mean, float precision). If you want my advice make a lookup table corresponding to quartertones or even finer, and don't bother representing frequencies that aren't tonal to the equal tempered scale. But I don't know if this will work for adjusting the keyboard track amount... it would be like snapping-to-grid for the filters cutoff.

[AdmiralQuality]

If you want my advice make a lookup table corresponding to quartertones or even finer, and don't bother representing frequencies that aren't tonal to the equal tempered scale. But I don't know if this will work for adjusting the keyboard track amount... it would be like snapping-to-grid for the filters cutoff.

...not to mention Global Tune, Fine Tune, Pitch Bend, and Frequency Mod. I need an "infinite" number of pitches because my customers do.

EDIT: oop, almost forgot Portamento!

[VladimirDimitrievich]

this is the reason why im not doing filters until ive run it through my little analyzer program. and that will have to wait because i havent added fft to it yet. and that will have to wait because im currently busy making my oscillators sound better.

[mistertoast]

liqih,

Where does your approximation come from? What are you approximating, exactly? Is the approximation getting rid of the transcendental that I'm using now?

[camsr]

arke, whatever synth you make, I want to see a Huuge knob on, just like tarnce VST. I usually find I gotta turn up the HUUGE in my tracks.

[VladimirDimitrievich]

arke, whatever synth you make, I want to see a Huuge knob on, just like tarnce VST. I usually find I gotta turn up the HUUGE in my tracks.

i cant believe anybody actually remembers that