Since you’re here...

... we have a small favour to ask. More people, like you, are reading and supporting our blog: "Chess Engines Diary". And unlike many other sites and blogs, we made the choice to keep our articles open for all, regardless of where they live or what they can afford to pay.

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



Labels

Chess engines: Swordfish 15.1NNUE (Stockfish family)


Swordfish has a reasonable balance between strength and tactical abilities. It still has solid style of playing and achieves good results in test suites. It can be useful for analyzing sharp positions where regular Stockfish can overlook some tactics.
UCI-compatible graphical user interface (GUI) (e.g. XBoard with PolyGlot, Scid,
Cute Chess, eboard, Arena, Sigma Chess, Shredder, Chess Partner or Fritz) in order
to be used comfortably. Read the documentation for your GUI of choice for information
about how to use Stockfish engine family with it.
Swordfish - Stockfish family.

A note on classical evaluation versus NNUE evaluation
Both approaches assign a value to a position that is used in alpha-beta (PVS) search to find the best move. The classical evaluation computes this value as a function of various chess concepts, handcrafted by experts, tested and tuned using fishtest. The NNUE evaluation computes this value with a neural network based on basic inputs (e.g. piece positions only). The network is optimized and trained on the evaluations of millions of positions at moderate search depth.

The NNUE evaluation was first introduced in shogi, and ported to Stockfish afterward. It can be evaluated efficiently on CPUs, and exploits the fact that only parts of the neural network need to be updated after a typical chess move. [The nodchip repository](https://github.com/nodchip/Stockfish) provided the first  version of the needed tools to train and develop the NNUE networks. Today, more advanced training tools are available in 
[the nnue-pytorch repository](https://github.com/glinscott/nnue-pytorch/), 
while data generation tools are available in [a dedicated branch](https://github.com/official-stockfish/Stockfish/tree/tools).

On CPUs supporting modern vector instructions (avx2 and similar), the NNUE evaluation
results in much stronger playing strength, even if the nodes per second computed by the engine is somewhat lower (roughly 80% of nps is typical).

Notes:
1) the NNUE evaluation depends on the Stockfish binary and the network parameter file 
(see the EvalFile UCI option). Not every parameter file is compatible with a given
Stockfish binary, but the default value of the EvalFile UCI option is the name of a 
network that is guaranteed to be compatible with that binary.

2) to use the NNUE evaluation, the additional data file with neural network parameters
needs to be available. Normally, this file is already embedded in the binary or it can
be downloaded. The filename for the default (recommended) net can be found as the default
value of the `EvalFile` UCI option, with the format `nn-[SHA256 first 12 digits].nnue`



Rating CEDR=3481 (15'+10")
Rating CEDR=3449 (3'+3")

Individual statistics Swordfish 14.7 : 112 games (+ 40,= 72,-  0), 67.9 %
Cfish 290721                  :   2 (+  0,=  2,-  0), 50.0 %
Fritz 17                      :   2 (+  1,=  1,-  0), 75.0 %
GreKo 2021.08 NNUE            :   1 (+  1,=  0,-  0), 100.0 %
Bluefish 180921               :   2 (+  0,=  2,-  0), 50.0 %
Stockfish 14.1                :   2 (+  0,=  2,-  0), 50.0 %
SlowChess 2.8                 :   2 (+  0,=  2,-  0), 50.0 %
SugaR AI 2.50                 :   2 (+  0,=  2,-  0), 50.0 %
Fisherov 0.98i                :   1 (+  0,=  1,-  0), 50.0 %
Arasan 23.2                   :   2 (+  0,=  2,-  0), 50.0 %
Fisherov 0.98j                :   5 (+  0,=  5,-  0), 50.0 %
Kayra 1.1                     :   2 (+  0,=  2,-  0), 50.0 %
Dragon 2.6                    :   2 (+  0,=  2,-  0), 50.0 %
Lc0 0.28.2                    :   2 (+  1,=  1,-  0), 75.0 %
Bagatur 3.0                   :   5 (+  5,=  0,-  0), 100.0 %
Stockfish 020122              :   7 (+  0,=  7,-  0), 50.0 %
Wasp 5.20                     :   7 (+  6,=  1,-  0), 92.9 %
Berserk 8.5                   :   4 (+  1,=  3,-  0), 62.5 %
GreKo 2021.12                 :   7 (+  7,=  0,-  0), 100.0 %
Wasp 5.20 second compiles     :   7 (+  6,=  1,-  0), 92.9 %
Halogen 10.22                 :   7 (+  3,=  4,-  0), 71.4 %
Blue Marlin 14.7              :   9 (+  1,=  8,-  0), 55.6 %
CorChess 030122               :   3 (+  0,=  3,-  0), 50.0 %
Berserk 8.5.1                 :   3 (+  1,=  2,-  0), 66.7 %
Fritz 15                      :   1 (+  1,=  0,-  0), 100.0 %
Minic 3.18                    :   3 (+  2,=  1,-  0), 83.3 %
Francesca 0.31a               :   1 (+  0,=  1,-  0), 50.0 %
Stockfish 030122 Ivec         :   3 (+  0,=  3,-  0), 50.0 %
CorChess 100122               :   2 (+  0,=  2,-  0), 50.0 %
Black Marlin 3.0              :   2 (+  2,=  0,-  0), 100.0 %
Hannibal 1.7 x64              :   2 (+  2,=  0,-  0), 100.0 %
Raubfisch X48c2               :   2 (+  0,=  2,-  0), 50.0 %
ShashChess 20.2               :   2 (+  0,=  2,-  0), 50.0 %
BrainLearn 15.2               :   2 (+  0,=  2,-  0), 50.0 %
Fat Titz 130122               :   2 (+  0,=  2,-  0), 50.0 %
Eman 7.80                     :   2 (+  0,=  2,-  0), 50.0 %
Redfish 210921                :   2 (+  0,=  2,-  0), 50.0 %



Our social media:

Email Post
Labels:

Comments

Post a Comment

Popular Posts