MoveAPiece
Offline chess against Stockfish, optionally with a real DGT Pegasus board
MoveAPiece is a native Android chess app that plays against a locally
bundled Stockfish engine or against Maia, a human-like opponent based on
a bundled neural network – fully offline, without accounts, without
network access, and without proprietary services.
Optionally, MoveAPiece can connect via Bluetooth LE to a physical
chess board – a DGT Pegasus or a Chessnut Air – so you can play against
Stockfish or Maia on a real board instead of on the screen.
Features:
* Local play: human vs. human, human vs. Stockfish, or human vs. Maia
* Maia: a human-like opponent that plays the moves a player of a chosen
rating (1100–1900) would typically play, instead of searching for the
strongest move. Nine pre-trained networks are bundled in the app and
run entirely on the device via ONNX Runtime (telemetry disabled) – no
network access
* Opening training: practice fixed lines from a built-in library of 20
well-known openings
* Adjustable Stockfish playing strength (UCI_LimitStrength / UCI_Elo,
1320–3190)
* Move history in algebraic notation, take back a move
* Evaluation display: optionally shows Stockfish's assessment of the
current position (can be toggled on/off)
* PGN import/export: export games via the share dialog or load a game
from a PGN file and continue playing
* Detection of check, checkmate, stalemate, and draws (repetition,
50-move rule, insufficient material)
* Move, capture, and check sound effects
* Portrait and landscape orientation
* Available in German, English, French, Spanish, Italian, and Dutch
* Support for physical boards (DGT Pegasus and Chessnut Air, BLE):
moves on the board are automatically detected and applied, and the
opponent's replies are shown via LEDs on the board. The Chessnut Air
additionally recognises each piece, so promotions need no dialog, a
position set up on the board can be taken over into the game, and
move sounds come from the board's speaker. Full touchscreen control
remains available at all times.
MoveAPiece is licensed under the GPLv3.
bundled Stockfish engine or against Maia, a human-like opponent based on
a bundled neural network – fully offline, without accounts, without
network access, and without proprietary services.
Optionally, MoveAPiece can connect via Bluetooth LE to a physical
chess board – a DGT Pegasus or a Chessnut Air – so you can play against
Stockfish or Maia on a real board instead of on the screen.
Features:
* Local play: human vs. human, human vs. Stockfish, or human vs. Maia
* Maia: a human-like opponent that plays the moves a player of a chosen
rating (1100–1900) would typically play, instead of searching for the
strongest move. Nine pre-trained networks are bundled in the app and
run entirely on the device via ONNX Runtime (telemetry disabled) – no
network access
* Opening training: practice fixed lines from a built-in library of 20
well-known openings
* Adjustable Stockfish playing strength (UCI_LimitStrength / UCI_Elo,
1320–3190)
* Move history in algebraic notation, take back a move
* Evaluation display: optionally shows Stockfish's assessment of the
current position (can be toggled on/off)
* PGN import/export: export games via the share dialog or load a game
from a PGN file and continue playing
* Detection of check, checkmate, stalemate, and draws (repetition,
50-move rule, insufficient material)
* Move, capture, and check sound effects
* Portrait and landscape orientation
* Available in German, English, French, Spanish, Italian, and Dutch
* Support for physical boards (DGT Pegasus and Chessnut Air, BLE):
moves on the board are automatically detected and applied, and the
opponent's replies are shown via LEDs on the board. The Chessnut Air
additionally recognises each piece, so promotions need no dialog, a
position set up on the board can be taken over into the game, and
move sounds come from the board's speaker. Full touchscreen control
remains available at all times.
MoveAPiece is licensed under the GPLv3.
版本
尽管下面提供了 APK 安装包的下载选项，但你应该注意，以这种方式安装将不会收到更新通知，这是一种不太安全的下载方式。 我们建议你安装使用 F-Droid 客户端。
下载 F-Droid-
x86_64该版本需要 Android 8.0 及以上版本。
此包由原始开发者构建并签名，并保证对应于此源代码 tarball。
-
arm64-v8a该版本需要 Android 8.0 及以上版本。
此包由原始开发者构建并签名，并保证对应于此源代码 tarball。
-
armeabi-v7a该版本需要 Android 8.0 及以上版本。
此包由原始开发者构建并签名，并保证对应于此源代码 tarball。