Nodal generative sequencer
  • Audulus users might also be interested in Nodal, a very interesting generative sequencer application for Mac and Windows (not iOS) by the Centre for Electronic Media Art at Monash University, in Melbourne, Australia. See

    My favourite Nodal examples are at

    I also mention Nodal here because, if Audulus supported MIDI output (hint, hint...), I think most, if not all, of the Nodal generative sequencing functionality could be replicated in Audulus, with many interesting extensions possible, such as generative timbral sequences alongside the pitch and rhythmic generation that Nodal allows.
  • @bennelong_bicyclist, cool! I had played with Nodal a while back. I suppose instead of using edge length to indicate how long a note is played, we could have a control on a "Note" node that determines how long the note is sustained. Is that what you were thinking?

    - Taylor
  • Yes, no need to keep the Nodal edge-length-is-note-duration paradigm. The ability to express note values as MIDI values, or even better as midicents (MIDI 61=6100 midicents) would be needed, but that should be possible with existing capabilities. Thus the generative graph (nodal network) would deal in midicents, and then either emit them as MIDI values using a new MIDI out widget, or convert the MIDI values to hertz for internal use in Audulus using the existing MIDI in widget. But I haven't fully thought it through. That said, Nodal is a fabulous and much under-appreciated sequencer, and it's limitations may also be strengths cf a more general purpose encironment like Max/MSP.
  • @bennelong_bicyclist, what if some sort of graph-based sequencer was implemented as a single node? I was thinking of having some sort of Markov-chain sequencer where each vertex in the graph is a note. The sequencer would have a clock input, so it could be externally clocked.
  • Yup, that could work if the graph inside the node could be edited. Or you can easily represent a Markov chain as a matrix of transition probabilities (unconnected vertices have probability of zero in the matrix), with the marginals of the matrix representing the current state and the next state values (or the previous state and the new state if you prefer).
  • The matrix representation would be easier to code up, but I'm not sure it makes a very good visualization of the Markov chain.
  • @bennelong_bicyclist this is so cool! thanks for sharing.
  • Hi all, know this is old so sorry about the noise. but check out , it is a nodal sequencer too . and it works with midi!