For the last couple of weeks I’ve been working on a new piece for digitally-emulated feedback. It’s basically an attempt to use a laptop computer to autonomously generate a wide range of interesting sounds, live, within a relatively simple software setup. I hadn’t worked seriously with this method since 2008, when I played a few gigs using this method. Although the system produced a nice palette of sounds, there turned out to be no real way to change the basic nature of it from one moment to the next, which made the performance feel rather undifferentiated. This new piece is working on that last point.
In an attempt to use the same principles I’d worked with when making pieces out of analogue feedback oscillation, I tried making a similar type of music with more portable and robust equipment i.e. a laptop computer. I made some pieces in AudioMulch by patching together various sound processing contraptions into circuits. The circuits created feedback loops which generated a signal that was modulated by the interaction of the circuit’s constituent parts. Because they were intended for live performance, I put some of the critical control points in the system under my own control, while the computer handled other controls independently.
For this new piece I have relinquished all direct control (so far). It’s still being built in AudioMulch, using only the contraptions that come with the programme. (A dizzying array of freeware plugins can be found on the web.) It’s a new layout of feedback circuits but instead of monitoring control points within the circuits, I’m getting the computer to constantly micro-manage every possible parameter within each contraption in the circuit. To do this I’ve written a simple script which generates chance-determined MIDI control signals for nearly 500 parameters within the system.
It’s going to take a lot more refining, both in the circuit design and in the controlling script, but it’s all sounding extremely promising. I’ve temporarily uploaded some sample recordings on SoundCloud, where you can hear how the early versions of the piece developed from crude to less crude to the not-bad state in which it exists today. The latest version is embedded below.
These recordings are purely a demonstration of the software doing its thing. I haven’t intervened at any stage, tweaked any of the sounds or edited anything out. As a comparison, I’ve uploaded a recording of what the circuit sounds like after the control script has been turned off. I’ll worry about how to present this piece as a live performance later.