A friend posted a picture of their cat, showing its belly as it lay spread across a threshold and suggested that it looked like an audio waveform and essentially dared someone to create such an audio waveform. So I did.
The sound is essentially arbitrary, but it was a fun little challenge to figure a smooth way to make a sound with a waveform that matched (closely) a given image. The length is totally arbitrary, but I figured 35 seconds was plenty.
Trying out some code to repeat chunks of an input sound file based on characteristics of the chunks. The input file here is one I made a long time ago, banging on a sheet of steel with a contact mic on it.
I was wanting to pack superellipses tightly, so I worked out how to get two of them to be tangent to each other, and this allows me to create superellipses that are as large as possible and so fill space well. For this image, I turned off the requirement that the superellipses not be inside one another, and I think it has a nice look to it.
These are all 4-exponent superellipses (i.e., their shapes are the same as that of x4+y4=1).
