Generative music patch

    Mar 20 2013 | 11:12 pm
    I'm a university student, studying music technology. For my final project I have chose to build a generative music patch, inspired by Brian Eno's IOS apps Bloom and Scape.
    I would like to create something similar to scape using Max and jitter. I've been told such a thing is possible and will require a lot of hours work. I've been researching into techniques on how to do this and I have become very interested in Cellular Automata (CA) which, Brian has been known to use to create music. In terms of using CA in Max, I'm drawing a bit of a blank at the minute and any help on this would be most appreciated. I would like to use simple shapes with certain properties and be able to move them, change their colour etc.. All which, would have an effect on the sound event created.
    After researching this for a report I produced, I quickly discovered that I may have "bitten off more than I can chew." so to speak... I have a basic knowledge of Max/msp/jitter and i've built a granular synthesis patch in the past, but I am constantly teaching myself as much as I can about Max in general. I've come to a point now where I have to start building the patch but I am totally lost on where to start with it. Please forgive me for my lack of knowledge, I do learn quickly.
    I would be most grateful if someone could help point me in the right direction with this? Any help, anyone can give me will be a massive benefit
    Thanks Jack

    • Mar 20 2013 | 11:39 pm
      There's tons you can do...pretty endless...
      I got interested in Cellular Automata awhile back too and got Stephen Wolfram's book "A New Kind of Science", which talks a lot about it. I was able to implement the system shown here:
      using [matrixctrl] to select the "rules" and Jitter to display the results. Took a decent amount of fiddling, but was able to get it working just as Wolfram's rules described, which was great to see. One could readily expand that to more cells and see what happens. Or you can choose lines at random (or by a pattern or algorithm) from some of the larger overall patterns and use them for sequencing. Could be interesting, especially if you have the visual element going alongside it.
      Lots and lots of other ideas, I imagine there will be plenty of people chiming in here...if there's one thing Maxers seem to like, it's generative music... ;)
      Edit: I totally forgot...check out [jit.bfg]. Check it out *now*. :)
    • Mar 21 2013 | 12:08 am
      OK, one other thing, which I just tried and is awesome. Check out
      Get the plug-in so you can interact with the CDF demo files. It's totally worth it. Once you've played with this one, go to the main Demos page and start exploring. Holy moly...
      This goes for any other Maxers reading this too...
    • Mar 21 2013 | 1:12 am
      Just curious, where are you studying?
    • Mar 21 2013 | 10:56 am
      I've written some generative MIDI stuff for max for live. the system is called nwdlbots (pronounced noodlebots). The approach is based around fuzzy logic and probability. Please feel free to look under the bonnet/ hood.
      also check out Peter Elsea's Fuzzy Logic papers.:
    • Mar 21 2013 | 11:01 am
      +1 for the nwdlbots, saw you present them last year and got me inspired to create something similar!
    • Mar 22 2013 | 10:56 am
      @seejayjames - Thanks for the information. It is much appreciated! I will be sure to check the sources out you've given me right away. I am curious though, how do you begin to write such a programme? My thoughts and ideas get all jumbled up and I can never seem to sort through them. I struggle getting over the first hurdle, so to speak.
      @Legless Sheep. I'm studying Music Technology Bsc.
      @Richard Garrett - nwdlbots is an amazing piece of software. As i've said to seejayjames, how did you begin to write this programme and what was your starting point to do so? Thanks for the information as well, I will check those papers on fuzzy logic out. I appreciate the help very much.
      Best wishes JS
    • Mar 23 2013 | 2:47 pm
      I know you are studying music technology, just wondering what university you are at?
    • Mar 24 2013 | 11:38 am
      starting points:
      I got into generative music by playing with Koan Pro in the late nineties but it didn't quite do everything I wanted. I went to Dave Cope's Workshop for Algorithmic Computer Music in 2004, where Peter Elsea taught us about fuzzy logic. I then applied a fuzzy generative approach to other projects and when M4L came along, realised that I could package these things in a modular way that could interact with a conventional sequencer.
      - it's all process really.
      don't know if that helps.
    • Mar 26 2013 | 4:36 pm
      @Legless Sheep - Ah. Sorry. Leeds Metropolitan University.
      @Richard Garrett - I see. Yea, I guess it is all a learning process. I guess I'm looking for a learning source that doesn't really exist, maybe. I looks like i'm going to have to try and "learn by doing." But that has problems in itself. Thanks for your help though, I appreciate it. I'm reading unto fuzzy logic at the minute. Interesting stuff!
    • Mar 27 2013 | 3:44 pm
      Dude, im at leeds met too
      small world
    • Mar 29 2013 | 3:52 pm
      No way man! Small world indeed.