This is a personal project to explore a topic from university that really interested me, game tree search!
I have recently been playing a lot of Mancala and wished to know how good my strategies, and specifically openers, are. Surely I could have just googled this but that is no fun. I am writing this in C since it is my perferred langauge and I wish for the search to be fast. I will be following a three phase structure to it's implementation:
- Phase 1:
- console IO
- Working mancala game
- Phase 2:
- Multiple search strategies for best openings
- Heuristics, such as seeing if the remaining seeds are enough to sway the game
- Timings and simple data output / python graphing
- Multiple search strategies for best openings
- Phase 3:
- Simple raylib visualization of games
I will be adding some of my results here when they are more complete.
This project is going to be kept small and simple.
A basic Makefile
is provided with which you can simply run make
in the current directory to build the mancala
executable.