Recent Releases of alexander
alexander -
Alexander 5.0 Stockfish patch Feb 13, 2025 FutilityValue formula tweak A lot of refactorig and improvements in the handicap mode. The tests results are very good:
- C++
Published by amchess about 1 year ago
alexander - Alexander 4.1
Alexander 4.1 Stockfish patch Commits on Feb 4, 2025 What's New
- C++
Published by amchess about 1 year ago
alexander - Alexander 4.0
Alexander 4.0 Aligned with Stockfish Jan 12, 2025 Increase the depth margin
More info: Alexander Presentation
- C++
Published by amchess about 1 year ago
alexander - Aleander 3.1
Fix livebook Oct 31, 2024 Remove moveCountPruning in search.cpp
- C++
Published by amchess over 1 year ago
alexander - Alexander 2.0
Aligned with Stockfish 17. Fixed mcts. Fixed persisted learning and added the experience book option and an utility for reset (to be done for old experience files) and display content given a fen. Extended livebook to LiChess in addition to ChessDB. Removed opening variety and replaced with variety which is now a combo with three values: -off the default -standard without loss of elo -psychological more risky because with elo loss, but not as significant, useful for inviting the engine to a game with psychological/human sacrifices, i.e., strictly speaking, incorrect, but almost impossible to refute.
- C++
Published by amchess over 1 year ago
alexander - Alexander 1.3
Improved Opening Variety (Idea implemented in Hypnos) Extended LiveBook (also egtbs) Now, the ChessDB livebook is usable, again not in analysis or search for mate, even in endgame positions with at most 7 units on the board)
- C++
Published by amchess almost 2 years ago
alexander - Alexander 1.2
Fix montecarlo Speedup WinProbability
- C++
Published by amchess almost 2 years ago
alexander - Alexander 1.1
Nodestime option corrected futility pruning: parent node
- C++
Published by amchess almost 2 years ago
alexander - Alexander 1.0
Alexander 1.0 ShashChess without nnue. Unfortunately, the Stockfish community has decided to shelve the classic evaluation function, despite the fact that the engine is far beyond any human being. In addition, the classical evaluation function is the only one well understood by a human player. In fact, even an IM/GM struggles to understand a neural network game. Conversely, even for a GM, it makes no difference to have a 3100- or 3500-elo engine.... We believe that the future is precisely trying to decrypt the neural network to rewrite classic strategy books in a modern way. The first step, in our opinion, is precisely to try to compare the two evaluation functions. In addition, the classical evaluation function allows the implementation of a true handicap mode, based on a real thinking system and the concept of avatars to simulate not only the playing strength, but also the style of a player. Alexander should be the only reference engine for an OTB player below the IM title. An IM or GM can occasionally consult ShashChess to try to penetrate the secrets of NNUE. This is also why we decided to separate the two worlds. "Scientific research, if it does not improve people's lives, is a sterile fruit of hypertrophied egos"
- C++
Published by amchess almost 2 years ago