An unstable sound can have a clear role

Hold a plain tone and slowly introduce instability. At first it may acquire a fine grain, then an uneven edge, then a roughness that competes with the pitch. The most useful position is not necessarily the most extreme. A small amount of instability can make a held note feel active while leaving room for phrasing. Once the pitch is nearly concealed, you may be working with a different musical object altogether.

Newfangled Audio describes Generate as a synthesizer built around chaotic generators, with development originating in an audio-rate double-pendulum model and continuing through modified dynamical systems. Its architecture also includes wavefolding and a low-pass gate. This article takes those ideas as a listening reference. It does not claim that another instrument reproduces Generate’s algorithms, nor that the word chaos alone describes a particular sound or guarantees a particular musical result.

Find three useful states

Choose one pitch and identify three settings: stable, textured and disrupted. Match their apparent loudness as closely as practical before comparing them. Otherwise the louder setting may seem more interesting simply because it occupies more attention. Move among the three slowly. Listen for what remains continuous and what changes abruptly. The point is to find a small vocabulary you can recognise while playing, not to catalogue every possible position of the control.

Use the stable state as a place to begin and return. Let the textured state sustain a phrase. Save disruption for a particular event: a pressure peak, the end of a breath or a transition into silence. The same timbre can sound purposeful or arbitrary depending on its placement. If every note starts at maximum complexity, the instrument has little further to travel when your performance asks for greater intensity.

Map pressure to a journey

An expressive controller can make this movement tangible. Map pressure to a restrained range of instability, with enough level compensation to keep the gesture understandable. Begin with one note and no effects. Increase pressure until the texture changes, then ease back until the stable centre returns. Practise arriving at the same intermediate state several times. A useful mapping allows you to revisit a region while preserving small variations within it.

Give another gesture a different task, such as changing spectral brightness or the amount of wavefolding. Avoid making every axis increase every parameter together. Independent dimensions let a quiet gesture become rough, or a bright gesture remain controlled. Start with two-note playing only after the single-note response feels legible. When several voices develop complex spectra, narrow the gesture ranges further so that differences between the voices remain audible.

Make repetition carry the gesture

A repeating or granular effect can preserve a brief expressive event and reorganise it over time. Hologram’s Microcosm is a hardware reference for processing through granular sampling, delay and looping; its own manual is the place to study its particular behaviour. In a separate browser instrument, begin by asking a simpler question: which part of this performed note should remain after the hand has left it? That question is useful regardless of the effect’s architecture.

Admit only a little processed signal at first. If adding a pitch-shifted shimmer, leave enough dry sound to hear the original gesture and keep the return from dominating each new phrase. Try playing less while the effect speaks. Then record an ending in which the source becomes simpler as the remaining trail grows more elaborate. The performance has handed activity to its own memory. Listen for the point at which that handover becomes clear, and give it time to finish.

Take it into a session

Try this.

Find stable, textured and disrupted states at one pitch. Perform a ninety-second phrase that reaches the disrupted state only twice. Add a restrained repeat effect for the final third, then stop playing and allow its trail to finish. Compare the dry and processed recordings at similar loudness.

Listen back

  • Which state preserves the clearest relationship between touch and sound?
  • Does instability change the phrase or only its surface?
  • What does the repeat remember that the dry performance has already left behind?
Open the studio →