fractal waveform virtual instrument

[SODDI]

There was a big fractal music community about 10 or 12 years ago - most of it dedicated to fractal MIDI composition. Most of the software is outdated and moribund.

http://www.fractalartist.com/art/fracta ... ic01.shtml

[Kalamata Kid]

It's not old!
It's ancient.

[thomni]

I've had some toying thoughts with the fractals in music idea.
Like making a composition, and then speeding it up x2 and layering it over and such.

Other thoughts like how can I get a speaker to play a tone. E.g. if I get an additive synth like alchemy and get the partials right so as to infer a square wave or something, then continually modulate the amplitude at lower and lower frequencies to try and get a sub-subsonic bass or something.

Unfortunately, none of those turn out well, but I like your idea. Trying to overcome limitations of human perception and modern-day hardware is where it's at though.

[Kalamata Kid]

I've had some toying thoughts with the fractals in music idea.
Like making a composition, and then speeding it up x2 and layering it over and such.

Other thoughts like how can I get a speaker to play a tone. E.g. if I get an additive synth like alchemy and get the partials right so as to infer a square wave or something, then continually modulate the amplitude at lower and lower frequencies to try and get a sub-subsonic bass or something.

Unfortunately, none of those turn out well, but I like your idea. Trying to overcome limitations of human perception and modern-day hardware is where it's at though.

I also love to experiment with audio.

My idea for fractal waveform was that fractal have an infinite detail that a waveform can use a part of that detail to make the audio thicker and more musical. Secondly there may be all kinds of audio fx that can be derived form a fractal waveform.

The equations that make fractals will have to be mangled and tweaked to make them look and have properties of a waveform. The mangling part I can do however the "tweaked to make them look and have properties of a waveform" will take some athematician/musician genius.

[Kalamata Kid]

Check out what Architeuthis is doing with Chaos.
viewtopic.php?f=1&t=468238&p=6696540#p6696540 He is creating a 2D linear wave form. My thinking was to do the same thing but converting a 3D fractal into 2D linear wave form. Then have a simulated 3D fractal controller similar to Serums which controls a wavetable. But since fractals have infinite depth (detail) there will be an additional dimension to probe.

Architeuthis; Feb 04 2017 18:05
Re: working on a tonal chaos generator vst
http://www.elanhickler.com/_/chaosgen/c ... growls.mp3


[/quote]

[deastman]

If you're still talking Mandelbrot set, the question is how are you going to take points on the complex plane and apply those values to audio parameters?

[Kalamata Kid]

If you're still talking Mandelbrot set, the question is how are you going to take points on the complex plane and apply those values to audio parameters?

Little that I know about this is to modify the equations so instead of a having a 2D image you have a 2D image that is linear. If they can take the rather simple fractal equation and modify it to make such beautiful images then with a little ingenuity they can make the fractal form a linear image. Look at the progress Architeuthis has made using Chaos theory.

Are you say this cannot be done?

[whyterabbyt]

If you're still talking Mandelbrot set, the question is how are you going to take points on the complex plane and apply those values to audio parameters?

Little that I know about this is to modify the equations so instead of a having a 2D image you have a 2D image that is linear. If they can take the rather simple fractal equation and modify it to make such beautiful images then with a little ingenuity they can make the fractal form a linear image. Look at the progress Architeuthis has made using Chaos theory.

Are you say this cannot be done?

No, he's saying you have to think about how you usefully map the one thing onto the other.

With Mandelbrot fractal images, for example, the colour at each point in a 2D plane is (to simplify) determined by whether a certain number of iterations of a recursive formula exceeds a given cutoff point. To apply that algorithm to audio synthesis you'd have to think not just about the technical side (ie handling this massive set of iterations), but what that plane and those cutoff points actually mean in audio terms.

And fractals arent one thing, they're a whole class of things, with lots of different sub-types. Some will be easier to leverage, but you'd have to think about which ones, and how.

For audio, IMO, Ive always considered that an L-system grammar driving scanned-wavetable lookup might be a good example of one possible mapping.

[Kalamata Kid]

There Will have to be a way to avoid the back or uninteresting part of the fractal image.

There will also have to be a way to control the path that is followed so the musician
has the means to express his/her musical creatively.

If I knew how this can be done I would very likely make a plugin.
See some links in the OP.

Using L-system grammar may be the way to go.

[VariKusBrainZ]

The downloadable version of this used to work Ok but buggy on my Win10.

cross platform

http://www.jdnmusic.com/fractal-sound-design

[deastman]

There Will have to be a way to avoid the back or uninteresting part of the fractal image.

There will also have to be a way to control the path that is followed so the musician
has the means to express his/her musical creatively.

If I knew how this can be done I would very likely make a plugin.
See some links in the OP.

Using L-system grammar may be the way to go.

Are you familiar with the actual algorithm for the Mandelbrot set? No offense, but your statements come across as the naive "gut instinct" of someone who thinks something would intuitively be "really cool" without any fundamental understanding of the subject. Whyterabbyt provided a basic overview of the algorithm involved, and you can easily learn the full details on, for example, the relevant Wikipedia page. It isn't a question of avoiding the blank space. The blank space actually MEANS SOMETHING in terms of the algorithm. The 2D visual you're familiar with is simply a visual representation of the points contained within the set, according to the algorithm. The question remains: how would you convert the points contained within the set in a meaningful way to an audio representation? Not in an abstract "yeah it would be cool" sort of way, in specific, concrete terms which can be implemented in software? How are real numbers represent? How are imaginary numbers represented? How is the iteration count represented? How is the escape threshold represented? What do these properties mean in terms of an audio signal? This is your concept, so the burden is on your shoulders to provide the details. Otherwise, you might as well be suggesting that cauliflower synthesis seems like an awesome idea, with no more specific details as to how that would work in practice. Again, I'm not trying to be rude or insulting here, but simply to get you to think harder about the details of your idea.

[VariKusBrainZ]

Can Mandelbrot set fractals be tweaked so to be viewed as a waveform and be used as a sound source?

Yes, convert an image to audio and use the resulting sample.

[Kalamata Kid]

The downloadable version of this used to work Ok but buggy on my Win10.

cross platform

http://www.jdnmusic.com/fractal-sound-design

Thanks for the link. Interesting.

Did a bit more research of what has been done previously.
https://www.google.com/webhp?sourceid=c ... +converter

[Kalamata Kid]

Can Mandelbrot set fractals be tweaked so to be viewed as a waveform and be used as a sound source?

Yes, convert an image to audio and use the resulting sample.

Import a fractal image into 2CAudio Kaleidoscope and get the audio output?
http://www.2caudio.com/products/kaleidoscope#_overview
Or
Photosounder? http://photosounder.com/ . This will do it.

Another way I was thinking about a fractal equation to be modified so that it would directly create a linear image similar to a waveform. Will it be able to be tweaked to output something musical?

The advantage of a fractal waveform will be that is has an infinite detail. But from the fractal images I have seen they do not have infinite variability. That it why I easily get bored with them. As beautiful as the fractal images are they all have some similarity. It will be like being stuck listening to Baroque music and only Baroque music. But then some creative minds might find a way to expand the fractal waveform possibilities. Maybe combine fractals and Chaos theory?

[deastman]

Can Mandelbrot set fractals be tweaked so to be viewed as a waveform and be used as a sound source?

Yes, convert an image to audio and use the resulting sample.

Import a fractal image into 2CAudio Kaleidoscope and get the audio output?
http://www.2caudio.com/products/kaleidoscope#_overview
Or
Photosounder? http://photosounder.com/ . This will do it.

Another way I was thinking about a fractal equation to be modified so that it would directly create a linear image similar to a waveform. Will it be able to be tweaked to output something musical?

The advantage of a fractal waveform will be that is has an infinite detail. But from the fractal images I have seen they do not have infinite variability. That it why I easily get bored with them. As beautiful as the fractal images are they all have some similarity. It will be like being stuck listening to Baroque music and only Baroque music. But then some creative minds might find a way to expand the fractal waveform possibilities. Maybe combine fractals and Chaos theory?

You haven't responded to my previous post yet. I kind of feel like you don't understand the algorithm behind the Mandelbrot set, and now you want to mix and match fractal geometry with chaos theory?!

Throwing a rendering of a fractal into Kaleidoscope or Photosounder or Metasynth is hardly a "fractal waveform virtual instrument". The results won't be substantially different from a multitude of other ways of deriving images to feed into those programs. If you want a fractal oscillator, you'd better understand how fractals are computed and how to apply those algorithms to a fundamentally different domain.