
CA generative audio. Species independent composition.
OK I picked an A minor scale to map the pixels to frequency, the rest lies in the topology. Listen on this FREESOUND link.

OK I picked an A minor scale to map the pixels to frequency, the rest lies in the topology. Listen on this FREESOUND link.
First experiments using FFT techniques to generate audio based on generative geometric structure. This example is a rendering as a polyphonic heavy metal nocturne, played pizzicato and tuned to an equal temperament Lochrian scale using a root frequency of 73.4Hz (D2). Each pixel is an oscillator and the surrounding pixels define its harmonic content. Main image is a section of the generative function and brighter centre section shows part of the the sonification data. Multi channel capability is obtained by a slight offset of the data for each channel. For example left data is plus two pixels offset on the y and right data is minus two pixels. Centre channel has no y offset and the final audio is a mix, right = 20% centre plus 80% right. This facilitates easy construction of 64 even 128 channel sound spaces. A 5.1 192 kHz audio version of this example can be located on this link at FreeSound.
Early experiments, ca. 2011, to interface a dancer with a CA structure to create a audio visual using dTAG, dance Trajectory Audio Generation. First part of video demonstrates technique. Dancer is hooked into motion capture equipment to give XY coordinates. Left hand maps to audio volume, right hand maps to trajectory in green CA background from which audio is generated. Second part of video a bit more ambitious, the “Tai Chi Ballet”.
Say pure mathematical reality lies in natural numbers and when we cant handle very large natural numbers or very large aggregates of natural numbers we invent a thing called the real number line, The down side might be anything done with real numbers to give "answers" about natural numbers is bound to be an approximation, incomplete and paints a fuzzy picture of the underlying math phenomena.
Previous post showed a section of a mod 13 function. Using the same input matrix and a scaled rule set to mod 23 the output, this 16k image section, is the same but bigger. Nice example showing the rule set defines the function for every prime number. Using High NA EUV lithography scaling cells to 10nm and using a prime 433 or bigger the resulting function will fill a whole 300mm wafer with a single unique graphic.
Top inset is a section of the input matrix schematic. Red outline shows portion depicted in main image. 16k by 16k image size. Full size of image is about 900,000 pixels on the axis.
Rendering tera, > 10^12 pixels, scale math art to media, some techniques and sizes. Carpet weave with a 5mm warp, 25 square kilometres. Knitting with a 2 ply wool, 4 square kilometres. Textile printing with a 0.1mm pixel, 10000 square meters, Extreme ultra violet lithography, current chip manufacturing technology, about one square centimetre. Electron beam lithography, probably a couple of square millimetres. Shows what is in a smartphone!
Input schematic top left inset. Main image output function showing small portion of central input.
Analogue schematic, ANSCHEM, top left inset, to PROTO-IV process mask. A bit like a digital logic diagram to CMOS or NMOS. Image size 9k by 9k,,,,zoom in,
Fresh grass, dandelions and bluebell. Image size 16384 x 16384. Zoom in.
A section from this monster structure. Image width 8K height 24k.