No tags yet.
I now have seeded randomness! It wasn't so hard after all - selecting a new Markov probability table preset dumps 62 chains of 16 note generations from [prob] into a [coll]. To ensure that the seeds remain constant when revisiting a project file, you can 'write' and 'read' the relevant coll file. This feels like a huge step forward; generations work a lot better together, and by using the same coll file for all melodic/harmonic functions, the audio generally works very well. This is a perfect example of how using [uzi], [t] and [zl] can help make a complicated algorithm with many steps work smoothly. Choosing a new preset effects the data in the correct way instantaneously. Currently this data can be saved and loaded, which can be used when I want to return to an Ableton project.