/gamereport

Primary LanguagePython

WintrCat's Game Report

About:

A Python program by wintrcat that provides game reports for your chess games, without the need to pay for the Diamond premium on Chess.com. A locally installed instance of Stockfish 16 is used (Chess.com usually uses an older version than this...) to evaluate your moves, which the program then classifies and gives insights into.

Usage:

  • Run git clone https://github.com/wintrcat/gamereport.git in a terminal to download the program. You can also download as a ZIP if you do not have Git installed.
  • Download the PGN file of the game you want to analyse; you can select and download Chess.com games from your game archive.
  • Replace the game.pgn file in this directory with the one you downloaded; make sure to rename yours to game.pgn as well.
  • Run pip install -r requirements.txt in a terminal to install all of the necessary libraries.
  • Use the cd terminal command to enter the directory where this README.md file is.
  • Run python src/main.py in a terminal to run the program.

Command-Line Arguments:

  • -d / --depth Set the depth of the chess engine
  • -f / --file Set the analysis savefile to load (you can load .asys files)

Example command: python src/main.py --depth 22 --file save.asys

Engine Depth:

The engine used for this program is Stockfish 15.1, although other versions installed locally can be used for this program and for the most part will be fine unless a very old version is used. The depth of the engine refers to how many moves ahead of time that the engine accounts for when evaluating moves in the game. A higher depth means a more accurate game report, but also one that takes longer to produce. The default depth is 18 (Chess.com's default depth) which produces reports for 50-move games in 20-30 seconds, although this can heavily depend on the power of your CPU. If you want a more in-depth report, you can use a depth of 22+ but this will become exponentially slower. Games on Chess.com are analysed on their servers which are generally faster than a desktop computer, which is why it can sometimes take a minute or so for the program to analyse a game.

Disclaimers:

  • Stockfish source code is not included in this repository. The source code has not been modified and is available @ https://github.com/official-stockfish/Stockfish
  • pgn.py is not by me; copyright license for the library is included within its script file, src/pgn.py.