Species Evolution


Proof of concept for learning Neural networks and genetics algorithms developed in 2012. Simulates an environment of species that must survive and create a kind of “society mind” on a 2D world.

The neural network and genetics algorithm code is based on the amazing docs created by Mat Buckland.



  • Fully customizable using external XMLs.
  • Supports OpenMP for a very fast evolution on multi-core computers.
  • Evolution can be shown on the screen (slow but nice to see) or can be done on background (very fast but no visual output is provided).
  • On screen evolution stats.
  • You can easily add new species, by default there is only one (fully customizable) named Minzoo.
  • Available for Windows. Also compiles on Linux and MacOS X but make files are not provided.
  • Developed in C++ using CRM32Pro.


Last version: v1.00 (16-February-2012)
ChangeLog | Roadmap | LGPL license

32bits binaries for supported operating systems.
Source code
Source code and make files for VS2008+.
Previous releases
Get access to all releases repository.