Title: Calculating sound using fractal algorithms, what's out there ? Post by: djbarney on April 25, 2016, 03:05:56 PM Hi, I looked around in this section and did a search but most, if not all, of the work being done with music using Fractal mathematics appears to involve the production of MIDI sequences and so forth. Great ! Love to see the experimentation and work :dink:. But what I'm interested in is doing the audio/sound equivalent of what Mandelbulb 3D or Fractinit are doing. I have not seen any software yet that is doing what those applications are doing BUT ARE FOR SOUND. I think people make the mistake of trying to port a VISUAL fractal into sound. This is wrong. That kind of software is programmed from the bottom up to produce graphics for us TO LOOK AT. This is an entirely different set of problems and an entirely different domain to sound. Everything I've heard over the last 20 years that claims to be "Fractal music" sounds to, my ears at least, very little like what I am seeing in Mandelbulb 3D. I want to select a Fractal synth sound and have it sound as amazing as those Mandelbulb renders look ! O0 But the field of Fractals for sound and music seems to have been left behind, unless I'm missing something out there.
So are there any applications out there like Mandelbulb 3D, but for sound ? I don't want applications that generate something visual and then transfer that to sound as that is fundamentally the wrong way to go about it. I have tinkered with Pure Data ( https://en.wikipedia.org/wiki/Pure_Data ) before so plan to put my programming fingers where my mouth is if necessary, unless there's some huge software application that I've completely missed for some reason. Title: Re: Calculating sound using fractal algorithms, what's out there ? Post by: quaz0r on April 25, 2016, 03:27:07 PM basically what you want doesnt exist. its easy to say "fractals look cool, wouldnt it be neat if we could do that for sound." but what would that actually be? ... is basically a question nobody has come up with a good answer to yet, if ever they will. short of like you say coming up with simple rules for generating sequences of notes.
Title: Re: Calculating sound using fractal algorithms, what's out there ? Post by: lycium on April 25, 2016, 06:20:49 PM Have you checked out Aexion's "Aural"?
http://www.auralfractals.net/ Title: Re: Calculating sound using fractal algorithms, what's out there ? Post by: Adam Majewski on April 25, 2016, 06:55:32 PM http://mathr.co.uk/blog/2016-04-15_nebulullaby_launch.html
Title: Re: Calculating sound using fractal algorithms, what's out there ? Post by: Ville on April 26, 2016, 05:09:25 PM I was also thinking a lot how to turn fractals into music.
Traditional music composes typically of a sequence of notes, which have pitch and rhythm. A still figure is then made of pixels with color. You can access each pixel whenever you want, and you dont have the concept of time there. There are not many similarities between a figure and music. Color is something similar to timbre (sound color), but pitch or rhythm have no direct counterparts in figures. I had to go to very basic fractals to find anything that could be turned into music relatively directly. (If you havent heard my fractal music, it is here: http://www.fractalforums.com/fractal-music/two-fractal-music-compositions/ ) My fractals are quite far from the elegance of Mandelbrot, resembling more e.g. Koch snowflake. However, at least some people seem to like how they sound. The first visual fractals were quite simple, and only after some time more complex ones appeared. Perhaps more complex musical fractals with with more sophisticated structures are waiting for their finders.. :-) Best, Ville Pulkki Title: Re: Calculating sound using fractal algorithms, what's out there ? Post by: matsoljare on April 27, 2016, 02:17:29 AM Did you see my thread on it? Haven't been much development since, but there are samples you can listen to....
http://www.fractalforums.com/new-theories-and-research/new-kind-of-'audio-fractal'-discovered Title: Re: Calculating sound using fractal algorithms, what's out there ? Post by: djbarney on April 27, 2016, 06:14:22 PM Thanks for your replies.
Been thinking about this. Of course much music is already a sonic equivalent of a Mandelbulber render. This is one of the reasons it sounds so good because it has inherent fractality in composition and at smaller and smaller levels. For example reverb and echo are inherently self similiar as evidenced by looking into a spectral waveform of the sound that when zooming in looks like finer and finer feather/wing like structures. I had a good example image of this and will post it here when I find it. So just as Fractal software mimics the beauty of nature, music also reflects what we see around us in nature. As is pointed out in the article below music has always been algorithmic with composers using mathematical approaches well before computers. Apparently Mozart had a dice game he used to construct compositions. Here is the article. A really good article about all this that goes into the entire history along with a few examples of actual applications. Realtime Algorithmic Music Systems From Fractals and Chaotic Functions: Toward an Active Musical Instrument (http://mate.dm.uba.ar/~umolter/materias/geo-fractal-2-11/papers%20fractales%20y%20musica/fractal%20algorithmic%20music.pdf) Title: Re: Calculating sound using fractal algorithms, what's out there ? Post by: Chillheimer on April 27, 2016, 08:04:48 PM .. a spectral waveform of the sound that when zooming in looks like finer and finer feather/wing like structures. I had a good example image of this and will post it here when I find it. please do!! would love to see this visualized! :)and while were at it: good drumming is fractal: http://www.sciencemag.org/news/2015/06/secret-groovy-drumming-may-be-math fractals in bachs music: http://mathtourist.blogspot.de/2008/09/fractal-in-bachs-cello-suite.html fractal 1/f noise and other fractal occurences in music: https://courses.physics.illinois.edu/phys406/Lecture_Notes/P406POM_Lecture_Notes/P406POM_Lect15.pdf and, my personal research: techno, especially psytrance/goa is a an acoustic cantor set. that one has been sitting on my to-do list a while, I need to write that article.. ;) I also love that microphone or guitar feedback is the result of a recursive 'formula' and in my broad fractal view also is an acoustic fractal. Title: Re: Calculating sound using fractal algorithms, what's out there ? Post by: hgjf2 on April 30, 2016, 05:05:57 PM Is obvoiusly that a reverb is complicated and is ordered like a fractal. As example into chamber cube shaped if you clap you arms, the acoustical waves, will reflecting by walls, floor and ceiling but neperiodical becuase the sperical wave will knotting increasing harder becoming increasing dense. An example how clap sounding at minute {0; 1; sqrt(2); sqrt(3); 2; sqrt(5);...} but missing sqrt(7), sqrt(15), sqrt(23), sqrt(28), sqrt(31),... sqrt (4k*(7+8l)), but the intensity: 1;6;12;8;6;24;24;0;8;30;24;24;8;24;48;0;6;48;36;... at intervaly sqrt(n)
Title: Re: Calculating sound using fractal algorithms, what's out there ? Post by: hgjf2 on April 30, 2016, 05:09:45 PM Is obvoiusly that a reverb is complicated and is ordered like a fractal. As example into chamber cube shaped if you clap you arms, the acoustical waves, will reflecting by walls, floor and ceiling but neperiodical becuase the sperical wave will knotting increasing harder becoming increasing dense. An example how clap sounding at minute {0; 1; sqrt(2); sqrt(3); 2; sqrt(5);...} but missing sqrt(7), sqrt(15), sqrt(23), sqrt(28), sqrt(31),... sqrt (4k*(7+8l)), but the intensity: 1;6;12;8;6;24;24;0;8;30;24;24;8;24;48;0;6;48;36;... at intervaly sqrt(n) If you clap hands into cubed box at irrational distance of walls comparative with length , the sound intervals are like as this example: 17 53.7544467966324 72.2786404500042 88.7157287525381 109.033087246637 125.47017554917 143.994369202542 145.284271247462 161.721359549996 180.245553203368 180.748815999175 182.038718044094 198.475806346628 200.562911697466 233.437088302534 235.524193653372 237.317358494099 251.961281955906 270.191535099166 290.005630797458 290.508893593265 307.23992240591 308.52982445083 324.966912753363 326.76007759409 327.557280900008 345.787534043269 360.431457505076 362.224622345803 363.808464900834 364.311727696641 396.682641505901 397.185904301709 399.273009652547 415.71009795508 417.503262795807 418.793164840727 435.230253143261 436.027456449179 452.464544751713 453.251184000825 455.84155214747 473.568542494924 474.071805290731 490.802834103376 492.595998944103 505.152817055072 506.442719099992 506.945981895799 510.322989291556 523.677010708444 528.847182944928 541.907263851704 543.197165896624 543.491106406735 562.51856285591 563.51452439072 565.60162974156 578.15844785253 578.955651158448 598.769746856739 601.066174493273 614.912894649162 615.206835159273 618.28990204492 619.087105350838 633.940351098341 634.726990347454 636.814095698291 655.044348841552 668.39837025844 669.688272303359 670.485475609277 670.988738405085 671.481437144086 671.775377654197 672.781903245811 688.212465956731 689.218991548345 692.092736148296 707.743185201717 727.054018104201 727.263340389897 729.350445740735 741.404001055897 747.077436088189 759.928194709269 778.661710648337 779.951612693256 781.535455248287 782.541980839902 782.835921350013 784.125823394932 797.972543550821 798.979069142435 802.356076538193 816.496737204193 819.590368146645 820.880270191565 832.147186257614 834.234291608452 849.874176605068 851.667341445795 854.048387306744 854.25770959244 854.551650102551 855.547611637359 868.901633054247 871.984699939893 886.6286234017 886.922563911811 887.425826707619 891.306096899183 891.798795638185 908.235883940718 910.826252087363 911.120192597475 924.180273504251 925.764116059282 942.201204361815 943.700428692431 944.49763199835 946.08147455338 947.3713765983 947.874639394107 963.011261594915 976.86854580761 982.332658554205 995.392739460982 999.765708391548 This is generated with VISUAL BASIC with the code: Sub test2() Dim a, b, c, et, nn, nm As Long Dim na, nb, nc, rez As Double Dim coada1(1 To 1000000) As Double Dim coada2(1 To 1000000) As Double et = 1 For a = 1 To 100 For b = 1 To 100 For c = 1 To 100 na = (Sqr(2) + (a - 50) * 10) nb = (Sqr(5) + (b - 50) * 10) nc = (Sqr(10) + (c - 50) * 10) rez = na * na + nb * nb + nc * nc If ((rez) < 10000) Then Debug.Print (rez); Debug.Print (" ; "); coada1(et) = rez et = et + 1 End If 'Debug.Print (rez); 'Debug.Print (" ; "); Next c Next b Next a Debug.Print ("=") For nn = 1 To et 'Debug.Print (coada1(nn)) For nm = 1 To nn If (coada1(nn) < coada1(nm)) Then coada2(nm) = coada1(nn) coada2(nn) = coada1(nm) coada1(nn) = coada2(nn) coada1(nm) = coada2(nm) End If Next nm 'Debug.Print (coada1(nn)) Next nn For nn = 1 To et Debug.Print (coada1(nn)); Debug.Print (" : "); Next nn Debug.Print ("=") End Sub Title: Re: Calculating sound using fractal algorithms, what's out there ? Post by: djbarney on April 30, 2016, 07:14:00 PM If you clap hands into cubed box at irrational distance of walls comparative with length , the sound intervals are like as this example: 17 53.7544467966324 72.2786404500042 88.7157287525381 109.033087246637 125.47017554917 143.994369202542 Yep. Most things have some kind fractal dimension including music and reverb, echoes. But its not quite the idealised, designed fractals that can be created in eg. Mandelbulber 3D. That's an intentional fractal approach from the start whereas a lot of music is accidentally - if naturally - created fractals. please do!! would love to see this visualized! :) See attachement. That's from a drum machine/303 Rebirth track that I did. That part to me at least looks fractally. Somewhat like bird feathers. Quote and while were at it: good drumming is fractal: http://www.sciencemag.org/news/2015/06/secret-groovy-drumming-may-be-math fractals in bachs music: http://mathtourist.blogspot.de/2008/09/fractal-in-bachs-cello-suite.html fractal 1/f noise and other fractal occurences in music: https://courses.physics.illinois.edu/phys406/Lecture_Notes/P406POM_Lecture_Notes/P406POM_Lect15.pdf and, my personal research: techno, especially psytrance/goa is a an acoustic cantor set. that one has been sitting on my to-do list a while, I need to write that article.. ;) Yes, I was just looking at the Cantor set related to yin/yang & Hexagrams. Quote I also love that microphone or guitar feedback is the result of a recursive 'formula' and in my broad fractal view also is an acoustic fractal. A lot of music is very mathematical. I suppose a mathematician would say everything can be explained mathematically. I've found guides/scientific papers using modular synths to recreate formula’s using the various modules of the synth. The basic thing is this: an image of a fractal is a single frame. A single "slice" of time. Rendered pcm audio to make a "synth" sound, maybe in real time (but complex fractals I guess ain't gonna happen in realtime) has a time domain to it. I'm thinking of something like SimSynth 1.0 from way back in 1994 where you could define the sine waves and so forth using numbers that were entered and it would render a sound (EDIT: actually I think I was thinking of an old version of Cool Edit, think the same thing can be done in Audacity). I think PureData and other software can do this. So. A fractal SOUND is happening in a time domain. This sounds a bit like rendering an animation of a fractal in Mandelbulber or whatever application. So a fractal synth sound would not be akin to a single fractal image but with THE CHANGES that happen over time that we can see in fractal animations. Of course that all has to be translated in some rational way to the pcm data; frequencies, amplitude, envelope, stereo (surround sound ? but most people just have 2 speakers). I'm thinking of the basic raw material of a sound effect before sequencing, arpeggiation, etc. Title: Re: Calculating sound using fractal algorithms, what's out there ? Post by: Chillheimer on June 08, 2016, 10:31:17 PM I wish topics wouldn't fall out of sight so fast here.
I missed your answer completely, djbarney! fascinating, I totally agree, especially the "single slice of time" part.. thx for sharing your thoughts! Title: Re: Calculating sound using fractal algorithms, what's out there ? Post by: djbarney on June 10, 2016, 11:03:44 AM Cheers. A lot of the problem seems to be the conceptualisation in order to actually start writing algorithms in SuperCollider or whatever. I got diverted by other matters but am subconsciously thinking about this. Although I did get into a twitter exchange with BT about this - https://twitter.com/BC_Man/status/732673365467660288 ... still puzzling over why he needs to make fractal sounds "by hand" ? |