soft: ableton live 7, reaktor 5, audiorealism everything
hardware: elektron machinedrum, nord micromodular, bass guitar, piano
Electronic Music other: Generating rhythms with the euclidean algorithm
Written October 24 2008 , Tags: rocketscience, rhythm, math, omg
This one is for you mathmusic heads out there (astroid i'm looking at you). I stumbled upon a paper by Godfried Toussaint called "The Euclidean Algorithm Generates Traditional Musical Rhythms", and played around a bit with it, and it is just terrific.
Here is the full post (with code and sound examples):
interesting. it's really funny, that one uses this intervals nearly „naturally“.
isn't it, it just sounds totally natural. Even more complicated things like 9 against 11 start pretty "obvious" (until you start to want to clap them out, of course .
this is kind of like what i spent most of my time at university trying to invent (im bad at motivating myself to do research so i have to (re)invent stuff myself)
evenly distributing stuff across time is defo something ive put a bit of thought to
shame i dont have the patience/attention span to do computer programming
I'd be very interested in hearing your theories/research bla, your music is a big source of inspiration for me, and all your crazy sounds
my work is pretty messy (it was for an art degree...)- spread across loads of squared paper
i cant remember if there was that much interesting stuff in it- it was certainly interesting for me to think about but i dont think i had much usuable results and conclusions
i think twice i posted here (but theyre gone now) my system for programming rhythms that i invented in the last 5 years- ive never quite perfected it- it just needs a bit of refining- its slow progress when you dont use a computer for this stuff!
i first thought of it because i was sending midi velocities from my drum machine to my effects unit and i wanted an even distribution of velocities across the pattern and i wanted to make sure the high velocities werent just on the 1/4 notes or 8th notes
it reduces the number of possible places for each beat but you still have to make some choices yourself but i think eventually ill get it to be a definite system that works by itself- obviously the drawback with this kind of thing is that the results are all pretty similar
ive got a recording of a liveset using my systems beats at link - its the live at bitjam one- but the drums are a bit too quiet and drowned in effects so you cant really hear it properly- but still theyre better than any beats i made without a system
right now i'm working on a sequencer that triggers samples from binary representations of tiny int numbers (1=play 0=stop). stole the concept from this guy but added some flare: link
mine has channel timing offsets for a shuffle/triple feel.
6 over 9 is great. Thanks for sharing! Thanks for the inspiration.
really cool. have a good gig!
What about note-length for sustained sounds like open hi-hats? If you want swing, can it only be applied as a groove quantize afterwards? How about accented notes? You know you could use those off-note events to trigger something else for some interesting syncopation possibilities.
i think that binary sequencer thingy on youtube would be much more interesting if instead the 1s represented the possibility of that drum sound for that step. then using something other than just randomness to decide if the drum actually got sounded. maybe velocity and the total number of active note ons could be used to weight the random chances of a drum hit when there is a 1.
i have a variation of the E() function from the paper working now but it produces unexpected sequences when k (the # of pulses) is greater than 1/2 n (the # of intervals). it just doesn't look equally distributed after that point. it would probably still sound interesting though.
i dig the descriptions of all the styles. great find! i'm going to work this function into some of my esoteric sound toys.
btw, in the example on your site, did you switch n and k from how the the paper described them? i'm a bit confused.
I keep getting broken link messages... I don't see anyone else having that problem so I guess it's just me.
My internets must be tangled.
I think it's the first article by Gottfried Toussaint here: link
But it's only an abstract! Do you have a link to the original article?
Signup to comment