i remember the inception of warhammer/wh40k and the publication of cosmic trigger, read some elric, rolled some polyhedrons, and have a sort of inverse star of chaos tattoo jotted from my idle youthful moments at arcosanti. my understanding is that "chaos" is about the threshold concerning what we deign to keep track of and what we do not, what is within our consideration of "predictable", which is really relative. i do not like to spin any unnecessary mystique about the subject and prefer to say "procedural" really i think we agree in our thesis.. my journey through cheese erm i mean audio dsp has involved the sort of monolithic references envaluating What We Do and What We Do Not Do. "we" basically don't do, or, mostly haven't done to date, anything that doesn't lend itself easily to antialiasing. clever methods like DSF are What We Want.
the result is that We have barely begun to simply *do things with numbers* because people want brilliance like 100+ pages for a filter by andrew simper. fair enough, reasonable. but we're not *doing things* because (vitd) "if only the brightest birds sang in the forest".
to contrast,
http://xoxos.net/temp/black1.mp3
i've mucked around with 3d forms for a bit and generally found them to be either too noisy or not noisy enough. mostly due to my limited thinking and approach. putting newer ideas together, the above.. many simple "3d rotation" arrangements will produce, say 9 partials in odd, mirrored arrangements. not worth publishing. and of limited benefit, eg. generally, the way i've done things at least, doesn't produce anything wildly dissimilar to 3 or 4 op FM.
i'm liking the example linked above, because it does produce a wide range of timbres (two rate params, two angles, a few scalars). the audio example is with constant frequency components, only modulating angles and scalars. it sounds like FM, pretty much. most things seem to.. and, it's already complex enough to be less than meaningful and intuitive. this arrangement is a little better than some, but i'm calling it "black" because it's hardly worth relating the very arbitrary form inside.
going with what i began the post with, if simple minded, ignorant people who don't care about AA practiced dsp more, there would be some really wild developments (check out my history of "physical models.."). maybe we can get there now that computers are faster and oversampling isn't such an issue. but the mystique we weave around audio dsp to sell it, is not constructive to our synergetic potential as a culture, by excluding so many willing participants.
also for developers, something i saw earlier today. had another dig around and encountered more caches of attractors (halvorsen, hadley, thomas.. "nose hoover" what a name) and this pdf which i haven't digested about a sort of intersecting rossler form -
http://www2.gsu.edu/~matals/lp/lp_paper ... ocus_s.pdf
(dreams of a platform where all the keys work) xoxos 64 bit is probably going to require (begin kirk voice) some.. sort.. of.. meaningful... public.. acknowledgement....
