Implement arbitrary precision complex numbers about romanesgo HOT 3 OPEN

That's a lot of zoom 😆

from romanesgo.

Replaced the use of float64s with big.Floats. Performance is significantly impacted but that could easily be due to my poor initial implementation.

Currently, without even increasing precision, the rendering time of the test fractal on my machine jumps from ~235ms to ~1m35s. This is obviously hilariously unacceptable & won't be merged into the main branch unless performance can be maintained.

Preliminary result, test fractal rendered using a precision of 237 (octuple precision as per IEEE 754):

from romanesgo.

I'm stumped on how to make it more performant without just writing my own implementation of arbitrary precision floats from scratch. I'll leave the arbitrary-prec branch behind as an example of how not to approach the problem.

from romanesgo.