Stefan Pohl Computer Chess

private website for chessengine-tests


Latest Website-News (2018/04/24): Testrun of asmFish 9 finished (tested with Contempt=+20, like the default-setting in Stockfish 9). +21 Elo to Stockfish 9. Next testrun: asmBrainFish 9 (playing with Cerebellum_Light Release 169). Result not before next Wednesday.

 

BrainFish/Cerebellum successes in the Infinity Chess Engine Masters Tournaments !!

The 32nd Engine Masters Tournament was dominated by The_Machine, Wilfried Eilmes (Germany), who dedicated his victory to the Brainfish developer Thomas Zipproth. As he said, his success was mainly based on the Cerebellum-Libary concept, integrating book lines and engine play in an effective way. Read more (written by Arno Nickel): here

 

 

First preview-result (only 134 games played) of LC Zero in the long thinking-time tournament is online (and the games for download)... Check the "View Games"-section for some nice wins of LC Zero.

 

From now, the long thinking-time tournament is for testing of LC Zero, which plays a gauntlet vs. 5 opponents: Arminius 170101 (Elo: 2502), Greko 2018.2 (Elo: 2507), Gogobello 1.4 (Elo: 2534), Pro Deo 2.2 (Elo 2535), Cheese 1.9 (Elo 2543). Average opponent CEGT Elo (singlecore, 40m/20'): 2524

Checkout the long thinking-time-section for further information.

Testing of LC Zero is not easy. Because it uses the GPU for its thinking, it is not possible to run more than one game simultaneously on one PC. So, I decided to use the long thinking-time tournament (with higher speed of 12'+5'' (for more games)) for testing LC Zero...

 

My new SALC V5 openings and books are ready for download. Check out the "SALC openings"-section on this website for further information. Download SALC V5.02 here

 

Stay tuned.


Stockfish testing

 

Playing conditions:

 

Hardware: i7-6700HQ 2.6GHz Notebook (Skylake CPU), Windows 10 64bit, 8GB RAM

Fritzmark: singlecore: 5.3 / 2521 (all engines running on one core, only), average meganodes/s displayed by LittleBlitzerGUI: Houdini: 2.6 mn/s, Stockfish: 2.2 mn/s, Komodo: 2.0 mn/s

Hash: 512MB per engine

GUI: LittleBlitzerGUI (draw at 130 moves, resign at 400cp (for 4 moves))

Tablebases: None

Openings: HERT testset (by Thomas Zipproth) (download the file at the "Download & Links"-section or here)(I use a version of HERT, where the positions in the file are ordered in a different way - makes no difference for testing-results, dont be confused, when you download my gamebase-file and the game-sequence doesnt match with the sequence of your HERT-set...)

Ponder, Large Memory Pages & learning: Off

Thinking time: 180''+1000ms (= 3'+1'') per game/engine (average game-duration: around  7.5 minutes). One 5000 games-testrun takes about 7 days.The version-numbers of the Stockfish-development engines are the release-date, written backwards (year,month,day))(example: 170526 = May, 26, 2017). I use BrainFish-compiles (bmi2) by Thomas Zipproth (without using the Cerebellum-Library, BrainFish is identical to Stockfish and BrainFish-compiles are the fastest compiles of the Stockfish C++ code at the moment, around +10% faster than the abrok.eu-compiles and around 4% faster than the ultimaiq-compiles).

Download BrainFish (and the Cerebellum-Library): here

 

Each Stockfish-version plays 1000 games versus Komodo 11.3.1, Houdini 6, Fire 6.1, Shredder 13, Fizbo 2. All engines are running with default-settings, except: Move Overhead is set to 300ms, if an engine allows to do so.

To avoid distortions in the Ordo Elo-calculation, from now, only 2x Stockfish (latest official release + the latest version) and 1x asmFish and 1x Brainfish are stored in the gamebase (all older engine-versions games will be deleted, every time, when a new version was tested). Stockfish, asmFish and BrainFish older Elo-results can still be seen in the Elo-diagrams below. BrainFish plays always with the latest Cerebellum-Library of course, because otherwise BrainFish = Stockfish.

 

Latest update: 2018/04/24: asmFish 9

 

(Ordo-calculation fixed to Stockfish 9 = 3450 Elo)

 

See the individual statistics of engine-results here

Download the current gamebase here

Download the archive (all played games with HERT (165000 games)) here

See a ORDO-rating of the complete HERT-archive-base here

 

     Program                    Elo    +    -   Games   Score   Av.Op.  Draws

   1 BrainFish 180227 bmi2    : 3509    8    8  5000    76.8 %   3283   39.4 %
   2 asmBrainFish 171107      : 3491    8    8  5000    75.5 %   3276   43.5 %
   3 asmFish 9 bmi2           : 3471    8    8  5000    72.8 %   3285   43.3 % (new)
   4 Stockfish 180407 bmi2    : 3464    7    7  5000    72.0 %   3285   45.9 %
   5 Stockfish 9 180201       : 3450    7    7  6000    68.7 %   3301   48.5 %
   6 Houdini 6 pext           : 3430    5    5 11000    58.9 %   3357   53.9 %
   7 Komodo 11.3.1 bmi2       : 3392    6    6  7000    55.6 %   3346   52.5 %
   8 Komodo 11.2.2 x64        : 3382    6    6  8000    56.3 %   3331   49.0 %
   9 Fire 6.1 popc            : 3210    5    5 11000    29.5 %   3377   40.1 %
  10 Fizbo 2 bmi2             : 3201    6    6  9000    27.0 %   3389   35.2 %
  11 Shredder 13 x64          : 3193    5    5 11000    27.6 %   3379   38.9 %
  12 Fizbo 1.9 bmi2           : 3166    8    8  5000    28.9 %   3341   36.6 %

 

The 4 different Fishes in this Elo-list:

 

- Stockfish

- asmFish = Stockfish manually rewritten in assembler (look here)

- BrainFish = Stockfish playing with Cerebellum-Library by Thomas Zipproth (look here)

- asmBrainFish = asmFish playing with Cerebellum-Library

Below you find a diagram of the progress of Stockfish in my tests since the end of 2016

And below that diagram, the older diagrams.

 

You can save the diagrams (as a JPG-picture (in originial size)) on your PC with mouseclick (right button) and then choose "save image"...

The Elo-ratings of older Stockfish dev-versions in the Ordo-calculation can be a little different to the Elo-"dots" in the diagram, because the results/games of new Stockfish dev-versions - when getting part of the Ordo-calculation - can change the Elo-ratings of the opponent engines and that can change the Elo-ratings of older Stockfish dev-versions (in the Ordo-calculation / ratinglist, but not in the diagram, where all Elo-"dots" are the rating of one Stockfish dev-version at the moment, when the testrun of that Stockfish dev-version was finished).


Sie sind Besucher Nr.