The fastest algorithm is very very slow

It takes me about 42 minutes to render a 1280x640 image on my M1 Pro MBP.

I am thinking about the possibility of parallel computing.

The issue names no files, tests, or entry points; it only reports that rendering a 1280x640 image takes about 42 minutes on an M1 Pro and suggests parallel computing. First locate the image-rendering entry point and establish a reproducible baseline, then define a measurable rendering-time target before deciding what completion means.

Written by the indexing model from the issue text.