A Branch in the Path
  • I’ve recently revisited one of the first patches I started piecing together when I first discovered Audulus 2 years ago, but never managed to bring to completion.

    It’s basically two overtone multiplexers, with nodes from the first providing the root note for the other, and with the durations simply divided down a number of octaves from the overtone frequencies. I originally added a simple probability switch to climb up or down the overtone series with the output feeding back to the input so that the thing could run by itself, and a Euclidean sequencer to divide up the pulses.

    The original patch was intended not so much as a patch in itself but rather as a means of developing material for what eventually developed into A Branch in the Path – a piece for alto flute, bass clarinet and mobile phones: The main idea being that the audience navigate a grid of related overtone series via a website visited on their smartphones. Crossings from one series to another can be made at nodes that share a matching frequency – a little like changing lines on a subway system. I’ve mapped it out visually on my site: https://rudigermeyer.com/music/a-branch-in-the-path

    I originally didn’t manage to find a way of successfully programming all the possbile shifts from one ‘line’ to another and gave up on getting it automated in Audulus. However taking a look at JJThrash’s Audulus version of Conway’s Game of Life last year made me realize that I needn’t be shy of stringing together a large number of nodes, and gave me the courage to make another attempt at programming all the possible shifts in the overall overtone grid.

    My solution consists of a little probablility module for each overtone node. For the simplest nodes the module chooses between remaining on a note, moving up, or moving down within the same series. Other nodes include the possbility of shifting left or right to another overtone series. It’s perhaps not the most efficient approach, chewing up quite a lot of CPU power, but it does work, even though it does sometimes glitch when note changes and the envelope gate don’t exactly line up. I haven’t figured out a way to debug that yet. As from the programming itself I gave up on an entirely DRY approach and chose to manually enter a number of the values for each node-module instead of increasing the density of the connecting cables.

    I’m curious to know what other approaches others might suggest. Perhaps there are some clever equations that can simplify it all.

    The node path probabilities are collected in the ‘pathfinder’ sub-patch. The rest is a mix of some updated bits alongside the original core elements.
    2224 x 1668 - 1M
    RM - A Branch in the Path.audulus
  • Very nice patching!
  • @RobertSyrett Thanks! It can be fun to set a few of them going simultaneously with different settings (on different devices).
  • Wow this is really fascinating! Deserves deep study :)