This thread was the coolest so I took the fluoddity repo and put a playground at
g.mino.mobi/fluoddity
OAP has described the explorer’s impulse when poking around these guys, and I can feel that. So! I have given the option to save your config to atproto.
An open-ended playground for Fluoddity-style emergent behavior: GPU agents that read and write a vector trail field, steered by an evolvable Fourier black box. Clean-room reimplementation; credit to a...
In general I'm just throwing around "x=slider*10"s and "y=slider*.0001"s for many global constants to see what happens. When starting from zero, I typically look at a sample of 64 randomly initialized behaviors at a time with ~10k particles per behavior. I'm trying to get something like this:
