Significant portions of this project have been built on livestream to document the learning process.

The livestream series is in progress on YouTube, and it starts here: https://www.youtube.com/watch?v=1QotIA4_jb4

The current version of Wake is run using Python 3.x from the terminal.

• Clone the directory

• pip install -r requirements.txt (this installs the single dependency, numPy)

• cd wake

• python3 game.py

You will be presented with an output of the board in the shell:

Wake Engine [0.1.0] running using interface mode: [uci]
White to move:
8  ♜ ♞ ♝ ♛ ♚ ♝ ♞ ♜
7  ♟︎ ♟︎ ♟︎ ♟︎ ♟︎ ♟︎ ♟︎ ♟︎
6  ░ ░ ░ ░ ░ ░ ░ ░
5  ░ ░ ░ ░ ░ ░ ░ ░
4  ░ ░ ░ ░ ░ ░ ░ ░
3  ░ ░ ░ ░ ░ ░ ░ ░
2  ♙ ♙ ♙ ♙ ♙ ♙ ♙ ♙
1  ♖ ♘ ♗ ♕ ♔ ♗ ♘ ♖
   A B C D E F G H

To make a move, use UCI-style (short algebraic notation) move inputs, e.g: e2e4. This will update the state of the game.

Black to move:
8  ♜ ♞ ♝ ♛ ♚ ♝ ♞ ♜
7  ♟︎ ♟︎ ♟︎ ♟︎ ♟︎ ♟︎ ♟︎ ♟︎
6  ░ ░ ░ ░ ░ ░ ░ ░
5  ░ ░ ░ ░ ░ ░ ░ ░
4  ░ ░ ░ ░ ♙ ░ ░ ░
3  ░ ░ ░ ░ ░ ░ ░ ░
2  ♙ ♙ ♙ ♙ ░ ♙ ♙ ♙
1  ♖ ♘ ♗ ♕ ♔ ♗ ♘ ♖
   A B C D E F G H

The engine code is currently under development, using the UCI protocol to define valid inputs.

The only valid inputs are currently:

• Any valid move (e.g. g1f3 for a normal move, a7a8q for promotion)
• quit to kill the engine and terminal input processes.

The engine code will be open for general contribution from the after v1 is complete.

• Board representation testing

  • Board representation is in alpha. A full suite of unit tests still needs to be written to validate behavior.
  • Stalemate, 50 move rule, 3-fold repetition need to be implemented

• Search and Evaluation

  • Crude material evaluation is an attribute of the Position class
  • Proper evaluation metrics and implementation TBD

• Rewrite

  • The project is effectively an experiment in building a chess engine from scratch. Ultimately, the Python code serves as a prototype for a v1 rewrite in Rust.