@Morphoice

You can map it to a function of best fit using python / matlab.
Ask chat gpt how to do this from a table of data.

Alternatively you could create a lookup table of accurate values.

I didn't realize you were going for analog tracking, that's a different case to digital, yes.
The tracking lookup only happens once per note, so you don't have to worry about efficiency here. Envelopes by comparison need optimization as they are applied per sample.

@clevername27

yep it was just to bring it into view. I'd rather have more parameters than use sliderpacks because I had issues with those in the past. But that is a last resort

@Morphoice

Oh dear! Well you've been working hard, hopefully you'll feel back to normal soon!

@Morphoice

Pseudocode. I'm sure you've figured it out by now, but might as well post it since I remembered that my original response sucked lol.

FUNCTION keytrack(midiNote, baseNote, baseFreq, trackingAmount)

    // Step 1: Find the difference in semitones
    semitoneDifference = midiNote - baseNote

    // Step 2: Convert to a freq multiplier
    frequencyScale = 2 ^ ( (semitoneDifference * trackingAmount) / 12 )

    // Step 3: Apply scaling to the base freq
    RETURN baseFreq * frequencyScale

END FUNCTION

Example usage:

midiNote        = 72        // Target note (C5)
baseNote        = 60        // Reference note (Middle C)
baseFreq        = 1000 Hz   // Reference frequency (e.g. filter cutoff)
trackingAmount  = 1         // Full keytracking

trackedFreq = keytrack(midiNote, baseNote, baseFreq, trackingAmount)

// Expected Output: 2000 Hz

@griffinboy

bump, apologies

@aaronventure

Ah, but event ID increments. So I assume you're saying that I have to keep track and figure out which voices are which at a particular time?

I need to be able to, for instance, say detune voice two up an octave. When I press and hold three notes on the keyboard, voice 1 2 and 3 are now active. When I release the note that voice 2 was playing, and play a new note, this new note is using voice two as voice 1 and 3 are active. And therefore should inherit voice 2 detune.

I just want to make sure there isn't already a way to do what I'm needing. Thanks for your responses

@aaronventure

Thanks and sorry: event is? I can't see that in the API

It doesn't seem to be part of event data. Is there any way to get the voice number to do voice specific processing? I want to use the voice number to send event data to specific voices.

function onNoteOn()
{
	Message.ignoreEvent(true);
	
	// I'd like to be able to get hold of the voice number to link the detune knobs via voice

	Event.noteIds[Message.getNoteNumber()][0] = Synth.playNote(Message.getNoteNumber(), Message.getVelocity());
//	Event.noteIds[Message.getNoteNumber()][1] = Synth.playNote(Message.getNoteNumber(), Message.getVelocity());
	
	globalRouting.setEventData(Event.noteIds[Message.getNoteNumber()][0], 0, ControlValue.detune1);
//	globalRouting.setEventData(Event.noteIds[Message.getNoteNumber()][1], 0, ControlValue.detune2);
	
}

An example @aaronventure wrote for me a while back (now in the snippet browser) exemplified how to create multiple notes from the same voice and detune them differently, but I'm not sure how to do this kind of thing across voices as well! I'm not very good with events yet.

@pelle

Sounds like a bit of work but certainly. You'll have to design the algorithm though and do lots of thinking and testing to actually get something like that behaving smoothly. Try to prototype something simple and then extend it.

There seems to be a cap limit of 18 parameters on the UI of c++ custom nodes.

Any more than that and they won't show up on the UI.
For the particular node I am making I need at least double that.
Please can you alter this @Christoph-Hart ?

@griffinboy

Update*

Adding the preprocessor definition that Christoph mentions in this post

NUM_POLYPHONIC_VOICES=1

in every available preprocessor box (including the juicer file when building Hise from GitHub, and the project preprocessor boxes) seems to have worked. Sometimes I get errors when running Hise debug and going over the limit. But in regular release Hise the limit seems to work. Now that I think about it, I wonder if I was supposed to track down this constant and change it directly in the source code.

I've noticed quirks such as the seems that it lets you play one less voice than the limit, and it also cancels voices in a bit of an odd way. Maybe there is still something to do, or this is just how it works.

Voice limit must be set to a power of 2, eg 8, or you'll get an error.

@cassettedeath

Sounds like you've already got the correct info.
There is a trick that some companies use where on the plugin page they have a disclaimer fully explaining that the product is not associated with the mentioned brand names and that the names are only mentioned for context and history on the technology.

I don't know anything about it, but perhaps that's an option that gets you out of trouble on the product page.

@hisefilo

Yeah it's c++ hise.

I mean, we have loris which is kind of better from the looks of it. Unless I decide to implement the latest research mine will not be better. But I can't say I care about additive to that extent. Maybe in the future I'll have a better reason to do all that work

@clevername27

Maybe I'll post about it.
Ah yes - you are correct. You can't simply break a sound into it's actual sine waves perfectly (although reading some of the latest research on fft, there are some algorithms that are so good at it that it almost doesn't matter, not that I've implemented those yet). It's more a case of applying multiple passes of different FFT techniques to construct a map of everything, and then try and rebuilt something close. But unless you do a ton of passes and maybe use the most complicated fft research, there will be information lost and resolution issues

@clevername27

Loris is additive. And it is possible to decompose sounds into sines. I did a crude job over the weekend. You use multiresolution FFT and then try and work out where the transients would have been using FFT reassignment. And then you increase the contrast of the spectrogram and track partials pitch, phase and amplitude over time, storing them in an array. You can then manipulate the partials in this array, and reconstruct the signal afterwards. The special thing that Loris does is that it can also analyse the noise components in a signal, which is beyond what my program can do.

I'm tempted to give in and just use Loris...

@Morphoice

Idk what is standard, but I always use middle C as the base frequency. And then the filter basically doubles frequency every octave, moving in these increments

@Christoph-Hart

Hi is there a specific place to do this in the source code? Just want to be sure, sorry for asking!

@Morphoice

True. A state space model is probably ideal for such a filter.
These things are a pain to derive though.

@Morphoice

Not sure I can help: my approach would've been component modelling. That's probably one of the easiest ways to get the instability implicitly. Modelling that kind of thing manually can get complex if accuracy is the goal. I think at somepoint you hit limits with what you can do with simpler dsp

@clevername27

Idk, but if this is c++, you can read the error messages by running the .sln of the node under

dsp networks > binaries > builds > visual studio

And then compiling it for hise debug

(for windows at least that is the way)