User Tools

Site Tools


This is an old revision of the document!

Generating 1/F and 1/F-squared note sequences

The fractal boundary traced out by the brownian motion of the spheres in Nel's CS 444 project is basically equivalent (I believe) to a sequence of random values generated by a 1/F (pink noise) or 1/F2 (brownian noise) process. Various powers of f (Frequency) can be used and collectively are called Fractional Noise (or Fractal Noise), a term coined by Mandelbrot [1].

Applied to SphereGen

The sequence of values that make up the paths traced out by each sphere's motion can be mapped to note values in a number of interesting ways.

Mapping ideas:

  • use the same Brownian process to generate notes centered initially around Middle C (= 60 midi) [2]
  • map the values to a pentatonic scale
  • map the values to microtones
  • use process to generate 1/F2 noise [2]

Playback and storage ideas:

  • Play the sounds along with the sphere's movement (more than 4 voices (heck, probably more than 1) = cacaphony)
  • Filter out note values which create various kinds of dissonances
  • Allow replay (from start or specified timestamp) while spheres are still moving
  • Save the note values as Midi along with Wave/MP3 recording of the 'music'
  • Timestamp saved information to make it possible to generate musical notation of a specific range of music

Conceptual Sketches with Minim

Applied to iPhone as Orchestra

The “iPhone as Orchestra” idea is based on wireless/Bluetooth connections between smart handheld devices in proximity to one another. The first idea is implementing a master/slave architecture where the first device to begin generating a sequence transmits to all others, which employ a Pd-inspired algorithm to generate harmonization. Again, we need to start simply. Some ideas:

  • Simply echo the original sequence, using some delay
  • Use delay or timestamps to do a “round” (like Row, Row, Row your Boat)
  • Use full-blown interactive algorithmic composition to generate the new voices

Other ensemble ideas using mobile devices are the PLORK, L2Ork, and iPhone Orchestras. See [4], [5] for video clips about Stanford's experimental ensembles MoPho.


[1] Algorithmic composition: paradigms of automated music generation By Gerhard Nierhaus, online at Google Books

[2] Algorithmic Composition: A Gentle Introduction to Music Composition Using Common LISP and Common Music by Mary Simoni, online at UMich Scholarly Monograph Series and following.

[3] Search Google using “algorithmic composition 1/f” for some choice links, especially Bruce Jacob

[4] Stanford's Mobile Phone Orchestra

[5] iPhone App by Smule: Ocarina (Stairway to Heaven)

prensf2010/spheregen.1274384353.txt.gz · Last modified: 2010/05/20 14:39 by nels_oscar