Code for scheme art contest for UC Berkeley's CS 61A, Fall 2022.

1st place in heavyweight division of the Scheme Art Contest.

See https://inst.eecs.berkeley.edu/~cs61a/fa22/proj/scheme_gallery/#frosty-s-existential-dread

Course instructions say to run pip3 install abstract-turtle

Undocumented dependecy is pillow and numpy: python3 -m pip install pillow numpy

To run a scheme file with the course's scheme interpreter, run

python3 scheme contest.scm --pillow-turtle --turtle-save-path output

This runs the file contest.scm and saves the drawing output to the file output.

The main code is in contest.scm. Configure the various variables defined to adjust the scene and render settings.

The script make_parallel.py generates scheme files so that multiple instances of the interpreter can run in parallel. When all processes are finished, it uses image magick to merge the images into the final result.

I found PyPy to speed up rendering significantly compared to the standard CPython interpreter. I can't use Cython or Numba or other compilers because the scheme interpreter from the course is obfuscated.