cellular_automata 0.8.0

A cellular automata library & playground

Cellular Automata

A cellular automata library / playground written in Dart. Happily accepting pull requests! ^_^

• Author: Jimmy Forrester-Fellowes (jimmyff)
• Github: https://github.com/jimmyff/cellular_automata
• Dart Pub: https://pub.dartlang.org/packages/cellular_automata
• Demos: http://jimmyff.github.io/cellular_automata

Features

• Simple system for writing optimised rules
• MCell rules compatible, current systems supported:
  • MCell Generations
• Generators to seed the simulation
• Map states to colours/characters for rendering
• Controls: pause, resume, stepBack, stepForward
• Stale simulation detection (repeating generations or patterns in activity)
• Utilises the package:logging for debug & info (view on pub)
• Decoupled renderers:
  • Canvas: (Web), fullscreen support
  • StageXL: (Web), WebGL, fullscreen support, (performance issues with big scenes)

Implemented rules/simulations

• Conway's Game of Life (source)
• Conway's Game of Life Simple (binary states) (source)
• Brian's Brain (source)
• Majority Vote (source)
• MCell Generations rule parser. See the demos including:
  • Star Wars
  • Rake
  • Star Wars
  • Sedi Mental
  • Frozen Sprials

Demos

See the demos/ folder for demos. You can try the hosted demos here.

Performance

Dartium performance is significantly faster than compiled performance at the moment. Also there are some generator defects that are yet to be resolved. To try and increase use the trust-type-annotations flag in your pubspec:

transformers:
- $dart2js:
    checked: false
    minify: true
    commandLineOptions: ['--trust-type-annotations', '--trust-primitives']

Roadmap

• Add more MCell rule parsers
• Add MCell rule database
• Add multiple states to 'Majority Vote'
• Add visual feedback to show processing of simulations - useful whilst building and debugging
• Support for entities to exist which would allow for simulations such as Langton's Ant
• Add a multiple states generator
• Add more tests
• Transitions for switching between two simulations
• Multiple simulations/rule-sets running in a single scene
• Generated color palettes
• Interactive mode (edit mode)
• Analysis tools (histograms for state counts)
• Add an LED Matrix renderer for raspberry pi using GPIO pins
• Flutter renderer
• CLI renderer
• Hexagonal grids
• Infinite world size support

Credits

Please see the credits file for a list of the authors that discovered the rules that have been implemented in this project.

Licence

Please see the cellular_automata license.