Can a physics simulator be used to generate sounds for music?
Does anyone use, or know of, a simulator for creating sounds based on simulating the vibrations of real objects?
I saw an art display called Tipping Point that created a sound-scape based on resonances of connected vessels of water that drained into each-other. It was a fascinating and complex sound that constantly change, giving moments of stillness, some of chaos and others almost lyrical.
That lead me to think about whether algorithmic music could be extended to a direct simulation of something like this.
Are there any physical simulators that are used for this sort of thing? At its simplest you could start with a church organ and adjust its length over time linked to others.
(3)3 Comments
Sorted by latest first Latest Oldest Best
This has been done for many years, probably more than 20. It's called physical modeling. Ableton Live Suite comes with at least two synth instruments based on physical modeling. Many digital pianos and other keyboard instruments meant to re-create real instruments, such as the Nord Electro line, use physical modeling along with sample banks. In the late 90s, a company called Line 6 developed circuit modeling based on the same concepts, and now circuit modeling is ubiquitous.
Right now, physical modeling and circuit modeling are probably two of the most widely used synthesis techniques across a range of musical products. Other popular methods are wavetable/sample based synthesis and analog subtractive synthesis.
People have made a good start in exploring these ideas, but there's plenty still to do. Comprehensive models that attempt to really simulate the behaviour of a vibrating, three dimensional system and then sample the sound produced at an audio rate are very expensive to run - The NESS project, for example, are still talking about the future possibility of making real-time models. Of course it's perfectly possible to make music using synthesizers that render offline, but mainstream products usually aim to be playable live.
When you hear about physical modelling being done in commercial products, this is often in the form of digital waveguides - it can yield impressive results, but it doesn't really attempt to reproduce the vibrating system in detail in the way that something like a finite element model does.
Advances in GPU-like processor technology bring the idea of detailed physical models being playable in real time closer. Who knows, when all the bitcoins have been mined, maybe we'll be able to make some new music!
Several; the book "Designing Sound"
Designing Sound teaches students and professional sound designers to
understand and create sound effects starting from nothing. Its thesis
is that any sound can be generated from first principles, guided by
analysis and synthesis. The text takes a practitioner’s perspective,
exploring the basic principles of making ordinary, everyday sounds
using an easily accessed free software. Readers use the Pure Data (Pd)
language to construct sound objects
uses Pure Data (Pd) and there is also Supercollider and probably other softwares in this department.
Terms of Use Privacy policy Contact About Cancellation policy © freshhoot.com2025 All Rights reserved.