Play it here -> https://iblunder.com/

Or challenge it on Lichess -> https://lichess.org/@/sapling-bot

You can browse all windows, linux or mac releases here

• Sapling makes use of hardware intrinsics to improve performance. Currently your CPU must support: Avx2, Bmi1, Bmi2, Popcnt, Sse. Most modern hardware shipped after 2013 should be supported.
• The releases come with a bundled version of the dotnet runtime, however if you want to run from source you'll need the dotnet 8 SDK installed.

cd ./scripts

// Windows
./build_windows_avx256.bat
./build_windows_avx512.bat

// Linux
./build_linux_avx256.sh
./build_linux_avx512.sh

// Osx
./build_osx_avx256.sh
./build_osx_avx512.sh

• quit : exit the program
• setoption name threads value 8 : sets the number of threads to use
• ucinewgame : initializes a new game
• position startpos : sets the engine to the starting chess position
• position startpos moves a2a3 a7a6 : sets the engine to the starting position then applies a set of moves
• position fen rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1 : sets the engine to the position given by the fen string
• d : outputs a diagram of the current position
• go perft 6 : Runs a pert test to a specific depth
• go depth 10 : Returns the best move after searching for the given depth
• go wtime 10000 btime 10000 winc 100 binc 100 : returns the best move after searching with the given time control
• go see a2a3 : [Dev] returns a the result of static exchange evaluation for a given move
• go eval : [Dev] returns the static evaluation of the current position
• datagen : [Dev] starts generating data in the bullet format, used when training a new NNUE network

After any changes to the engine a SPRT test must be ran to ensure that the changes have a positive effect.

There is a script sprt.bat which contains the command to run a cutechess-cli SPRT test. Ensure that you've configured CuteChess to point to both dev and base engines before hand, and also update the opening book + endgame table base to point to one on your system.

I'm in the process of training a (768x8->1024)x2-8 network starting from random weights using self play data generation and bullet trainer. Expect the engine to get much stronger as I improve the network. Check here to see the sequence of networks starting from scratch and the training logs.

• Chess Programming Wiki
• Talk Chess Forum
• Coding Adventure