Since you’re here...

We hope you will consider supporting us today. We need your support to continue to exist, because good entries are more and more work time. Every reader contribution, however big or small, is so valuable. Support "Chess Engines Diary" even a small amount– and it only takes a minute. Thank you.
============================== My email: jotes@go2.pl



Chess engines: Heimdall 1.3.1

-

Heimdall - UCI chess engine written in Nim.
Author: Mattia Giambirtone

More info
Heimdall is available on Lichess under its old name (Nimfish), feel free to challenge it! I try to keep the engine running on there always up to date with the changes on the master branch

v.1.3.1:
Greetings! Today I'm releasing a bugfix version of Heimdall that fixes the pondering issues introduced in the previous release. Thanks to RBB (Discord) and @FischerRandomChess in the TCEC twitch chat for reporting the issue to me! :)

What changed
Honestly, not a lot strenght-wise, but still:

UCI option names and values are now case insensitive in accordance with the spec
Pondering is now fixed and behaves as expected
"Fail mid" is now implemented in reverse futility pruning
4-ply continuation history
A slightly improved neural network
A completely new SMP system that uses a custom thread pool and no longer wastes time spawning threads at each go command
The UCI command loop now also uses a persistent worker thread instead of creating one each time
Triple extensions (after singular search result)
Child node LMR extensions
TT initialization is now multithreaded: Heimdall will use as many workers as configured by Threads
TT prefetching in quiescence search
Support for the mimalloc allocator has been dropped
Use "fail-medium" in stand-pat cutoffs in quiescence search
Improvements and fixes to repetition detection
Many small fixes (OOB errors, safety, etc.)
Refactored many helpers to be branchless
Fix bugs related to EnableWeirdTCs
Fix a bug that caused crashes when parsing FENs without full-move numbers
PV nodes are reduced if the TT probe result suggests a cutoff
Development/beta builds will now feature the commit hash in the version string even on Windows
Strength
This release isn't that much stronger than the previous one (it is a bugfix, after all), but there have been some gainers still. I won't be publishing numbers with balanced books, as this kind of testing already takes a lot of time and I only want to bother with it for strength-significant releases, so these tests were run with the usual UHO_Lichess_4852_v1.epd book:

STC:
Elo   | 15.76 +- 8.81 (95%)
Conf  | 8.0+0.08s Threads=1 Hash=16MB
Games | N: 2008 W: 537 L: 446 D: 1025
Penta | [20, 207, 465, 286, 26]
https://chess.n9x.co/test/1130/

LTC:
Elo   | 6.25 +- 7.76 (95%)
Conf  | 40.0+0.40s Threads=1 Hash=64MB
Games | N: 2000 W: 447 L: 411 D: 1142
Penta | [4, 218, 527, 240, 11]
https://chess.n9x.co/test/1131/

Here also follows a test showing the scaling for the new SMP system (2v1 went from +55 to +68!):

Elo   | 68.32 +- 11.03 (95%)
SPRT  | 8.0+0.08s Threads=2 Hash=16MB
LLR   | 2.92 (-2.25, 2.89) [0.00, 3.00]
Games | N: 1128 W: 363 L: 144 D: 621
Penta | [2, 54, 256, 227, 25]


Heimdall 1.3.0 vs other engines:
Horsie 1.0.6 JA1.5/12-912 Games
Viridithas 16.0.01/8-68 Games
Sting Black Hole 96/6+66 Games
Deep HIARCS 15.4u4/6+26 Games
Ethereal 14.252/6-26 Games
Chess-System-Tal-22/6-26 Games
Caissa 1.212/6-26 Games
Alexandria 7.0.14 JA2/6-26 Games
Revenge 4.02/6-26 Games
Motor 0.8.0 JA2/6-26 Games
Lizard 11.2.01.5/6-36 Games
PlentyChess 4.0.11.5/6-36 Games
Clover 8.1 JA1.5/6-36 Games
HypnoS ++ 1.011.5/6-36 Games
Artemis 7TR1.5/6-36 Games
Dragon 3.31.5/6-36 Games
Velvet 8.1.1 JA1/6-46 Games
Obsidian 15.01 SE1/6-46 Games
Starzix 6.01/6-46 Games
Igel 3.6.0 JA1/6-46 Games
Stockfish dev-202502020.5/6-56 Games


Github: https://github.com/nocturn9x/heimdall

Comments

Post a Comment

Popular posts from this blog

New Strong Engines Test, by Chess Engines Diary, 2024.04.12

-
Image
  đź’ľ  552 (!) games from the tournament download   đź‘Ť @chessenginesdiary  Country - Poland, City -  Malbork đź•“ Time 3'+3" đź’»HP Pavilion i5-1035G1 8GB RAM đź–¬ GUI-Banksia All  CEDR  317.321 games download  (01.04.2024 - 3'+3")  HypnoS 030424, Yuli GM Pro 16, Stockfish 16.1 and ShashChess 35 . These four engines scored the same number of points and placed at the top of the table in this strong tournament. Tech table: Engine KN/move NPS dep/mov time/mov mov/game time/game fails Alexandria 6.1.0 3218 576429 30.2 5.6 56.7 316.2 Berserk 13 4779 841025 36.7 5.7 56.0 318.1 Brainlearn 28 2275 366977 31.4 6.2 43.5 269.4 Caissa 1.18 5492 923486 30.2 5.9 50.4 299.7 Clover 6.1.19 4803 854377 31.4 5.6 62.2 349.4 Cool Iris 12.10 2201 356730 27.6 6.2 51.2 315.5 CorChess 20240331 2253 372774 27.4 6.0 52.2 315.8 Crystal 8 3582 571994 20.3 6.3 47.7 298.8 Fire 9.3 5225 876565 22.0 6.0 55.5 331.1 Fisherov chess monk 1.2 34...
Read more

New chess opening book: M11.2 (bin and ctg)

-
Image
  We provide two opening books, in two different formats: bin and ctg. This is enough to be used in almost any chess program (GUI). Of course, every book may contain errors, but this one is based on a high rating. Happy testing! Book - CTG format Customers who wish to use the CTG format book will need a GUI capable of reading this format, ie. Chessbase GUIs.  Please save the book in your book folder. We strongly recommend that you "Optimize" the book in the Book options dialog so the tournament moves are chosen. The book is very wide even in tournament mode unlike most other books. A type of book format mainly used by chess programs from Chessbase , such as Fritz, Chess Base, Komodo, FatFritz, Houdini , etc. Book - BIN format (  Used by chess GUIs Arena, Banksia and Droidfish) Binary books have some important advantages, especially for computer chess engines: space-efficient, fast on accessing and searching, more useful information.  At the most simple form, they ar...
Read more

Dragon NNUE by Komodo Chess - it's free!

-
Image
    Dragon by Komodo Chess Larry Kaufman: We have released Dragon at komodochess.com. Dragon uses NNUE (Neural Network Updated Efficiently) technology, originally developed for the game of shogi. Komodo has a great deal of chess knowledge in its evaluation. Training an NNUE network based on this evaluation was both an advantage and a challenge, requiring experimentation with architectures and data generation of billions of positions. The reinforcement learning phase for Dragon is in its infancy, but is already showing great promise. Dragon is a huge strength improvement over 14.1, the last release, about 197 elo in MCTS mode and 189 elo in standard mode, at CCRL blitz time control on one thread based on direct matches. With four threads, the gains were 156 elo in MCTS and 170 in standard mode. The improvement is due to the embedded neural network providing much more accurate evaluation and also in some sense gaining an extra ply or so of search by “seeing” some tactics that a ...
Read more

Chess engines: ChessKiss 1.8

-
Image
ChessKiss - UCI engines,  Thanks to the Author: Abel Belzunces v.1.8 -Lazy eval improved -Fixed some issues in pawn evaluation -New now pawn penalty when ending -Some performance gains... -New recapture extension -Pawn cache improved -Evaluation performance improved a bit -New book -Seems book moves were not very random -Some bugs fixed here and there -Now rook and knight pairs are also evaluated -Now root nodes are evaluated differently -Time for move is calculated differently Chesskiss 1.8 download
Read more

New version chess engine: Lc0 BT4

-
Image
LcZero - CEDR Rating=3683 vBT4: Big transformer 4. New network architecture which builds off of BT3 by adding two types of auxiliary heads, future heads and categorical value heads. The categorical value heads predict a distribution over values of q rather than a WDL outcome distribution, and the future heads predict the moves that will be played over the next two plies. The hope is that these heads will give additional information to the net to improve training speed. We've also fixed half-precision training, so this model will be larger. BT4 training started in mid-October and is expected to take a few months. It has 15 layers with 1024 embedding size, 32 heads per layer, and 1536 emb size, for roughly a doubling in size over BT3.  Individual statistics:  Lc0 0.30.0 Stockfish 16 5/13 -3 13 Games Booot 7.2 6/10 +2 10 Games Critter 1.6a 8/9 +7 9 Games Dragon 3.2 4/8 +0 8 Games Stockfish 20230729 3.5/8 -1 8 Games ShashChess 33.2 3.5/8 -1 8 Games Fisherov chess monk 1...
Read more

Houdidit 6.03

-
Image
Michael Byrne: It has been alleged that the chess engine Houdini 6.03 was based on part, the open source chess engine Stockfish. If true, this will allows anyone who purchased the engine access to the source under Stockfish's GPL license. I purchased the engine multiple times over the years and I am now pleased to make the source available to everyone, as it is my opinion that the allegations are true and the evidence provided is overwhelming. In addition, you will find some custom compiled versions of the engine available to all ( allowed under Stockfish's GPL license). The engine here has been renamed to Houdidit , as in guess "Who did it?" and to keep it different from the engine Mr. Houdart was selling which in my opinion was in violation of Stockfish's GPL , i.e, selling Houdini without disclosing it was based, in part, on the open source Stockfish, nor releasing its source. A full discussion of this issue may be found here: http://talkchess.com/foru...
Read more