Stefan Pohl Computer Chess

Latest Website-News (2026/09/09): Testrun of Stockfish 19 finished: -2 Celo to Stockfish 260829, +28 Celo to Stockfish 18. Stockfish 18 is still included in the UHO-Top15 Ratinglist for comparison with Stockfish 19. Stockfish 18 will be deleted with the next ratinglist-update...

 

Dont forget to take a look at my EAS-Ratinglist (The world's first engine-ratinglist not measuring strength of engines, but engines's style of play). 

 

Next testrun: Gaia 4.3.0 - will it be strong enough to enter the UHO-Top15 Ratinglist?

 

Stay tuned.

UHO-Top15 Engines Ratinglist (+ regular testing of Stockfish Dev-versions)

 

 

Playing conditions:

 

The 15 strongest engines and the latest StockfishDev version are playing 1000 games vs. each opponent: 15000 games per engine are played. A RoundRobin Tournament with 120000 games overall !

 

Hardware: AMD Ryzen 9 7945HX 16-core (32 threads) notebook with 64GB RAM (since 2024/10/08), 2 identical machines. Windows 11 64bit

Speed: (singlethread, TurboBoost-mode switched off, chess starting position) Stockfish 17: 600 kn/s (when 26 games are running simultaneously)

Hash: 512MB per engine

GUI: Cutechess-cli (GUI ends game, when a 5-piece endgame is on the board, all other games are played until mate or draw by chess-rules (3fold, 50-moves, stalemate))

Tablebases: None for engines, 5 Syzygy for cutechess-cli

Openings: My UHO_2022_6mvs_+120_+129 openings are used (first 500 lines) (UHO 2022 openings are part of my Anti Draw Openings download-package). 

Ponder, Large Memory Pages & learning: Off

Thinking time: 3min+1sec per game/engine. One testrun (15000 games) takes around 80-85 hours (=3.5 days)

The version-numbers of the Stockfish engines are the date of the latest patch, which was included in the Stockfish sourcecode, not the release-date of the engine-file, written backwards (year,month,day))(example: 200807 = August, 7, 2020). The used SF compile is the AVX512-compile, which is the fastest on my AMD Ryzen CPU. SF binaries are taken from https://github.com/official-stockfish/Stockfish/releases (except the official SF-release versions, which are taken form the official Stockfish website).

Read the explanation, why I chose my testing-conditions, like I did here

 

Latest update: 2026/09/09: Stockfish 19 (-2 Celo to Stockfish 260829, +28 Celo to Stockfish 18)

 

(Ordo-calculation fixed to Stockfish 19 = 3905 Celo. Mention, there are no human or "realistic" Elo numbers in enginechess, since the engines reached superhuman strength, so I decided to use "Celo" (=Computer Elo) instead)

 

See the tournament cross-table as a picture (done by Fritz20 GUI) here

See the individual statistics of engine-results here

See the Engines Aggressiveness Score Ratinglist here

See the Gamepair-statistics of the UHO-Top15 Ratinglist here

See the Decision Time statistics of the UHO-Top15 Ratinglist here

See the full UHO Ratinglist (with EAS-Ratinglist): here

Download the current gamebase, the full-list gamebase and the archive here

See the old SPCC-Ratinglist (full list) and EAS-List from 2020 until 2023/08/28: here

 

(best Stockfish Celo: Stockfish 260829: 3907 Celo, latest full release (Stockfish 19) has 3905 Celo)

     Program                   Celo    +    - Games    Score   Av.Op. Draws

   1 Stockfish 260829        : 3907    4    4 16000    67.1%   3780   48.5%
   2 Stockfish 19 260905     : 3905    4    4 15000    67.6%   3774   48.7%
   3 Stockfish 18 260131     : 3873    4    4 15000    63.6%   3774   48.7%
   4 Reckless 260502 a512    : 3830    4    4 16000    56.2%   3785   49.3%
   5 Torch 4d a512           : 3811    4    4 16000    53.5%   3786   49.6%
   6 PlentyChess 8.0.0       : 3806    4    4 16000    52.8%   3786   48.9%
   7 Pawnocchio 2.0 a512     : 3783    4    4 16000    49.4%   3788   49.6%
   8 Coda 0.9.4 AI           : 3780    4    4 16000    49.0%   3788   49.2%
   9 Cinder 6.0 a512         : 3775    4    4 16000    48.3%   3788   48.9%
  10 Obsidian 250706 a512    : 3766    4    4 16000    47.0%   3789   49.3%
  11 Viridithas 20.0 a512    : 3761    4    4 16000    46.1%   3789   49.5%
  12 Alexandria 9.0 a512     : 3740    4    4 16000    43.0%   3791   49.1%
  13 Integral 250805 a512    : 3736    4    4 16000    42.4%   3791   49.7%
  14 Stormphrax 8.0.0        : 3733    4    4 16000    42.1%   3791   49.1%
  15 Quanticade 260810       : 3733    4    4 16000    42.0%   3791   49.8%
  16 Triumviratus 6.0 AI     : 3729    4    4 16000    41.5%   3791   51.4%
  17 Hobbes 3.0 a512         : 3721    4    4 16000    40.4%   3792   49.5%

 

Games        : 135000 (finished)
White Wins   : 67555 (50.0%)
Black Wins   : 821    (0.6%)
Draws        : 66624 (49.4%)

Below the UHO-Top15 Ratinglist, recalculated with Gamepairs (using my Gamepair Rescoring Tool V1.5), Ordo-calculation fixed to Stockfish 17.1 = 3854 Celo, realizing Vondele's (Stockfish Maintainer) idea: "Thinking uniquely in game pairs makes sense with the biased openings used these days. While pentanomial makes  sense it is a bit complicated so we could simplify and score game pairs only (not games) as W-L-D (a traditional score of 2-0, or 1.5-0.5 is just a W)."

Mention, the errorbar here is +/-5.5 Celo (=11 overall Error), compared to +/-4 Celo in the classical ratinglist above (=1.375 wider errorbar here). But the result-gaps are (at least) 2x up to 2.5x bigger (in average), calculating with gamepairs. So, the statistical reliability of the results is much better in the Gamepair-ratinglist, compared to the classical ratinglist...

CFS means Chance For Superiority in percent, calculated by ORDO.

See the head-to-head gamepair-statistics of the UHO-Top15 Ratinglist here

   # PLAYER                  :    Celo  Error   Pairs     W     D     L   (%)  CFS(%)
   1 Stockfish 260829        :    3906     10    8000  5569  2220   211  83.5      60
   2 Stockfish 19 260905     :    3905   ----    7500  5364  1975   161  84.7     100
   3 Stockfish 18 260131     :    3806     11    7500  4428  2661   411  76.8     100
   4 Reckless 260502 a512    :    3701     10    8000  3155  3631  1214  62.1     100
   5 Torch 4d a512           :    3659     10    8000  2699  3698  1603  56.9     100
   6 PlentyChess 8.0.0       :    3647     10    8000  2559  3734  1707  55.3     100
   7 Pawnocchio 2.0 a512     :    3597     10    8000  2018  3756  2226  48.7      94
   8 Coda 0.9.4 AI           :    3591     10    8000  1954  3755  2291  47.9      99
   9 Cinder 6.0 a512         :    3582     10    8000  1833  3795  2372  46.6     100
  10 Obsidian 250706 a512    :    3562     10    8000  1646  3752  2602  44.0     100
  11 Viridithas 20.0 a512    :    3549     10    8000  1488  3776  2736  42.2     100
  12 Alexandria 9.0 a512     :    3503     10    8000  1101  3598  3301  36.3      98
  13 Integral 250805 a512    :    3494     10    8000  1046  3529  3425  35.1      80
  14 Stormphrax 8.0.0        :    3491     10    8000  1019  3509  3472  34.7      83
  15 Quanticade 260810       :    3487     10    8000   991  3491  3518  34.2      93
  16 Triumviratus 6.0 AI     :    3481     10    8000   986  3382  3632  33.5     100
  17 Hobbes 3.0 a512         :    3464     10    8000   815  3396  3789  31.4     ---

------------------------------------------------------------------- 
--- Number of all Gamepairs          : 67500 
--- Number of drawn Gamepairs overall: 28829 (= 42.71%) 
--- Number of 1:1 drawn Gamepairs    : 14570 (= 21.58%) 
--- Number of 2-draws drawn Gamepairs: 14259 (= 21.12%) 
------------------------------------------------------------------- 

Aborted testruns, because the tested engine was too weak to enter the UHO-Top15 Ratinglist:

Icarus 1.1, Heimdall 1.5, Renegade 1.2.0, Igel 3.7.0, Minic 3.41, Booot 7.3, Black Marlin 9.0, Equisetum 1.0, Clarity 8, PZChessBot 7.1, Prune 4.0.1

Below you find a diagram of the progress of Stockfish in my tests since February 2026. 

And below that diagram, the older diagrams.

 

You can save the diagrams on your PC with mouseclick (right button) and then choose "save image"...