Skip to content

matteonaccarato/chess

Repository files navigation

CHESS ♟️

WHAT

♟️Developed chess game from scratch implementing an autonomous player using specific algorithms, such as MiniMax with Alpha‑beta pruning
💻 Three autonomous players ➡️ Random / MiniMax Base / MiniMax Advanced
📊 A python script has also been integrated to show bot performance data against each other.

LEARNING

☁️ Decision algorithms
🏎️ Fast prototyping

STACK

🧑‍💻 C++ / Unreal Engine 5.2 / Python

HOW TO PLAY ♟️

MOVE PIECES

  • When it is your turn, click on the piece you want to move. After that, possible moves of that piece are graphically shown
  • To confirm a move click on the tile you want to move on or the opponent's piece you want to capture

REPLAY

  • History of the moves done in the game is shown on the right part of the screen (in proper buttons)
  • To view previous moves, you can just click on the button of the move you want to view. The board configuration will be loaded as it was at that move.
  • REWIND: ONLY if you click on an opponent turn, you can active the rewind function by clicking one of your pieces. Then you can resume the game from this position

SAMPLE GAME

Sample game


UML

UML


(BOT) PERFORMANCE (data)

1) RANDOM (red) vs MINIMAX BASE (green)

Random  vs Minimax base

2) MINIMAX BASE (red) vs MINIMAX ADVANCED (green)

MiniMax base vs MiniMax peSTO

About

Algorithms applied to Chess ♟️

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages