Latest update: 2020/08/09: Stockfish 200807 9931db908a9b (+66 Elo to Stockfish 200731) (Ordo-calculation fixed to Stockfish 11 = 3554 Elo)
See the individual statistics of engine-results [You must be registered and logged in to see this link.]See the ORDO-rating of the archive-gamebase since 2020 [You must be registered and logged in to see this link.]Download the current gamebase [You must be registered and logged in to see this link.]Download the archive-gamebase since 2020 [You must be registered and logged in to see this link.] Program Elo + - Games Score Av.Op. Draws
1 SF 200807 9931db908a9b : 3653 8 8 7000 81.8 % 3374 33.9 %
2 BrainFish-2 200724 : 3631 8 8 7000 84.1 % 3330 31.0 %
3 Stockfish 200731 popc : 3587 8 8 7000 80.5 % 3330 36.2 %
4 Stockfish 11 200118 : 3554 5 5 16000 76.2 % 3335 39.3 %
5 Stockfish 10 181129 : 3510 5 5 15000 78.5 % 3274 37.7 %
6 Stockfish 9 180201 : 3461 9 9 5000 74.9 % 3259 41.7 %
7 Komodo 14 bmi2 : 3435 5 5 11000 53.0 % 3413 51.7 %
8 Houdini 6 pext : 3429 3 3 32000 63.1 % 3327 47.1 %
9 Komodo 13.3 bmi2 : 3424 6 6 8000 62.8 % 3328 49.9 %
10 Komodo 13.1 bmi2 : 3411 5 5 11000 62.0 % 3319 48.8 %
11 Komodo 12.3 bmi2 : 3398 7 7 7000 62.7 % 3300 49.4 %
12 Komodo 14 MCTS : 3325 7 7 5000 44.4 % 3371 53.4 %
13 Ethereal 12.25 pext : 3318 6 6 10000 36.2 % 3435 48.3 %
14 Slow Chess 2.2 popc : 3315 6 6 9000 32.3 % 3464 45.8 %
15 Ethereal 12.00 pext : 3303 6 6 9000 43.1 % 3357 50.8 %
16 Ethereal 11.75 pext : 3294 6 6 9000 39.3 % 3378 53.2 %
17 Fire 7.1 popc : 3285 3 3 32000 44.8 % 3331 51.9 %
18 Xiphos 0.6 bmi2 : 3284 4 4 19000 37.8 % 3387 50.3 %
19 Xiphos 0.5.6 bmi2 : 3274 6 6 7000 41.2 % 3342 54.6 %
20 Ethereal 11.53 pext : 3267 7 7 7000 42.2 % 3328 53.4 %
21 Komodo 12.3 MCTS : 3262 6 6 7000 42.7 % 3320 46.3 %
22 Ethereal 11.25 pext : 3257 7 7 6000 38.4 % 3348 51.0 %
23 rofChade 2.3 bmi2 : 3245 6 6 8000 27.5 % 3437 41.4 %
24 Booot 6.4 popc : 3230 7 7 6000 31.1 % 3380 46.5 %
25 Schooner 2.2 popc : 3227 7 7 6000 31.3 % 3376 50.3 %
26 Laser 1.7 bmi2 : 3204 7 7 6000 30.8 % 3357 45.8 %
27 Fizbo 2 bmi2 : 3199 9 9 5000 36.0 % 3311 39.0 %
28 Fritz 17 : 3198 7 7 6000 29.4 % 3363 44.2 %
29 Shredder 13 x64 : 3196 8 8 6000 31.9 % 3345 42.6 %
30 Defenchess 2.2 popc : 3191 8 8 5000 26.6 % 3380 41.8 %
31 Booot 6.3.1 popc : 3186 8 8 5000 34.0 % 3314 44.1 %
32 RubiChess 1.7.3 popc : 3180 7 7 6000 25.2 % 3384 39.5 %
33 Andscacs 0.95 popc : 3153 8 8 5000 23.1 % 3377 35.4 %
The version-numbers (180622 for example) of the engines are the date of the latest patch, which was included in the Stockfish sourcecode,
not the release-date of the engine-file. Especially the asmFish-engines are often released much later!!
Below you find a diagram of the progress of Stockfish in my tests since the end of 2018. 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). Latest Website-News (2020/08/09): Testrun of Stockfish 200807 avx2 with "default"-Framework nnue-net 9931db908a9b (= sv200806_1802 net) is finished. +66 Elo to Stockfish 200731. +99 Elo to Stockfish 11. Mention, that in my NNUE-net quick tests, this net is -24 Elo weaker than the best nets sv200805_1056 and sv200805_0633. So, I will repeat this testrun with the same binary (Stockfish 200807 avx2) and the net sv200805_0633 (0633 has less losses and more draws vs. Stockfish 200717 than 1056, and I think, that, when playing against much weaker engines in my full 7000 games testrun, this a little more solid play will give slightly better results - but the difference should be very small).Let's see, if there is a measureable Elo-gain. I dont believe in +24 Elo, but hopefully around +10 Elo or so... Result on Tuesday.