itch.io

Nice ! Maybe I should give the possibility of copying and setting the seed of a run, so people could replicate what others saw :)

Thank you ! Sadly the overall shape remains quite the same, but the colours are varied enough.

Introduction:

Set an aquarium of evolving critters on your desktop !

https://cedric-l-martin.itch.io/hydras

This cellular automaton is designed to do two things :

- Produce self-replicating structures. They do so by laying out patterns that resemble Sierpinsky triangles.

- Make structures that evolve over time. This happens in a deterministic fashion, as the rules for their evolution are embedded in the rules of their interactions. 

Warning:
I post this software now specifically because I lack the skills to optimize it and have given up on trying. Launch it if you dare put your eyes on a fps count below 10 ! I'm actually new to programming and cellular automata, and I will try to explain what I was trying to do as clearly as possible.

Cellular automata:

This cellular automaton is inspired by life-like automata, in which a grid contains cells that update their state at each time-step, based on the states of their neighbors. Depending on the rules governing these interactions between individual cells, complexe structures can emerge, that are stable, periodic or can move around. The most well-known of these is the eponymous Game of Life, by John Conway, probably because it is one of the simplest and yet it can yield a fantastic array of complex behaviours.

Details:

Including the empty one, there are 15 states in this cellular automaton.

Only empty cells can change states (In the software, filled cells fade away because I have integrated in them a timer in order to free space so that structures could continue to grow indefinitely).

To update its state, each cell checks its Moore's neighborhood, which designates the eight adjacent cells.

A cell changes its state based on that of its neighbors, but it won't do so only according to a set of pre-defined rules. Each cell contains not one state, but a string of states, acting like a genome. Then each unique neighborhood would point at a different position in the genome, and the character located at this position tells the state of the cell in the next time-step.