Bluebetafish - UCI Chess engine in C++

Overview:

Bluebetafish is a chess engine with an approximate rating of 1900. It evaluates chess positions and determines the best moves using minimax search algorithm with alpha-beta pruning. It supports integration with UCI (Universal Chess Interface Protocol) compatible GUIs.

Try it:

https://chess-webapp-production-79f7.up.railway.app/index.html
If first link does not work then try this: https://chess-webapp-1.onrender.com/index.html.
- (The website takes ~1 minute to load for the first time, due to the hosting platform issue. Please wait for 1 minute while loading for the first time, then it will work fine)

Getting Started

Download and run the executable from the bin directory, or compile the source code using the make command. Or download any UCI protocol supported GUI and load the engine.

1. Use the engine via command line using UCI commands: [Each command is explained later]

position fen r3k2r/p1ppqpb1/bn2pnp1/3PN3/1p2P3/2N2Q1p/PPPBBPPP/R3K2R w KQkq -: will initialize the board with given position (input FEN string).

$ ./bluebetafish_64bit_linux

id name BlueBetaFish
uciok
position fen r3k2r/p1ppqpb1/bn2pnp1/3PN3/1p2P3/2N2Q1p/PPPBBPPP/R3K2R w KQkq -
Piece Board: 
+---+---+---+---+---+---+---+---+
| r |   |   |   | k |   |   | r |   8
+---+---+---+---+---+---+---+---+
| p |   | p | p | q | p | b |   |   7
+---+---+---+---+---+---+---+---+
| b | n |   |   | p | n | p |   |   6
+---+---+---+---+---+---+---+---+
|   |   |   | P | N |   |   |   |   5
+---+---+---+---+---+---+---+---+
|   | p |   |   | P |   |   |   |   4
+---+---+---+---+---+---+---+---+
|   |   | N |   |   | Q |   | p |   3
+---+---+---+---+---+---+---+---+
| P | P | P | B | B | P | P | P |   2
+---+---+---+---+---+---+---+---+
| R |   |   |   | K |   |   | R |   1
+---+---+---+---+---+---+---+---+---+---+
a   b   c   d   e   f   g   h   --------> File

CurrentPlayer : WHITE ,  EnPassant square : -1 ,  Castling Rights = KQkq

go movetime 5000: engine finds the best move for 5000 milliseconds.

go movetime 5000

info score cp 30 depth 1 nodes 1073 time 2 pv e2a6
info score cp 30 depth 2 nodes 1528 time 6 pv e2a6 b4c3
info score cp 15 depth 3 nodes 5730 time 9 pv e2a6 b4c3 b2c3
info score cp 15 depth 4 nodes 23335 time 27 pv e2a6 b4c3 b2c3 e6d5
info score cp 5 depth 5 nodes 85732 time 88 pv e2a6 b4c3 d2c3 e6d5 e4d5
info score cp 0 depth 6 nodes 209717 time 254 pv e2a6 e6d5 c3d5 b6d5 e4d5 e7e5
info score cp -20 depth 7 nodes 505761 time 568 pv e2a6 e6d5 c3d5 b6d5 a6b7 a8b8 e4d5
info score cp -20 depth 8 nodes 1203547 time 1286 pv e2a6 e6d5 c3d5 b6d5 a6b7 a8b8 e4d5 b8b7
info score cp -60 depth 9 nodes 4158122 time 4238 pv e2a6 e6d5 e5g4 f6g4 c3d5 b6d5 f3g4 h3g2 g4g2
bestmove e2a6

Total Nodes of iterative deepening : 6262784

Output (bestmove e2a6): within 5 seconds, the engine could search upto depth 9 and best move is e2a6.

2. Play against the engine using GUI apps:

UCI-Compatible GUIs: Use any GUI that supports the UCI protocol, such as Arena (for Windows), SCID (for both Windows and Linux), etc.

SCID_GUI_play_against_engine.mp4

Supported UCI commands

1. `position [fen | startpos] moves ....`: Sets up the board in memory with given position

Sets up the position described in FEN string on the internal board and plays the moves on the internal chess board.
Pass startpos as argument if you want to initialize the board to starting position.
moves is an optional argument. If it is not passed, no move is played after the given fen/startpos.

Examples:

`position startpos`: initializes the board with starting position.

Output: current position in memory is the starting position.

Piece Board: 
+---+---+---+---+---+---+---+---+
| r | n | b | q | k | b | n | r |   8
+---+---+---+---+---+---+---+---+
| p | p | p | p | p | p | p | p |   7
+---+---+---+---+---+---+---+---+
|   |   |   |   |   |   |   |   |   6
+---+---+---+---+---+---+---+---+
|   |   |   |   |   |   |   |   |   5
+---+---+---+---+---+---+---+---+
|   |   |   |   |   |   |   |   |   4
+---+---+---+---+---+---+---+---+
|   |   |   |   |   |   |   |   |   3
+---+---+---+---+---+---+---+---+
| P | P | P | P | P | P | P | P |   2
+---+---+---+---+---+---+---+---+
| R | N | B | Q | K | B | N | R |   1
+---+---+---+---+---+---+---+---+---+---+
a   b   c   d   e   f   g   h   --------> File

CurrentPlayer : WHITE ,  EnPassant square : -1 ,  Castling Rights = KQkq

position fen {FEN_STRING}: initializes the board with given position using FEN notation.

The corresponding FEN notation of the given position is: r3k2r/p1ppqpb1/bn2pnp1/3PN3/1p2P3/2N2Q1p/PPPBBPPP/R3K2R w KQkq -
Command: position fen r3k2r/p1ppqpb1/bn2pnp1/3PN3/1p2P3/2N2Q1p/PPPBBPPP/R3K2R w KQkq -

`position startpos moves e2e4 e7e5 g2g3`: initializes the board with starting position and plays the given moves.

position startpos moves e2e4 e7e5 g2g3:
Piece Board: 
+---+---+---+---+---+---+---+---+
| r | n | b | q | k | b | n | r |   8
+---+---+---+---+---+---+---+---+
| p | p | p | p |   | p | p | p |   7
+---+---+---+---+---+---+---+---+
|   |   |   |   |   |   |   |   |   6
+---+---+---+---+---+---+---+---+
|   |   |   |   | p |   |   |   |   5
+---+---+---+---+---+---+---+---+
|   |   |   |   | P |   |   |   |   4
+---+---+---+---+---+---+---+---+
|   |   |   |   |   |   | P |   |   3
+---+---+---+---+---+---+---+---+
| P | P | P | P |   | P |   | P |   2
+---+---+---+---+---+---+---+---+
| R | N | B | Q | K | B | N | R |   1
+---+---+---+---+---+---+---+---+---+---+
a   b   c   d   e   f   g   h   --------> File

CurrentPlayer : BLACK ,  EnPassant square : -1 ,  Castling Rights = KQkq

2. `go`: For calculating best moves

Starts calculating best move for the current position set up with the position command.
There are different arguments for this command.

`depth {DEPTH_VALUE}`

Searches the position upto depth DEPTH_VALUE

Example: go depth 6

Set up position:

position fen r3k2r/p1ppqpb1/bn2pnp1/3PN3/1p2P3/2N2Q1p/PPPBBPPP/R3K2R w KQkq -
Piece Board: 
+---+---+---+---+---+---+---+---+
| r |   |   |   | k |   |   | r |   8
+---+---+---+---+---+---+---+---+
| p |   | p | p | q | p | b |   |   7
+---+---+---+---+---+---+---+---+
| b | n |   |   | p | n | p |   |   6
+---+---+---+---+---+---+---+---+
|   |   |   | P | N |   |   |   |   5
+---+---+---+---+---+---+---+---+
|   | p |   |   | P |   |   |   |   4
+---+---+---+---+---+---+---+---+
|   |   | N |   |   | Q |   | p |   3
+---+---+---+---+---+---+---+---+
| P | P | P | B | B | P | P | P |   2
+---+---+---+---+---+---+---+---+
| R |   |   |   | K |   |   | R |   1
+---+---+---+---+---+---+---+---+---+---+
a   b   c   d   e   f   g   h   --------> File

CurrentPlayer : WHITE ,  EnPassant square : -1 ,  Castling Rights = KQkq

Find best move: -
Input: go depth 6
Output:

info score cp 30 depth 1 nodes 1073 time 1 pv e2a6
info score cp 30 depth 2 nodes 1528 time 3 pv e2a6 b4c3
info score cp 15 depth 3 nodes 5730 time 6 pv e2a6 b4c3 b2c3
info score cp 15 depth 4 nodes 23335 time 23 pv e2a6 b4c3 b2c3 e6d5
info score cp 5 depth 5 nodes 85732 time 90 pv e2a6 b4c3 d2c3 e6d5 e4d5
info score cp 0 depth 6 nodes 209717 time 252 pv e2a6 e6d5 c3d5 b6d5 e4d5 e7e5
bestmove e2a6

Total Nodes of iterative deepening : 295449

Output interpretation:
- bestmove {MOVE}
  - The engine has stopped searching and found the best move for the current position.
  - Example: bestmove e2a6 [Here best move is e2a6]
- info: Engine output with more informations.
  - depth : current depth upto which search has been done.
  - time : time (in millisecond) taken to complete the current DFS.
  - nodes : number of nodes searched
  - pv {MOVE1} {MOVE2} {MOVE3} ..... : the principal variation line (best move sequence for both players) found.
  - score cp {SCORE} : the score (evaluation) in centi-pawn unit from engine's POV.

`movetime {MOVETIME_VALUE}`

Searches the position for MOVE_TIME milliseconds

Example: go movetime 5000 (Searches for 5 seconds)
Output:

go movetime 5000
info score cp 30 depth 1 nodes 1073 time 1 pv e2a6
info score cp 30 depth 2 nodes 1528 time 3 pv e2a6 b4c3
info score cp 15 depth 3 nodes 5730 time 6 pv e2a6 b4c3 b2c3
info score cp 15 depth 4 nodes 23335 time 26 pv e2a6 b4c3 b2c3 e6d5
info score cp 5 depth 5 nodes 85732 time 71 pv e2a6 b4c3 d2c3 e6d5 e4d5
info score cp 0 depth 6 nodes 209717 time 215 pv e2a6 e6d5 c3d5 b6d5 e4d5 e7e5
info score cp -20 depth 7 nodes 505761 time 503 pv e2a6 e6d5 c3d5 b6d5 a6b7 a8b8 e4d5
info score cp -20 depth 8 nodes 1203547 time 1153 pv e2a6 e6d5 c3d5 b6d5 a6b7 a8b8 e4d5 b8b7
info score cp -60 depth 9 nodes 4158122 time 3873 pv e2a6 e6d5 e5g4 f6g4 c3d5 b6d5 f3g4 h3g2 g4g2
bestmove e2a6

Total Nodes of iterative deepening : 6670336

wtime {WTIME_VALUE} : Searches the position assuming white has WTIME_VALUE milliseconds remaining

`btime {BTIME_VALUE}` : Searches the position assuming black has `BTIME_VALUE` milliseconds remaining

Example: go wtime 180000 (Searches the position assuming white has 180 seconds)

info score cp 30 depth 1 nodes 1073 time 2 pv e2a6
info score cp 30 depth 2 nodes 1528 time 3 pv e2a6 b4c3
info score cp 15 depth 3 nodes 5730 time 6 pv e2a6 b4c3 b2c3
info score cp 15 depth 4 nodes 23335 time 23 pv e2a6 b4c3 b2c3 e6d5
info score cp 5 depth 5 nodes 85732 time 73 pv e2a6 b4c3 d2c3 e6d5 e4d5
info score cp 0 depth 6 nodes 209717 time 226 pv e2a6 e6d5 c3d5 b6d5 e4d5 e7e5
info score cp -20 depth 7 nodes 505761 time 525 pv e2a6 e6d5 c3d5 b6d5 a6b7 a8b8 e4d5
info score cp -20 depth 8 nodes 1203547 time 1255 pv e2a6 e6d5 c3d5 b6d5 a6b7 a8b8 e4d5 b8b7
info score cp -60 depth 9 nodes 4158122 time 4100 pv e2a6 e6d5 e5g4 f6g4 c3d5 b6d5 f3g4 h3g2 g4g2
bestmove e2a6

Total Nodes of iterative deepening : 7729152

winc : white increment per move in milliseconds
binc : black increment per move in milliseconds
movestogo {MOVESTOGO} : time increment begins after MOVESTOGO moves

Name		Name	Last commit message	Last commit date
Latest commit History 85 Commits
Bugs		Bugs
README-FILES		README-FILES
TestCases		TestCases
bin		bin
src		src
.gitignore		.gitignore
README.md		README.md
UCI_protocol.txt		UCI_protocol.txt

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Repository files navigation

Bluebetafish - UCI Chess engine in C++

Overview:

Try it:

Getting Started

1. Use the engine via command line using UCI commands: [Each command is explained later]

2. Play against the engine using GUI apps:

Supported UCI commands

1. `position [fen | startpos] moves ....`: Sets up the board in memory with given position

`position startpos`: initializes the board with starting position.

`position fen {FEN_STRING}`: initializes the board with given position using FEN notation.

`position startpos moves e2e4 e7e5 g2g3`: initializes the board with starting position and plays the given moves.

2. `go`: For calculating best moves

`depth {DEPTH_VALUE}`

`movetime {MOVETIME_VALUE}`

`wtime {WTIME_VALUE}` : Searches the position assuming white has `WTIME_VALUE` milliseconds remaining

`btime {BTIME_VALUE}` : Searches the position assuming black has `BTIME_VALUE` milliseconds remaining

`winc` : white increment per move in milliseconds

`binc` : black increment per move in milliseconds

`movestogo {MOVESTOGO}` : time increment begins after `MOVESTOGO` moves

3. `quit`: The process quits.

About

Releases 1

Packages

Languages

BlueBetaFish/chess-engine

Folders and files

Latest commit

History

Repository files navigation

Bluebetafish - UCI Chess engine in C++

Overview:

Try it:

Getting Started

1. Use the engine via command line using UCI commands: [Each command is explained later]

2. Play against the engine using GUI apps:

Supported UCI commands

1. position [fen | startpos] moves ....: Sets up the board in memory with given position

position startpos: initializes the board with starting position.

position fen {FEN_STRING}: initializes the board with given position using FEN notation.

position startpos moves e2e4 e7e5 g2g3: initializes the board with starting position and plays the given moves.

2. go: For calculating best moves

depth {DEPTH_VALUE}

movetime {MOVETIME_VALUE}

wtime {WTIME_VALUE} : Searches the position assuming white has WTIME_VALUE milliseconds remaining

btime {BTIME_VALUE} : Searches the position assuming black has BTIME_VALUE milliseconds remaining

winc : white increment per move in milliseconds

binc : black increment per move in milliseconds

movestogo {MOVESTOGO} : time increment begins after MOVESTOGO moves

3. quit: The process quits.

About

Topics

Resources

Stars

Watchers

Forks

Releases 1

Packages 0

Languages

1. `position [fen | startpos] moves ....`: Sets up the board in memory with given position

`position startpos`: initializes the board with starting position.

`position fen {FEN_STRING}`: initializes the board with given position using FEN notation.

`position startpos moves e2e4 e7e5 g2g3`: initializes the board with starting position and plays the given moves.

2. `go`: For calculating best moves

`depth {DEPTH_VALUE}`

`movetime {MOVETIME_VALUE}`

`wtime {WTIME_VALUE}` : Searches the position assuming white has `WTIME_VALUE` milliseconds remaining

`btime {BTIME_VALUE}` : Searches the position assuming black has `BTIME_VALUE` milliseconds remaining

`winc` : white increment per move in milliseconds

`binc` : black increment per move in milliseconds

`movestogo {MOVESTOGO}` : time increment begins after `MOVESTOGO` moves

3. `quit`: The process quits.

Packages