1 unstable release
|0.1.0||Sep 22, 2021|
#7 in #chess-engine
21 downloads per month
A Universal Chess Interface (UCI) chess engine.
Blunders is currently a work in progress.
- Stoppable Negamax with alpha-beta pruning, Quiescence, and Iterative Deepening search
- Shared Transposition Table
- Move ordering
- Bitboard position representation
- Incremental Zobrist hashing
- Responsive UCI compatible I/O
- FEN string parsing
Building and running Blunders
Blunders supports 64-bit Windows 10 and Linux primarily, MacOS is tested only through CI. The following commands should work on both platforms.
Install the most recent stable rust compiler and cargo through rustup. Download or clone the Blunders repository and navigate into the root folder.
To build only, run the command
cargo build --release
To build and run, run the command
cargo run --release
The default location for the Blunders executable is
Blunders is a UCI compatible chess engine, and is most easily used from a chess GUI or CLI program instead of running it directly. Blunders uses the chess GUI Cute Chess during development and is known to work well within it.
To use Blunders directly, look at the UCI specification to find complete instructions on how to interact with Blunders in UCI mode. Eventually Blunders will get a non-standard set of commands to make it easy to use directly.
Blunders runtime settings
Hash x: an integer size in megabytes
xto set the size of the engine's hash table
Clear Hash: a button command telling the engine to clear its hash table, effectively forgetting its search history
Ponder bool: tells engine whether pondering is allowed or not. Allowing this means the engine may be allowed to search during an opponent's turn
Threads x: an integer
xtelling engine the maximum number of threads it may use to search. This is best set to the number of threads your computer cpu supports
Debug bool: tell engine to print debugging or extra information strings
Direct use through UCI examples
Change default settings, then quit:
blunders/target/release>./blunders setoption name Hash value 20 setoption name Ponder value false setoption name Threads value 4 setoption name Debug value true setoption name Clear Hash quit blunders/target/release>
Search the starting position to depth 3 to get info and bestmove output, then quit:
blunders/target/release>./blunders position startpos go depth 3 info depth 3 score cp +10 time 6 nodes 10000 nps 1666666 pv d2d4 d7d5 c2c4 bestmove d2d4 quit blunders/target/release>
Testing is done through
cargo. There are several commands that can be run to test all crates. Note that there are extra debug assertions so it may be worth it to test in both debug and release modes.
Run relatively quick tests:
cargo test --all or
cargo test --all --release
Run all tests:
cargo test --all -- --include-ignored or
cargo test --all --release -- --include-ignored
Blunders has some simple benchmarks that can be run with
cargo bench --all.
Checklist for 1.0
- Develop stable engine API
- Support single and multithreaded search
- Blocking and non-blocking
- Compile for WASM
- Add Blunders non-UCI commands for GUI-less play vs engine
- Clean library docs for
- Write User Starting Guide
This project is licensed under GNU GPL v3.0.
Copyright (C) 2021 Paulo Lemus This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with this program. If not, see <https://www.gnu.org/licenses/>.