Thrawn is a free UCI-compliant chess engine. I made this to learn about C++ and performance optmizations.

Thrawn does not include a graphical user interface (GUI) that is required to display a chessboard and to make it easy to input moves. These GUIs are developed independently from Thrawn and are available online. Some recommended GUIs include:

• CuteChess: https://github.com/cutechess/cutechess
• Arena GUI: http://www.playwitharena.de/
• Scid vs PC: https://scidvspc.sourceforge.net/

The rating of Thrawn is evaluated by CCRL

• Thrawn v2.2 ~3000 ELO (SF NNUE)
• Thrawn v2.0: ~2800 ELO (SF NNUE)
• Thrawn v1.1: ~1900 ELO

Version v2.1 and later: supports x64 and ARM chips. Can compile for linux, macos, or windows.

Older verions: only supports x64 and windows compilation.

Ensure you have a gcc compiler version 7.3 or later.

Compiling Thrawn v2.0 and newer:

git clone https://github.com/feftywacky/Thrawn.git
cd Thrawn
cd src

# release build
make # or mingw32-make

# debug build (only available for v3.0 or newer)
make BUILD=debug # or mingw32-make BUILD=debug

To clean the build:

make clean # or mingw32-make clean

Compiling Thrawn v1.1 and older:

git clone https://github.com/feftywacky/Thrawn.git
cd Thrawn
cd src
g++ -std=c++17 -Ofast -flto -o Thrawn *.cpp

Open source CLI tool to run matches between chess engines: https://github.com/Disservin/fastchess

• Thrawn v2.0 and newer uses NNUE
• Thrawn v1.1 uses handcrafted evaluation

• More move ordering heuristics (counter moves, history moves etc.)
• Train neural network

• NNUE for evaluation
• Bitboard data structure
• Multithreaded search (Lazy SMP)
• Transposition table
• Time allocation/Control
• UCI protocol

• https://www.chessprogramming.org/Main_Page
• https://github.com/bluefeversoft/vice
• https://github.com/official-stockfish/Stockfish
• https://github.com/mkd/gargantua

• https://web.archive.org/web/20071006042845/http://www.brucemo.com/compchess/programming/index.htm

• https://hxim.github.io/Stockfish-Evaluation-Guide/
  
• nnue probe framework: https://github.com/dshawul/nnue-probe
• nnue dataset https://tests.stockfishchess.org/nns