Skip to content
On this page

All of the information below has been generated from the results of tests performed on the Fishtest framework. All data and graphs can be downloaded here. Please get involved if you have questions or suggestions!

Current Testing Criteria

  • 1 Thread 60 seconds + 0.6 seconds for 60,000 games (21/11/2019 - current)
  • 8 Threads 30 seconds + 0.3 seconds for 40,000 games (13/12/2018 - current)
  • UHO_4060_v2.epd opening book (29/06/2023 - current)
Previous Testing Criteria
  • 1 Thread 60 seconds + 0.6 seconds for 40,000 games (02/01/2016 - 21/11/2019)
  • 1 Thread 60 seconds + 0.5 seconds for 40,000 games (13/10/2013 - 02/01/2016)
  • 1 Thread 60 seconds + 0.5 seconds for 20,000 games (04/03/2013 - 13/10/2013)
  • 8moves_GM.pgn opening book (10/04/13 - 01/11/13)
  • varied.bin opening book (04/03/13 - 07/04/13)
  • 8moves_v3.pgn opening book (09/11/2013 - 29/06/2023)
  • UHO_XXL_+0.90_+1.19.epd opening book (17/04/2022 - 29/06/2023)

Current Development

DateVersion1 Thread8 Threads1 Thread (UHO)
2022‑05‑14master vs Stockfish 15
Bench: 6481017
SE depth scaling using the previous depth
[differences] 9
Elo: 4.54 ±0.8
Ptnml: 18, 1850, 25507, 2580, 45
nElo: 16.29 ±2.8
PairsRatio: 1.41
[raw statistics]
Elo: 3.12 ±0.8
Ptnml: 5, 803, 18029, 1154, 9
nElo: 13.92 ±3.4
PairsRatio: 1.44
[raw statistics]
2022‑06‑16master vs Stockfish 15
Bench: 5845802
Simplify away condition in ttSave in probCut
[differences] 25
Elo: 5.72 ±0.8
Ptnml: 31, 1838, 25288, 2799, 44
nElo: 19.99 ±2.8
PairsRatio: 1.52
[raw statistics]
2022‑07‑13master vs Stockfish 15
Bench: 5905619
Update default net to nn-ad9b42354671.nnue
[differences] 34
Elo: 5.70 ±0.8
Ptnml: 42, 1867, 25207, 2832, 52
nElo: 19.67 ±2.8
PairsRatio: 1.51
[raw statistics]
Elo: 4.86 ±0.8
Ptnml: 3, 734, 17975, 1276, 12
nElo: 21.46 ±3.3
PairsRatio: 1.75
[raw statistics]
2022‑08‑12master vs Stockfish 15
Bench: 5868987
Remove an unneeded randomization of evals
[differences] 43
Elo: 7.18 ±0.8
Ptnml: 48, 1784, 25112, 2993, 63
nElo: 24.44 ±2.8
PairsRatio: 1.67
[raw statistics]
Elo: 4.60 ±0.8
Ptnml: 5, 737, 17987, 1265, 6
nElo: 20.43 ±3.4
PairsRatio: 1.71
[raw statistics]
Elo: 14.61 ±1.3
Ptnml: 149, 5043, 17073, 7608, 127
nElo: 30.73 ±2.8
PairsRatio: 1.49
[raw statistics]
2022‑09‑07master vs Stockfish 15
Bench: 5609606
VLTC tuning
[differences] 52
Elo: 6.87 ±0.8
Ptnml: 40, 1796, 25154, 2958, 52
nElo: 23.61 ±2.8
PairsRatio: 1.64
[raw statistics]
Elo: 5.45 ±0.8
Ptnml: 3, 732, 17913, 1339, 13
nElo: 23.68 ±3.3
PairsRatio: 1.84
[raw statistics]
2022‑10‑05master vs Stockfish 15
Bench: 4114228
Revert "Mix alpha and statScore for reduction"
[differences] 66
Elo: 6.86 ±0.8
Ptnml: 21, 1580, 25632, 2727, 40
nElo: 25.04 ±2.7
PairsRatio: 1.73
[raw statistics]
Elo: 5.98 ±0.8
Ptnml: 6, 643, 18015, 1329, 7
nElo: 26.72 ±3.3
PairsRatio: 2.06
[raw statistics]
2022‑10‑30master vs Stockfish 15
Bench: 4271738
Adjust reduction less at medium depths
[differences] 81
Elo: 8.52 ±0.8
Ptnml: 23, 1469, 25573, 2884, 51
nElo: 30.84 ±2.7
PairsRatio: 1.97
[raw statistics]
Elo: 5.91 ±0.7
Ptnml: 6, 589, 18129, 1271, 5
nElo: 27.24 ±3.3
PairsRatio: 2.14
[raw statistics]
Elo: 17.17 ±1.8
Ptnml: 38, 2430, 8572, 3933, 27
nElo: 36.91 ±4.0
PairsRatio: 1.60
[raw statistics]
2022‑12‑02master vs Stockfish 15
Bench: 3467381
Fix bestThread selection
[differences] 97
Elo: 7.46 ±0.7
Ptnml: 21, 1389, 25902, 2657, 31
nElo: 28.19 ±2.7
PairsRatio: 1.91
[raw statistics]
Elo: 5.97 ±0.7
Ptnml: 3, 573, 18164, 1254, 6
nElo: 27.83 ±3.3
PairsRatio: 2.19
[raw statistics]
Elo: 17.04 ±1.3
Ptnml: 73, 4844, 17208, 7820, 55
nElo: 36.73 ±2.8
PairsRatio: 1.60
[raw statistics]
2022‑12‑04Stockfish 15.1 [[https://stockfishchess.org/images/logo/icon_128x128.png|width=20px]]
Bench: 3467381
[differences] 98
2022‑12‑19master vs Stockfish 15
Bench: 3727508
Sometimes do a reduced search if LMR is skipped
[differences] 119
Elo: 8.55 ±0.8
Ptnml: 16, 1420, 25675, 2850, 39
nElo: 31.50 ±2.7
PairsRatio: 2.01
[raw statistics]
2023‑01‑01master vs Stockfish 15
Bench: 4015511
Update default net to nn-60fa44e376d9.nnue
[differences] 126
Elo: 10.00 ±0.8
Ptnml: 15, 1301, 25667, 2976, 41
nElo: 36.92 ±2.7
PairsRatio: 2.29
[raw statistics]
Elo: 7.10 ±0.7
Ptnml: 6, 529, 18118, 1336, 11
nElo: 32.56 ±3.2
PairsRatio: 2.52
[raw statistics]
Elo: 23.55 ±1.3
Ptnml: 59, 4354, 17132, 8377, 78
nElo: 51.05 ±2.8
PairsRatio: 1.92
[raw statistics]
2023‑01‑23master vs Stockfish 15
Bench: 3941848
Update default net to nn-bc24c101ada0.nnue
[differences] 143
Elo: 11.37 ±0.7
Ptnml: 9, 1127, 25786, 3048, 30
nElo: 42.95 ±2.7
PairsRatio: 2.71
[raw statistics]
Elo: 9.09 ±0.8
Ptnml: 2, 484, 17984, 1526, 4
nElo: 40.84 ±3.2
PairsRatio: 3.15
[raw statistics]
Elo: 29.49 ±1.3
Ptnml: 43, 3893, 17099, 8870, 95
nElo: 64.58 ±2.9
PairsRatio: 2.28
[raw statistics]
2023‑02‑09master vs Stockfish 15
Bench: 3841998
Update default net to nn-1337b1adec5b.nnue
[differences] 162
Elo: 12.76 ±0.8
Ptnml: 9, 1089, 25635, 3224, 43
nElo: 47.33 ±2.6
PairsRatio: 2.98
[raw statistics]
Elo: 8.49 ±0.8
Ptnml: 2, 502, 18018, 1473, 5
nElo: 38.38 ±3.2
PairsRatio: 2.93
[raw statistics]
Elo: 32.71 ±1.3
Ptnml: 44, 3760, 16821, 9269, 106
nElo: 71.26 ±2.9
PairsRatio: 2.46
[raw statistics]
2023‑02‑18master vs Stockfish 15
Bench: 4283297
Remove one reduction call
[differences] 168
Elo: 12.58 ±0.8
Ptnml: 11, 1127, 25591, 3221, 50
nElo: 46.26 ±2.6
PairsRatio: 2.87
[raw statistics]
Elo: 31.91 ±1.3
Ptnml: 47, 3828, 16804, 9225, 96
nElo: 69.38 ±2.9
PairsRatio: 2.41
[raw statistics]
2023‑02‑24master vs Stockfish 15
Bench: 4705194
Search tuning at very long time control
[differences] 174
Elo: 11.69 ±0.7
Ptnml: 10, 1151, 25692, 3105, 42
nElo: 43.51 ±2.6
PairsRatio: 2.71
[raw statistics]
Elo: 9.37 ±0.8
Ptnml: 1, 455, 18013, 1526, 5
nElo: 42.48 ±3.1
PairsRatio: 3.36
[raw statistics]
Elo: 32.42 ±1.3
Ptnml: 44, 3804, 16761, 9307, 84
nElo: 70.59 ±2.9
PairsRatio: 2.44
[raw statistics]
2023‑03‑19master vs Stockfish 15
Bench: 4980082
Remove 'si' StateInfo variable/parameter.
[differences] 196
Elo: 13.36 ±0.8
Ptnml: 9, 1098, 25532, 3300, 61
nElo: 48.75 ±2.6
PairsRatio: 3.04
[raw statistics]
Elo: 10.10 ±0.8
Ptnml: 0, 460, 17934, 1589, 17
nElo: 44.63 ±3.1
PairsRatio: 3.49
[raw statistics]
Elo: 35.27 ±1.3
Ptnml: 48, 3610, 16657, 9594, 91
nElo: 76.91 ±2.9
PairsRatio: 2.65
[raw statistics]
2023‑04‑01master vs Stockfish 15
Bench: 4380438
Decrease Depth more for positions not in TT.
[differences] 211
Elo: 14.42 ±0.8
Ptnml: 7, 1031, 25480, 3431, 51
nElo: 52.68 ±2.6
PairsRatio: 3.35
[raw statistics]
Elo: 9.90 ±0.8
Ptnml: 2, 450, 17964, 1575, 9
nElo: 44.20 ±3.1
PairsRatio: 3.50
[raw statistics]
Elo: 37.37 ±1.3
Ptnml: 48, 3485, 16555, 9815, 97
nElo: 81.54 ±2.9
PairsRatio: 2.81
[raw statistics]
2023‑04‑22master vs Stockfish 15
Bench: 3548023
Less reduction for tt move.
[differences] 235
Elo: 14.60 ±0.8
Ptnml: 9, 967, 25569, 3405, 50
nElo: 53.91 ±2.6
PairsRatio: 3.54
[raw statistics]
Elo: 11.32 ±0.8
Ptnml: 0, 386, 17939, 1661, 14
nElo: 50.39 ±3.0
PairsRatio: 4.34
[raw statistics]
Elo: 38.07 ±1.3
Ptnml: 27, 3392, 16687, 9793, 101
nElo: 83.87 ±2.9
PairsRatio: 2.89
[raw statistics]
2023‑05‑07master vs Stockfish 15
Bench: 3808503
Refine deeper post-lmr searches
[differences] 244
Elo: 14.36 ±0.8
Ptnml: 11, 1004, 25521, 3423, 41
nElo: 52.83 ±2.6
PairsRatio: 3.41
[raw statistics]
Elo: 10.63 ±0.7
Ptnml: 0, 379, 18029, 1581, 11
nElo: 48.41 ±3.0
PairsRatio: 4.20
[raw statistics]
Elo: 39.62 ±1.3
Ptnml: 34, 3309, 16568, 9988, 101
nElo: 87.17 ±2.9
PairsRatio: 3.02
[raw statistics]
2023‑06‑04master vs Stockfish 15
Bench: 2551691
Move internal iterative reduction before probcut
[differences] 265
Elo: 15.01 ±0.8
Ptnml: 9, 1037, 25355, 3552, 47
nElo: 54.25 ±2.6
PairsRatio: 3.44
[raw statistics]
Elo: 11.87 ±0.8
Ptnml: 3, 362, 17908, 1720, 7
nElo: 52.68 ±3.0
PairsRatio: 4.73
[raw statistics]
Elo: 38.10 ±1.3
Ptnml: 48, 3387, 16634, 9826, 105
nElo: 83.49 ±2.9
PairsRatio: 2.89
[raw statistics]
2023‑06‑12master vs Stockfish 15
Bench: 2370027
Use block sparse input for the first layer.
[differences] 274
Elo: 17.57 ±0.8
Ptnml: 10, 881, 25223, 3840, 46
nElo: 63.12 ±2.5
PairsRatio: 4.36
[raw statistics]
Elo: 13.95 ±0.8
Ptnml: 1, 290, 17828, 1865, 16
nElo: 60.93 ±2.8
PairsRatio: 6.46
[raw statistics]
Elo: 44.18 ±1.3
Ptnml: 45, 3031, 16345, 10449, 130
nElo: 97.20 ±3.0
PairsRatio: 3.44
[raw statistics]
2023‑06‑22master vs Stockfish 15
Bench: 2593605
Update default net to nn-5af11540bbfe.nnue
[differences] 289
Elo: 18.30 ±0.8
Ptnml: 3, 812, 25265, 3864, 56
nElo: 66.19 ±2.5
PairsRatio: 4.81
[raw statistics]
Elo: 14.33 ±0.8
Ptnml: 3, 297, 17767, 1914, 19
nElo: 61.59 ±2.9
PairsRatio: 6.44
[raw statistics]
Elo: 47.03 ±1.3
Ptnml: 26, 2938, 16102, 10805, 129
nElo: 103.71 ±3.0
PairsRatio: 3.69
[raw statistics]
2023‑06‑29Stockfish 16 [[https://stockfishchess.org/images/logo/icon_128x128.png|width=20px]] [[https://github.githubassets.com/images/icons/emoji/unicode/1f4c8.png|width=20px]]
Bench: 2593605
[differences] 290

Version Comparison

Elo Progression

Game Pair Ratio Progression

30 Day Average

Draw Percentage vs Elo


Stockfish Progression


Historical Information

Stockfish 3 Development (01/03/13 - 30/04/13)
DateVersion1 Thread
2013‑03‑04master vs Stockfish 2.3.1
Bench: 4968764
Increase see prune depth
[differences] 226
Elo: 15.00 ±2.8
WDL: 2906, 13325, 3769
nElo: 26.02 ±4.8
[raw statistics]
2013‑03‑11master vs Stockfish 2.3.1
Bench: 4968764
Be more aggressive on trying to finish iterations
[differences] 227
Elo: 15.49 ±2.8
WDL: 3016, 13077, 3907
nElo: 26.38 ±4.8
[raw statistics]
2013‑03‑11master vs Stockfish 2.3.1
Bench: 4968764
Check for easy move just once
[differences] 228
Elo: 13.42 ±2.8
WDL: 2974, 13280, 3746
nElo: 23.19 ±4.8
[raw statistics]
2013‑03‑16master vs Stockfish 2.3.1
Bench: 5442365
Further increase SEE prune depth
[differences] 232
Elo: 17.77 ±2.8
WDL: 2897, 13184, 3919
nElo: 30.53 ±4.8
[raw statistics]
2013‑03‑24master vs Stockfish 2.3.1
Bench: 4985829
Update bestValue when futility pruning (2)
[differences] 237
Elo: 16.71 ±2.8
WDL: 2874, 13291, 3835
nElo: 28.92 ±4.8
[raw statistics]
2013‑03‑30master vs Stockfish 2.3.1
Bench: 4781239
Set IID half way between d/2 and d-4
[differences] 241
Elo: 18.76 ±2.8
WDL: 2824, 13273, 3903
nElo: 32.46 ±4.8
[raw statistics]
2013‑04‑03master vs Stockfish 2.3.1
Bench: 4705335
Double Impact of Gain tables
[differences] 242
Elo: 15.44 ±2.8
WDL: 3040, 13032, 3928
nElo: 26.21 ±4.8
[raw statistics]
2013‑04‑06master vs Stockfish 2.3.1
Bench: 4361224
Increase null verification threshold to 12 plies
[differences] 249
Elo: 17.11 ±2.8
WDL: 2774, 12861, 3727
nElo: 29.62 ±4.9
[raw statistics]
2013‑04‑07master vs Stockfish 2.3.1
Bench: 5473339
Rescale UCI parameters to 100
[differences] 252
Elo: 19.02 ±2.8
WDL: 2948, 13010, 4042
nElo: 32.29 ±4.8
[raw statistics]
2013‑04‑10master vs Stockfish 2.3.1
Bench: 5157061
De-templetize Position::is_draw()
[differences] 257
Elo: 24.13 ±2.8
WDL: 2791, 13031, 4178
nElo: 41.10 ±4.8
[raw statistics]
2013‑04‑19master vs Stockfish 2.3.1
Bench: 5274705
Skip a couple of popcount in previous patch
[differences] 262
Elo: 28.27 ±2.9
WDL: 2754, 12868, 4378
nElo: 47.69 ±4.9
[raw statistics]
2013‑04‑26master vs Stockfish 2.3.1
Bench: 4311634
Fix a crash introduced few days ago
[differences] 270
Elo: 33.67 ±2.9
WDL: 2642, 12784, 4574
nElo: 56.61 ±4.9
[raw statistics]
2013‑04‑28master vs Stockfish 2.3.1
Bench: 4176431
Temporary revert "Expose EvalInfo struct to search"
[differences] 273
Elo: 30.86 ±2.9
WDL: 2721, 12786, 4493
nElo: 51.82 ±4.9
[raw statistics]
2013‑04‑30Stockfish 3 [[https://stockfishchess.org/images/logo/icon_128x128.png|width=20px]] [[https://github.githubassets.com/images/icons/emoji/unicode/1f4c8.png|width=20px]]
Bench: 4176431
[differences] 275
Stockfish 4 Development (30/04/13 - 20/08/13)
DateVersion1 Thread
2013‑05‑16master vs Stockfish 3
Bench: 4327405
Use two counter moves instead of one
[differences] 28
Elo: 22.11 ±3.0
WDL: 3212, 12305, 4483
nElo: 35.78 ±4.8
[raw statistics]
2013‑05‑23master vs Stockfish 3
Bench: 4821467
Bunch of 3 small patches
[differences] 33
Elo: 26.70 ±2.1
WDL: 6435, 24062, 9503
nElo: 42.53 ±3.4
[raw statistics]
2013‑05‑31master vs Stockfish 3
Bench: 4931544
Passed pawn tuning
[differences] 38
Elo: 29.50 ±2.2
WDL: 6588, 23436, 9976
nElo: 46.13 ±3.4
[raw statistics]
2013‑06‑23master vs Stockfish 3
Bench: 4609948
Fix some stale comments
[differences] 72
Elo: 35.47 ±2.2
WDL: 6196, 23539, 10265
nElo: 55.80 ±3.4
[raw statistics]
2013‑07‑03master vs Stockfish 3
Bench: 4507288
Simplify aspiration window code
[differences] 88
Elo: 37.36 ±2.2
WDL: 6223, 23269, 10508
nElo: 58.35 ±3.4
[raw statistics]
2013‑07‑13master vs Stockfish 3
Bench: 4558173
Fully qualify memset and memcpy
[differences] 100
Elo: 39.27 ±3.1
WDL: 3052, 11645, 5303
nElo: 61.44 ±4.9
[raw statistics]
2013‑07‑19master vs Stockfish 3
Bench: 4769737
Halve king eval margin
[differences] 110
Elo: 39.83 ±3.1
WDL: 3067, 11583, 5350
nElo: 62.10 ±4.9
[raw statistics]
2013‑07‑25master vs Stockfish 3
Bench: 4727133
Rewrite pawn shield and storm code
[differences] 133
Elo: 48.84 ±3.3
WDL: 3203, 10801, 5996
nElo: 73.11 ±4.9
[raw statistics]
2013‑08‑03master vs Stockfish 3
Bench: 4424151
Streamline time computation
[differences] 147
Elo: 50.95 ±3.2
WDL: 3056, 10976, 5968
nElo: 77.14 ±4.9
[raw statistics]
2013‑08‑18master vs Stockfish 3
Bench: 4132374
Further tweak movecount pruning
[differences] 162
Elo: 56.66 ±3.3
WDL: 2988, 10791, 6221
nElo: 85.22 ±5.0
[raw statistics]
2013‑08‑20Stockfish 4 [[https://stockfishchess.org/images/logo/icon_128x128.png|width=20px]] [[https://github.githubassets.com/images/icons/emoji/unicode/1f4c8.png|width=20px]]
Bench: 4132374
[differences] 165
Stockfish DD Development (20/08/13 - 29/11/13)
DateVersion1 Thread
2013‑08‑29master vs Stockfish 4
Bench: 4620975
Enable LMR for dangerous moves
[differences] 12
Elo: 16.18 ±2.9
WDL: 3221, 12627, 4152
nElo: 26.72 ±4.8
[raw statistics]
2013‑09‑01master vs Stockfish 4
Bench: 3453945
Don't use lpthread for Android
[differences] 18
Elo: 19.30 ±2.9
WDL: 3083, 12724, 4193
nElo: 32.11 ±4.8
[raw statistics]
2013‑09‑05master vs Stockfish 4
Bench: 4633330
Do not prune useless checks in QS
[differences] 23
Elo: 23.80 ±2.9
WDL: 2932, 12768, 4300
nElo: 39.78 ±4.8
[raw statistics]
2013‑09‑07master vs Stockfish 4
Bench: 3864419
Remove unreachable values in mobility table
[differences] 27
Elo: 27.66 ±2.9
WDL: 2766, 12879, 4355
nElo: 46.68 ±4.9
[raw statistics]
2013‑09‑11master vs Stockfish 4
Bench: 4554576
Extend checks more when below alpha
[differences] 35
Elo: 24.87 ±2.9
WDL: 2824, 12923, 4253
nElo: 42.04 ±4.8
[raw statistics]
2013‑09‑12master vs Stockfish 4
Bench: 4554579
Revert "Move draw by material check"
[differences] 37
Elo: 24.34 ±2.9
WDL: 2825, 12951, 4224
nElo: 41.22 ±4.8
[raw statistics]
2013‑09‑13master vs Stockfish 4
Bench: 3846852
Increase passed bonus for having more pieces
[differences] 38
Elo: 26.21 ±2.9
WDL: 2871, 12752, 4377
nElo: 43.80 ±4.8
[raw statistics]
2013‑09‑16master vs Stockfish 4
Bench: 3884003
Fix time parameters for blitz games
[differences] 47
Elo: 31.56 ±2.1
WDL: 5588, 25200, 9212
nElo: 52.33 ±3.4
[raw statistics]
2013‑09‑23master vs Stockfish 4
Bench: 3529630
Update disabled warnings for Intel compiler
[differences] 54
Elo: 34.03 ±2.1
WDL: 5449, 25197, 9354
nElo: 56.49 ±3.4
[raw statistics]
2013‑09‑28master vs Stockfish 4
Bench: 3172206
Drop 'is' prefix from query functions
[differences] 62
Elo: 33.49 ±2.9
WDL: 2571, 12936, 4493
nElo: 56.93 ±4.9
[raw statistics]
2013‑09‑29master vs Stockfish 4
Bench: 8336338
Add more depth/positions to bench
[differences] 63
Elo: 33.90 ±2.0
WDL: 5292, 25525, 9183
nElo: 56.93 ±3.4
[raw statistics]
2013‑10‑08master vs Stockfish 4
Bench: 8340585
Use TT refined value to stand pat
[differences] 66
Elo: 36.58 ±2.9
WDL: 2623, 12656, 4721
nElo: 61.07 ±4.9
[raw statistics]
2013‑10‑08master vs Stockfish 4
Bench: 8340585
Increase slowmover and reduce instability
[differences] 67
Elo: 38.91 ±2.1
WDL: 5102, 25335, 9563
nElo: 65.11 ±3.5
[raw statistics]
2013‑10‑09master vs Stockfish 4
Bench: 8279065
Smoother transition for LMR
[differences] 68
Elo: 39.29 ±2.1
WDL: 5071, 25354, 9575
nElo: 65.80 ±3.4
[raw statistics]
2013‑10‑14master vs Stockfish 4
Bench: 7700683
Double king safety weights
[differences] 75
Elo: 43.64 ±2.9
WDL: 2432, 12637, 4931
nElo: 73.12 ±4.9
[raw statistics]
2013‑10‑18master vs Stockfish 4
Bench: 8440524
Score chain pawn also by rank
[differences] 78
Elo: 49.51 ±2.1
WDL: 4821, 24696, 10483
nElo: 81.68 ±3.5
[raw statistics]
2013‑10‑19master vs Stockfish 4
Bench: 9160831
Further increase safe checks bonus
[differences] 80
Elo: 50.11 ±2.1
WDL: 4817, 24636, 10547
nElo: 82.54 ±3.5
[raw statistics]
2013‑10‑20master vs Stockfish 4
Bench: 9294116
Further improve chain pawn evaluation
[differences] 84
Elo: 49.72 ±3.1
WDL: 2677, 11803, 5520
nElo: 79.12 ±4.9
[raw statistics]
2013‑10‑22master vs Stockfish 4
Bench: 8455956
Tweak again chain pawn bonus
[differences] 87
Elo: 53.85 ±3.0
WDL: 2425, 12075, 5500
nElo: 87.51 ±4.9
[raw statistics]
2013‑10‑24master vs Stockfish 4
Bench: 8291883
Retire mirror()
[differences] 94
Elo: 55.18 ±3.0
WDL: 2385, 12080, 5535
nElo: 89.82 ±4.9
[raw statistics]
2013‑10‑28master vs Stockfish 4
Bench: 8029334
Tweak bishop pair and knight weight
[differences] 96
Elo: 52.84 ±2.1
WDL: 4866, 24231, 10903
nElo: 86.04 ±3.5
[raw statistics]
2013‑11‑01master vs Stockfish 4
Bench: 7995098
Set timer to a fixed interval
[differences] 98
Elo: 59.73 ±3.0
WDL: 2324, 11947, 5729
nElo: 96.77 ±5.0
[raw statistics]
2013‑11‑09master vs Stockfish 4
Bench: 7243575
Futility pruning simplification
[differences] 106
Elo: 60.68 ±3.0
WDL: 2198, 12146, 5656
nElo: 99.73 ±5.0
[raw statistics]
2013‑11‑10master vs Stockfish 4
Bench: 9282549
Remove opposed flag for doubled pawns
[differences] 113
Elo: 61.23 ±2.1
WDL: 4333, 24357, 11310
nElo: 100.91 ±3.5
[raw statistics]
2013‑11‑11master vs Stockfish 4
Bench: 8331357
Simplify generate<EVASIONS>
[differences] 116
Elo: 63.85 ±2.1
WDL: 4185, 24361, 11454
nElo: 105.53 ±3.5
[raw statistics]
2013‑11‑29master vs Stockfish 4
Bench: 8596156
Add support for PPC 64bit on Linux
[differences] 123
Elo: 67.44 ±2.1
WDL: 4119, 24094, 11787
nElo: 110.87 ±3.5
[raw statistics]
2013‑11‑29Stockfish DD [[https://stockfishchess.org/images/logo/icon_128x128.png|width=20px]] [[https://github.githubassets.com/images/icons/emoji/unicode/1f4c8.png|width=20px]]
Bench: 8596156
[differences] 124
Stockfish 5 Development (29/11/13 - 31/05/14)
DateVersion1 Thread
2013‑12‑09master vs Stockfish DD
Bench: 7869223
Research at intermediate depth if LMR is very high
[differences] 23
Elo: 7.94 ±1.9
WDL: 5662, 27762, 6576
nElo: 14.37 ±3.4
[raw statistics]
2013‑12‑19master vs Stockfish DD
Bench: 7425809
Faster and simplified threat eval
[differences] 29
Elo: 11.35 ±1.9
WDL: 5766, 27162, 7072
nElo: 20.06 ±3.4
[raw statistics]
2013‑12‑23master vs Stockfish DD
Bench: 6835416
Loosened trigger condition for king safety
[differences] 31
Elo: 18.62 ±2.0
WDL: 5619, 26620, 7761
nElo: 32.31 ±3.4
[raw statistics]
2013‑12‑29master vs Stockfish DD
Bench: 7762310
Retire asymmThreshold
[differences] 36
Elo: 19.54 ±2.0
WDL: 5580, 26593, 7827
nElo: 33.87 ±3.4
[raw statistics]
2014‑01‑02master vs Stockfish DD
Bench: 7602383
Ensure move_importance() is non-zero
[differences] 46
Elo: 25.36 ±2.0
WDL: 5371, 26343, 8286
nElo: 43.67 ±3.4
[raw statistics]
2014‑01‑08master vs Stockfish DD
Bench: 8502826
Position::gives_check - use ci.ksq
[differences] 55
Elo: 29.85 ±2.0
WDL: 5165, 26242, 8593
nElo: 51.32 ±3.4
[raw statistics]
2014‑01‑14master vs Stockfish DD
Bench: 7205153
Introduce 'follow up' moves
[differences] 59
Elo: 29.84 ±2.0
WDL: 5222, 26129, 8649
nElo: 51.09 ±3.4
[raw statistics]
2014‑01‑19master vs Stockfish DD
Bench: 7804908
Small simplification to Position::see
[differences] 64
Elo: 32.49 ±2.0
WDL: 5088, 26094, 8818
nElo: 55.65 ±3.4
[raw statistics]
2014‑01‑29master vs Stockfish DD
Bench: 6875743
Tweak bishop PSQT tables
[differences] 70
Elo: 36.66 ±2.0
WDL: 4905, 25985, 9110
nElo: 62.70 ±3.4
[raw statistics]
2014‑02‑09master vs Stockfish DD
Bench: 8347121
Faster handling of king captures in Position::see
[differences] 79
Elo: 38.63 ±2.0
WDL: 4956, 25659, 9385
nElo: 65.38 ±3.4
[raw statistics]
2014‑02‑22master vs Stockfish DD
Bench: 8430785
Fix a warning with Intel compiler
[differences] 99
Elo: 38.93 ±2.0
WDL: 4944, 25649, 9407
nElo: 65.87 ±3.4
[raw statistics]
2014‑02‑26master vs Stockfish DD
Bench: 7990513
Dynamic draw value
[differences] 100
Elo: 39.25 ±2.1
WDL: 5164, 25172, 9664
nElo: 65.32 ±3.4
[raw statistics]
2014‑03‑14master vs Stockfish DD
Bench: 7451319
Depth dependent aspiration window delta
[differences] 122
Elo: 40.85 ±2.0
WDL: 4925, 25469, 9606
nElo: 68.77 ±3.5
[raw statistics]
2014‑03‑24master vs Stockfish DD
Bench: 7682173
Simplify TT replace strategy
[differences] 138
Elo: 43.70 ±2.1
WDL: 4887, 25221, 9892
nElo: 73.08 ±3.5
[raw statistics]
2014‑04‑08master vs Stockfish DD
Bench: 7533692
Restrict queen mobility to safe squares
[differences] 159
Elo: 47.70 ±2.1
WDL: 4689, 25165, 10146
nElo: 79.86 ±3.5
[raw statistics]
2014‑04‑12master vs Stockfish DD
Bench: 6921356
Move args parsing to UCI::loop
[differences] 174
Elo: 49.21 ±2.1
WDL: 4717, 24938, 10345
nElo: 81.84 ±3.5
[raw statistics]
2014‑04‑21master vs Stockfish DD
Bench: 7384368
Reset DrawValue[] before new search
[differences] 184
Elo: 54.53 ±2.1
WDL: 4742, 24289, 10969
nElo: 89.09 ±3.5
[raw statistics]
2014‑04‑25master vs Stockfish DD
Bench: 7905850
Speed up picking of killers
[differences] 191
Elo: 57.08 ±2.2
WDL: 4858, 23771, 11371
nElo: 91.87 ±3.5
[raw statistics]
2014‑05‑04master vs Stockfish DD
Bench: 8802105
Revert dynamic contempt
[differences] 216
Elo: 53.27 ±2.2
WDL: 5183, 23549, 11268
nElo: 84.84 ±3.5
[raw statistics]
2014‑05‑13master vs Stockfish DD
Bench: 8739659
Drop to qsearch at low depth in razoring
[differences] 227
Elo: 57.15 ±1.8
WDL: 7289, 35641, 17070
nElo: 91.95 ±2.9
[raw statistics]
2014‑05‑17master vs Stockfish DD
Bench: 8732553
Fix an off-by-one bug in extract_pv_from_tt
[differences] 229
Elo: 55.26 ±2.2
WDL: 5108, 23475, 11417
nElo: 87.95 ±3.5
[raw statistics]
2014‑05‑24master vs Stockfish DD
Bench: 7396783
Fix a warning with Intel compiler
[differences] 234
Elo: 53.28 ±2.1
WDL: 4858, 24198, 10944
nElo: 86.68 ±3.5
[raw statistics]
2014‑05‑31Stockfish 5 [[https://stockfishchess.org/images/logo/icon_128x128.png|width=20px]] [[https://github.githubassets.com/images/icons/emoji/unicode/1f4c8.png|width=20px]]
Bench: 8732553
[differences] 236
Stockfish 6 Development (31/05/14 - 27/01/15)
DateVersion1 Thread
2014‑06‑03master vs Stockfish 5
Bench: 8205159
Symmetric King Safety: take 2
[differences] 8
Elo: 5.35 ±1.8
WDL: 5386, 28612, 6002
nElo: 10.03 ±3.4
[raw statistics]
2014‑06‑11master vs Stockfish 5
Bench: 7875814
Simplify pawn threats and merge into ThreatenedByPawn[]
[differences] 22
Elo: 8.30 ±2.0
WDL: 4589, 24491, 5413
nElo: 15.43 ±3.7
[raw statistics]
2014‑06‑29master vs Stockfish 5
Bench: 8759675
Fix Singular extension condition to handle mate scores
[differences] 46
Elo: 10.78 ±1.8
WDL: 5152, 28455, 6393
nElo: 20.10 ±3.4
[raw statistics]
2014‑07‑22master vs Stockfish 5
Bench: 7831429
Outpost tuning
[differences] 62
Elo: 19.63 ±1.8
WDL: 4775, 28192, 7033
nElo: 36.29 ±3.4
[raw statistics]
2014‑08‑06master vs Stockfish 5
Bench: 7461881
Remove insufficient material rule
[differences] 73
Elo: 19.68 ±1.8
WDL: 4737, 28263, 7000
nElo: 36.49 ±3.4
[raw statistics]
2014‑09‑04master vs Stockfish 5
Bench: 7461881
Small tweak to idle_loop()
[differences] 82
Elo: 15.90 ±1.8
WDL: 4986, 28199, 6815
nElo: 29.35 ±3.4
[raw statistics]
2014‑09‑27master vs Stockfish 5
Bench: 6545733
Remove use of half-ply reductions
[differences] 93
Elo: 22.80 ±1.9
WDL: 4664, 28051, 7285
nElo: 41.96 ±3.4
[raw statistics]
2014‑10‑15master vs Stockfish 5
Bench: 7328585
Document why initing eval tables
[differences] 108
Elo: 27.58 ±1.9
WDL: 4613, 27605, 7782
nElo: 49.96 ±3.4
[raw statistics]
2014‑11‑01master vs Stockfish 5
Bench: 6564212
Merge pull request #89 from official-stockfish/pull_no_pretty
[differences] 129
Elo: 31.00 ±1.9
WDL: 4484, 27472, 8044
nElo: 55.96 ±3.4
[raw statistics]
2014‑11‑10master vs Stockfish 5
Bench: 6807896
Profile Build with Hash=16
[differences] 148
Elo: 36.21 ±1.9
WDL: 4357, 27132, 8511
nElo: 64.71 ±3.4
[raw statistics]
2014‑11‑25master vs Stockfish 5
Bench: 8255966
Bitbase index() from ADD to OR
[differences] 168
Elo: 39.02 ±1.9
WDL: 4151, 27225, 8624
nElo: 70.14 ±3.4
[raw statistics]
2014‑12‑07master vs Stockfish 5
Bench: 9324905
Simpler PRNG and faster magics search
[differences] 181
Elo: 41.42 ±1.9
WDL: 4108, 27038, 8854
nElo: 74.04 ±3.4
[raw statistics]
2014‑12‑22master vs Stockfish 5
Bench: 9498821
Prefer names to numbers in storm code
[differences] 197
Elo: 46.26 ±1.9
WDL: 4011, 26683, 9306
nElo: 81.89 ±3.4
[raw statistics]
2015‑01‑07master vs Stockfish 5
Bench: 7604776
Assorted formatting and comment tweaks in position.h
[differences] 213
Elo: 52.10 ±2.0
WDL: 3913, 26688, 9948
nElo: 91.46 ±3.4
[raw statistics]
2015‑01‑18master vs Stockfish 5
Bench: 8080602
Stockfish 6 Release Candidate 1
[differences] 228
Elo: 51.71 ±1.9
WDL: 3723, 26644, 9633
nElo: 91.89 ±3.5
[raw statistics]
2015‑01‑27Stockfish 6 [[https://stockfishchess.org/images/logo/icon_128x128.png|width=20px]] [[https://github.githubassets.com/images/icons/emoji/unicode/1f4c8.png|width=20px]]
Bench: 8918745
[differences] 236
Stockfish 7 Development (27/01/15 - 02/01/16)
DateVersion1 Thread
2015‑02‑08master vs Stockfish 6
Bench: 7699138
Pawn Center Bind Bonus
[differences] 12
Elo: 7.50 ±2.0
WDL: 6423, 26291, 7286
nElo: 12.81 ±3.4
[raw statistics]
2015‑03‑19master vs Stockfish 6
Bench: 8226843
Retire ConditionVariable
[differences] 123
Elo: 15.92 ±1.8
WDL: 4796, 28576, 6628
nElo: 29.89 ±3.4
[raw statistics]
2015‑03‑29master vs Stockfish 6
Bench: 7658627
Remove some difficult to understand C++11 constructs
[differences] 137
Elo: 19.02 ±1.8
WDL: 4590, 28633, 6777
nElo: 35.82 ±3.4
[raw statistics]
2015‑04‑10master vs Stockfish 6
Bench: 6985247
Allow Position::init() to be called more than once
[differences] 151
Elo: 22.12 ±1.8
WDL: 4480, 28497, 7023
nElo: 41.48 ±3.4
[raw statistics]
2015‑05‑09master vs Stockfish 6
Bench: 8787152
Smart TT save
[differences] 164
Elo: 28.82 ±1.8
WDL: 4197, 28295, 7508
nElo: 53.80 ±3.4
[raw statistics]
2015‑06‑07master vs Stockfish 6
Bench: 6716940
Simplify outpost evaluation
[differences] 176
Elo: 27.85 ±1.9
WDL: 4501, 27798, 7701
nElo: 50.86 ±3.4
[raw statistics]
2015‑07‑16master vs Stockfish 6
Bench: 6943812
Fix formatting of previous patch
[differences] 187
Elo: 30.74 ±1.9
WDL: 4367, 27736, 7897
nElo: 56.09 ±3.4
[raw statistics]
2015‑07‑30master vs Stockfish 6
Bench: 8040572
Simplify IID depth formula
[differences] 192
Elo: 34.04 ±1.9
WDL: 4283, 27527, 8190
nElo: 61.72 ±3.4
[raw statistics]
2015‑10‑03master vs Stockfish 6
Bench: 8073614
File based passed pawn bonus
[differences] 214
Elo: 44.23 ±1.9
WDL: 3690, 27555, 8755
nElo: 80.99 ±3.4
[raw statistics]
2015‑10‑25master vs Stockfish 6
Bench: 8004751
Use atomics instead of volatile
[differences] 232
Elo: 61.76 ±1.9
WDL: 3197, 26570, 10233
nElo: 110.69 ±3.5
[raw statistics]
2015‑12‑27master vs Stockfish 6
Bench: 8355485
Stockfish 7 Beta 1
[differences] 267
Elo: 62.62 ±1.6
WDL: 4472, 40358, 15170
nElo: 113.94 ±2.8
[raw statistics]
2016‑01‑02Stockfish 7 [[https://stockfishchess.org/images/logo/icon_128x128.png|width=20px]] [[https://github.githubassets.com/images/icons/emoji/unicode/1f4c8.png|width=20px]]
Bench: 8355485
[differences] 273
Stockfish 8 Development (02/01/16 - 01/11/16)
DateVersion1 Thread
2016‑01‑28master vs Stockfish 7
Bench: 7751425
Time management simplification
[differences] 16
Elo: 4.93 ±1.5
WDL: 3814, 31804, 4382
nElo: 10.90 ±3.4
[raw statistics]
2016‑03‑10master vs Stockfish 7
Bench: 8261839
Add follow up moves history for move ordering
[differences] 28
Elo: 12.83 ±1.5
WDL: 3354, 31816, 4830
nElo: 28.44 ±3.4
[raw statistics]
2016‑04‑08master vs Stockfish 7
Bench: 7482426
Small passed pawn simplification
[differences] 42
Elo: 15.27 ±1.5
WDL: 3251, 31741, 5008
nElo: 33.74 ±3.4
[raw statistics]
2016‑05‑20master vs Stockfish 7
Bench: 8428997
More detailed dependence of time allocation
[differences] 64
Elo: 29.44 ±1.6
WDL: 2719, 31217, 6104
nElo: 63.61 ±3.4
[raw statistics]
2016‑06‑10master vs Stockfish 7
Bench: 8276130
Stat Formula Tweak
[differences] 76
Elo: 36.29 ±1.6
WDL: 2606, 30625, 6769
nElo: 76.48 ±3.4
[raw statistics]
2016‑07‑24master vs Stockfish 7
Bench: 8145304
Allow null pruning at depth 1
[differences] 94
Elo: 49.73 ±1.7
WDL: 2415, 29483, 8102
nElo: 100.27 ±3.4
[raw statistics]
2016‑08‑18master vs Stockfish 7
Bench: 7662861
Remove a stale assignment
[differences] 101
Elo: 52.59 ±1.7
WDL: 2413, 29165, 8422
nElo: 104.74 ±3.4
[raw statistics]
2016‑09‑07master vs Stockfish 7
Bench: 6024713
Refactor previous patch
[differences] 124
Elo: 66.53 ±1.8
WDL: 1893, 28647, 9460
nElo: 131.97 ±3.3
[raw statistics]
2016‑10‑07master vs Stockfish 7
Bench: 6421663
Optimisation of Position::see and Position::see_sign
[differences] 149
Elo: 76.60 ±1.8
WDL: 1571, 28179, 10250
nElo: 151.24 ±3.3
[raw statistics]
2016‑11‑01Stockfish 8 [[https://stockfishchess.org/images/logo/icon_128x128.png|width=20px]] [[https://github.githubassets.com/images/icons/emoji/unicode/1f4c8.png|width=20px]]
Bench: 5926706
[differences] 168
Stockfish 9 Development (01/11/16 - 31/01/18)
DateVersion1 Thread
2016‑12‑31master vs Stockfish 8
Bench: 5468995
Small eval cleanup and renaming
[differences] 55
Elo: 5.21 ±1.5
WDL: 3656, 32088, 4256
nElo: 11.72 ±3.4
[raw statistics]
2017‑01‑29master vs Stockfish 8
Bench: 5941174
Simplify TT penalty stat (#980)
[differences] 86
Elo: 8.82 ±1.5
WDL: 3580, 31825, 4595
nElo: 19.53 ±3.4
[raw statistics]
2017‑03‑08master vs Stockfish 8
Bench: 5803228
Helper functions to count material for both sides
[differences] 106
Elo: 10.84 ±1.6
WDL: 3569, 31614, 4817
nElo: 23.73 ±3.4
[raw statistics]
2017‑04‑20master vs Stockfish 8
Bench: 6581936
simplify logic for history based pruning
[differences] 127
Elo: 15.17 ±1.6
WDL: 3430, 31395, 5175
nElo: 32.82 ±3.4
[raw statistics]
2017‑05‑07master vs Stockfish 8
Bench: 6107863
Linear Protector bonus by distance
[differences] 144
Elo: 20.25 ±1.6
WDL: 3258, 31155, 5587
nElo: 43.35 ±3.4
[raw statistics]
2017‑06‑21master vs Stockfish 8
Bench: 5725676
Increase reduction if tt-move is a capture
[differences] 167
Elo: 27.41 ±1.6
WDL: 2918, 31015, 6067
nElo: 58.52 ±3.4
[raw statistics]
2017‑08‑26master vs Stockfish 8
Bench: 5965302
Improve multi-threaded mate finding
[differences] 218
Elo: 29.32 ±1.6
WDL: 2886, 30860, 6254
nElo: 62.17 ±3.4
[raw statistics]
2017‑10‑02master vs Stockfish 8
Bench: 5620312
Good bishops on the main diagonals
[differences] 237
Elo: 32.61 ±1.6
WDL: 2688, 30881, 6431
nElo: 69.44 ±3.4
[raw statistics]
2017‑11‑03master vs Stockfish 8
Bench: 5536775
Introduce capture history table for capture move sorting
[differences] 247
Elo: 35.18 ±1.6
WDL: 2682, 30600, 6718
nElo: 73.93 ±3.4
[raw statistics]
2017‑12‑03master vs Stockfish 8
Bench: 5051254
Use constexpr when makes sense
[differences] 261
Elo: 43.31 ±1.7
WDL: 2406, 30227, 7367
nElo: 90.06 ±3.3
[raw statistics]
2018‑01‑23master vs Stockfish 8
Bench: 5783344
Contempt 20
[differences] 286
Elo: 57.25 ±1.9
WDL: 2917, 27634, 9449
nElo: 106.75 ±3.4
[raw statistics]
2018‑01‑31Stockfish 9 [[https://stockfishchess.org/images/logo/icon_128x128.png|width=20px]] [[https://github.githubassets.com/images/icons/emoji/unicode/1f4c8.png|width=20px]]
Bench: 5023629
[differences] 291
Stockfish 10 Development (31/01/18 - 29/11/18)
DateVersion1 Thread
2018‑02‑28master vs Stockfish 9
Bench: 5765806
Reintroduce depth 2 razoring (with additional margin)
[differences] 41
Elo: 5.16 ±1.7
WDL: 5143, 30105, 5752
nElo: 10.02 ±3.4
[raw statistics]
2018‑03‑07master vs Stockfish 9
Bench: 5544908
Simplification: use Arctan for the optimism S-curve
[differences] 53
Elo: 6.94 ±1.7
WDL: 4813, 29575, 5612
nElo: 13.60 ±3.4
[raw statistics]
2018‑03‑13master vs Stockfish 9
Bench: 5741807
Use intrinsics only for LSB/MSB
[differences] 64
Elo: 13.03 ±1.7
WDL: 4463, 29574, 5963
nElo: 25.59 ±3.4
[raw statistics]
2018‑03‑26master vs Stockfish 9
Bench: 5934103
Make kingRing always 8 squares
[differences] 75
Elo: 13.77 ±1.7
WDL: 4339, 29737, 5924
nElo: 27.26 ±3.4
[raw statistics]
2018‑04‑03master vs Stockfish 9
Bench: 4989125
Remove the Queen from the mobility area of minor pieces
[differences] 92
Elo: 16.59 ±1.7
WDL: 4283, 29525, 6192
nElo: 32.54 ±3.4
[raw statistics]
2018‑04‑07master vs Stockfish 9
Bench: 5170165
Reset negative statScore on fail high
[differences] 94
Elo: 16.63 ±1.8
WDL: 4397, 29293, 6310
nElo: 32.25 ±3.4
[raw statistics]
2018‑04‑23master vs Stockfish 9
Bench: 5549801
Alternative formula for dynamic contempt
[differences] 106
Elo: 15.90 ±1.8
WDL: 4462, 29247, 6291
nElo: 30.76 ±3.4
[raw statistics]
2018‑04‑29master vs Stockfish 9
Bench: 5254862
Always scale using pawn contribution
[differences] 112
Elo: 14.61 ±1.7
WDL: 4430, 29459, 6111
nElo: 28.54 ±3.4
[raw statistics]
2018‑05‑03master vs Stockfish 9
Bench: 5186783
Tweak the connected[] array value for pawns on rank 5
[differences] 116
Elo: 18.52 ±1.8
WDL: 4385, 29100, 6515
nElo: 35.63 ±3.4
[raw statistics]
2018‑05‑13master vs Stockfish 9
Bench: 5294316
Update search.cpp
[differences] 128
Elo: 24.92 ±1.8
WDL: 4175, 28786, 7039
nElo: 47.42 ±3.4
[raw statistics]
2018‑05‑24master vs Stockfish 9
Bench: 5167159
LMR Capture Tweak
[differences] 137
Elo: 26.72 ±1.8
WDL: 4033, 28864, 7103
nElo: 51.08 ±3.4
[raw statistics]
2018‑06‑05master vs Stockfish 9
Bench: 4326784
Call cycle detection before qsearch()
[differences] 148
Elo: 28.16 ±1.8
WDL: 3971, 28823, 7206
nElo: 53.79 ±3.4
[raw statistics]
2018‑06‑11master vs Stockfish 9
Bench: 4980482
Optimize an expression in endgame.cpp
[differences] 154
Elo: 29.72 ±1.9
WDL: 4335, 27917, 7748
nElo: 54.60 ±3.4
[raw statistics]
2018‑06‑23master vs Stockfish 9
Bench: 4557946
Another set of tuned values after one million games
[differences] 162
Elo: 31.98 ±1.9
WDL: 4354, 27621, 8025
nElo: 58.11 ±3.4
[raw statistics]
2018‑07‑19master vs Stockfish 9
Bench: 4817583
Better check evasion move sorting
[differences] 179
Elo: 36.70 ±1.9
WDL: 4312, 27166, 8522
nElo: 65.70 ±3.4
[raw statistics]
2018‑07‑27master vs Stockfish 9
Bench: 4905530
Simplify cmh pruning
[differences] 199
Elo: 37.45 ±1.9
WDL: 4183, 27339, 8478
nElo: 67.55 ±3.4
[raw statistics]
2018‑07‑28master vs Stockfish 9
Bench: 4883742
Increase the mg->eg gradient for the PawnlessFlank malus
[differences] 200
Elo: 35.84 ±1.9
WDL: 4235, 27418, 8347
nElo: 64.78 ±3.4
[raw statistics]
2018‑07‑31master vs Stockfish 9
Bench: 5591925
Small tweaks to recent code changes
[differences] 203
Elo: 37.67 ±1.9
WDL: 4019, 27642, 8339
nElo: 68.82 ±3.4
[raw statistics]
2018‑08‑08master vs Stockfish 9
Bench: 4669050
First check threshold in space evaluation
[differences] 207
Elo: 37.78 ±1.9
WDL: 4224, 27220, 8556
nElo: 67.82 ±3.4
[raw statistics]
2018‑08‑12master vs Stockfish 9
Bench: 4694813
Combo of several promising parameter tweaks
[differences] 211
Elo: 40.88 ±1.9
WDL: 4069, 27177, 8754
nElo: 73.46 ±3.4
[raw statistics]
2018‑08‑14master vs Stockfish 9
Bench: 4272361
Double weight of capture history
[differences] 214
Elo: 41.69 ±1.9
WDL: 3942, 27339, 8719
nElo: 75.47 ±3.4
[raw statistics]
2018‑08‑17master vs Stockfish 9
Bench: 4592766
Use an affine formula to mix stats and eval
[differences] 217
Elo: 43.15 ±1.9
WDL: 3922, 27213, 8865
nElo: 77.82 ±3.4
[raw statistics]
2018‑08‑28master vs Stockfish 9
Bench: 4172767
Tweak stat bonus formula
[differences] 220
Elo: 44.20 ±1.9
WDL: 3861, 27217, 8922
nElo: 79.79 ±3.4
[raw statistics]
2018‑08‑28master vs Stockfish 9
Bench: 4413173
Remove PawnsOnBothFlanks
[differences] 225
Elo: 42.37 ±1.9
WDL: 3952, 27242, 8806
nElo: 76.44 ±3.4
[raw statistics]
2018‑09‑01master vs Stockfish 9
Bench: 4609645
Re-introduce "keep pawns on both flanks"
[differences] 227
Elo: 46.46 ±1.9
WDL: 3804, 27075, 9121
nElo: 83.56 ±3.4
[raw statistics]
2018‑09‑10master vs Stockfish 9
Bench: 4248710
Tweak opposite colored bishops endgame scaling
[differences] 230
Elo: 45.47 ±1.9
WDL: 3911, 26973, 9116
nElo: 81.36 ±3.4
[raw statistics]
2018‑09‑27master vs Stockfish 9
Bench: 4059356
Fix two typos in comments
[differences] 235
Elo: 46.93 ±1.9
WDL: 3883, 26864, 9253
nElo: 83.72 ±3.4
[raw statistics]
2018‑10‑14master vs Stockfish 9
Bench: 4274207
Simplify check extensions
[differences] 241
Elo: 49.01 ±1.9
WDL: 3783, 26829, 9388
nElo: 87.49 ±3.4
[raw statistics]
2018‑10‑25master vs Stockfish 9
Bench: 3314347
On main thread: reduce depth after fail high
[differences] 245
Elo: 52.82 ±1.9
WDL: 3514, 26937, 9549
nElo: 95.10 ±3.4
[raw statistics]
2018‑11‑01master vs Stockfish 9
Bench: 3556672
Fix issues from using adjustedDepth too broadly
[differences] 248
Elo: 51.68 ±1.9
WDL: 3581, 26932, 9487
nElo: 92.90 ±3.4
[raw statistics]
2018‑11‑08master vs Stockfish 9
Bench: 3647775
Update list of top CPU contributors
[differences] 254
Elo: 50.43 ±1.9
WDL: 3773, 26689, 9538
nElo: 89.65 ±3.5
[raw statistics]
2018‑11‑19master vs Stockfish 9
Bench: 3717396
Stockfish 10-beta
[differences] 267
Elo: 53.77 ±1.9
WDL: 3612, 26634, 9754
nElo: 95.73 ±3.5
[raw statistics]
2018‑11‑27master vs Stockfish 9
Bench: 3939338
Simplify casting extension
[differences] 274
Elo: 54.21 ±1.9
WDL: 3562, 26685, 9753
nElo: 96.75 ±3.5
[raw statistics]
2018‑11‑29Stockfish 10 [[https://stockfishchess.org/images/logo/icon_128x128.png|width=20px]] [[https://github.githubassets.com/images/icons/emoji/unicode/1f4c8.png|width=20px]]
Bench: 3939338
[differences] 277
Stockfish 11 Development (29/11/18 - 17/01/20)
DateVersion1 Thread8 Threads
2018‑12‑06master vs Stockfish 10
Bench: 3773021
Revert "pseudo_legal() and MOVE_NONE"
[differences] 8
Elo: 4.32 ±1.7
WDL: 5033, 29437, 5530
nElo: 8.40 ±3.4
[raw statistics]
2018‑12‑13master vs Stockfish 10
Bench: 3332460
A combo of parameter tweaks
[differences] 16
Elo: 8.06 ±1.8
WDL: 5020, 29032, 5948
nElo: 15.41 ±3.4
[raw statistics]
Elo: 12.70 ±1.7
WDL: 4068, 30403, 5529
nElo: 25.98 ±3.4
[raw statistics]
2018‑12‑16master vs Stockfish 10
Bench: 3646542
Use stronglyProtected
[differences] 21
Elo: 7.85 ±1.7
WDL: 5107, 30004, 6037
nElo: 15.10 ±3.4
[raw statistics]
2019‑01‑04master vs Stockfish 10
Bench: 3559104
Check tablebase files
[differences] 40
Elo: 7.20 ±1.8
WDL: 5164, 28843, 5993
nElo: 13.64 ±3.4
[raw statistics]
2019‑01‑10master vs Stockfish 10
Bench: 3739723
Remove pvExact
[differences] 45
Elo: 13.14 ±1.8
WDL: 4779, 28930, 6291
nElo: 25.03 ±3.4
[raw statistics]
Elo: 16.17 ±1.7
WDL: 3887, 30366, 5747
nElo: 33.07 ±3.4
[raw statistics]
2019‑01‑22master vs Stockfish 10
Bench: 3665090
Simplify TrappedRook
[differences] 53
Elo: 13.98 ±1.8
WDL: 4714, 28963, 6323
nElo: 26.68 ±3.4
[raw statistics]
2019‑02‑03master vs Stockfish 10
Bench: 3653942
Less king danger if we have a knight
[differences] 61
Elo: 17.71 ±1.8
WDL: 4624, 28715, 6661
nElo: 33.46 ±3.4
[raw statistics]
Elo: 19.77 ±1.7
WDL: 3728, 30270, 6002
nElo: 40.32 ±3.4
[raw statistics]
2019‑03‑12master vs Stockfish 10
Bench: 3318033
Increase thread stack for OS X (#2035)
[differences] 80
Elo: 16.58 ±1.8
WDL: 4635, 28823, 6542
nElo: 31.46 ±3.4
[raw statistics]
2019‑03‑31master vs Stockfish 10
Bench: 3548313
Assorted trivial cleanups 3/2019 (#2030)
[differences] 91
Elo: 16.58 ±1.8
WDL: 4742, 28609, 6649
nElo: 31.16 ±3.4
[raw statistics]
Elo: 24.33 ±1.7
WDL: 3633, 29937, 6430
nElo: 48.91 ±3.4
[raw statistics]
2019‑04‑24master vs Stockfish 10
Bench: 3402947
Remove useless initializations (#2115)
[differences] 113
Elo: 16.39 ±1.8
WDL: 4634, 28847, 6519
nElo: 31.13 ±3.4
[raw statistics]
2019‑05‑05master vs Stockfish 10
Bench: 3644175
LMR for captures not cracking alpha
[differences] 121
Elo: 16.65 ±1.8
WDL: 4788, 28508, 6704
nElo: 31.17 ±3.4
[raw statistics]
2019‑05‑15master vs Stockfish 10
Bench: 3824325
Update failedHighCnt rule #2063
[differences] 136
Elo: 19.76 ±1.8
WDL: 4665, 28397, 6938
nElo: 36.86 ±3.4
[raw statistics]
Elo: 28.93 ±1.7
WDL: 3573, 29531, 6896
nElo: 57.18 ±3.4
[raw statistics]
2019‑06‑09master vs Stockfish 10
Bench: 3424592
Remove depth condition for ttPv (#2166)
[differences] 151
Elo: 19.87 ±1.9
WDL: 4796, 28123, 7081
nElo: 36.62 ±3.4
[raw statistics]
2019‑06‑20master vs Stockfish 10
Bench: 3398333
More bonus for free passed pawn
[differences] 161
Elo: 24.06 ±1.8
WDL: 4547, 28140, 7313
nElo: 44.48 ±3.4
[raw statistics]
Elo: 30.76 ±1.7
WDL: 3462, 29544, 6994
nElo: 60.92 ±3.4
[raw statistics]
2019‑06‑27master vs Stockfish 10
Bench: 3633546
Bonus for double attacks
[differences] 167
Elo: 22.75 ±1.9
WDL: 4644, 28096, 7260
nElo: 41.95 ±3.4
[raw statistics]
2019‑07‑11master vs Stockfish 10
Bench: 3206912
Assorted trivial cleanups June 2019
[differences] 176
Elo: 24.39 ±1.9
WDL: 4596, 28005, 7399
nElo: 44.83 ±3.4
[raw statistics]
2019‑07‑25master vs Stockfish 10
Bench: 3935523
Tweak of SEE pruning condition
[differences] 192
Elo: 25.72 ±1.9
WDL: 4519, 28006, 7475
nElo: 47.32 ±3.4
[raw statistics]
Elo: 37.49 ±1.7
WDL: 3225, 29251, 7524
nElo: 73.63 ±3.4
[raw statistics]
2019‑08‑14master vs Stockfish 10
Bench: 4139590
Tweak unsafe checks
[differences] 198
Elo: 32.24 ±1.9
WDL: 4168, 27963, 7869
nElo: 59.45 ±3.4
[raw statistics]
2019‑08‑26master vs Stockfish 10
Bench: 3568210
Tweak Late Move Reduction at root
[differences] 207
Elo: 35.63 ±1.9
WDL: 4021, 27870, 8109
nElo: 65.62 ±3.4
[raw statistics]
Elo: 44.52 ±1.8
WDL: 2958, 28986, 8056
nElo: 86.99 ±3.4
[raw statistics]
2019‑09‑12master vs Stockfish 10
Bench: 3954190
Scale down complexity
[differences] 211
Elo: 39.10 ±1.9
WDL: 3824, 27869, 8307
nElo: 72.22 ±3.4
[raw statistics]
2019‑09‑16master vs Stockfish 10
Bench: 4272173
Raise stack size to 8MB for pthreads
[differences] 218
Elo: 37.63 ±1.9
WDL: 4007, 27670, 8323
nElo: 68.83 ±3.4
[raw statistics]
Elo: 46.57 ±1.8
WDL: 2764, 28492, 7994
nElo: 91.44 ±3.4
[raw statistics]
2019‑09‑24master vs Stockfish 10
Bench: 3618154
Increase weight for supported pawns
[differences] 226
Elo: 38.97 ±1.9
WDL: 3857, 27818, 8325
nElo: 71.81 ±3.4
[raw statistics]
2019‑10‑05master vs Stockfish 10
Bench: 4131643
Introduce separate counter-move tables
[differences] 239
Elo: 41.96 ±1.9
WDL: 3746, 27701, 8553
nElo: 77.13 ±3.4
[raw statistics]
Elo: 51.76 ±1.8
WDL: 2628, 28829, 8543
nElo: 101.26 ±3.4
[raw statistics]
2019‑10‑18master vs Stockfish 10
Bench: 4423737
Current capture for Counter-Move history
[differences] 247
Elo: 44.63 ±1.9
WDL: 3641, 27608, 8751
nElo: 81.93 ±3.4
[raw statistics]
2019‑11‑04master vs Stockfish 10
Bench: 4707799
Rook PSQT Tuned
[differences] 259
Elo: 42.20 ±1.9
WDL: 3686, 27793, 8521
nElo: 77.91 ±3.4
[raw statistics]
Elo: 52.90 ±1.8
WDL: 2601, 28754, 8645
nElo: 103.29 ±3.4
[raw statistics]
2019‑11‑14master vs Stockfish 10
Bench: 4532366
Prune before extension
[differences] 266
Elo: 43.12 ±1.9
WDL: 3676, 27709, 8615
nElo: 79.39 ±3.4
[raw statistics]
2019‑11‑21master vs Stockfish 10
Bench: 5067870
Do lmr for more captures
[differences] 271
Elo: 46.56 ±1.5
WDL: 5306, 41395, 13299
nElo: 85.60 ±2.8
[raw statistics]
Elo: 53.93 ±1.8
WDL: 2502, 28837, 8661
nElo: 105.86 ±3.4
[raw statistics]
2019‑12‑02master vs Stockfish 10
Bench: 5122362
UnblockedStorm tuned
[differences] 278
Elo: 44.88 ±1.5
WDL: 5273, 41746, 12981
nElo: 83.21 ±2.8
[raw statistics]
2019‑12‑10master vs Stockfish 10
Bench: 5371271
Refine improving-logic
[differences] 288
Elo: 47.27 ±1.5
WDL: 5329, 41229, 13442
nElo: 86.56 ±2.8
[raw statistics]
Elo: 56.62 ±1.8
WDL: 2365, 28809, 8826
nElo: 111.42 ±3.4
[raw statistics]
2020‑01‑07master vs Stockfish 10
Bench: 4747984
Tuned nullmove search
[differences] 294
Elo: 51.50 ±1.5
Ptnml: 160, 3173, 15729, 9546, 1387
nElo: 98.05 ±2.8
PairsRatio: 3.28
[raw statistics]
Elo: 58.15 ±1.7
Ptnml: 36, 1527, 11059, 6509, 860
nElo: 118.69 ±3.4
PairsRatio: 4.71
[raw statistics]
2020‑01‑17master vs Stockfish 10
Bench: 5156767
Stockfish 11
[differences] 307
Elo: 53.59 ±1.5
Ptnml: 138, 2988, 15833, 9631, 1407
nElo: 102.99 ±2.8
PairsRatio: 3.53
[raw statistics]
Elo: 58.07 ±1.7
Ptnml: 36, 1478, 11159, 6463, 854
nElo: 119.25 ±3.4
PairsRatio: 4.83
[raw statistics]
2020‑01‑17Stockfish 11 [[https://stockfishchess.org/images/logo/icon_128x128.png|width=20px]] [[https://github.githubassets.com/images/icons/emoji/unicode/1f4c8.png|width=20px]]
Bench: 5156767
[differences] 307
Stockfish 12 Development (17/01/20 - 02/09/20)
DateVersion1 Thread8 Threads
2020‑01‑28master vs Stockfish 11
Bench: 5545845
More bonus for bestMoves
[differences] 16
Elo: -2.47 ±1.3
Ptnml: 330, 5657, 18424, 5285, 303
nElo: -5.21 ±2.8
PairsRatio: 0.93
[raw statistics]
2020‑01‑31master vs Stockfish 11
Bench: 5153165
Revert 5 recent patches
[differences] 19
Elo: 0.85 ±1.3
Ptnml: 306, 5327, 18593, 5457, 314
nElo: 1.80 ±2.8
PairsRatio: 1.02
[raw statistics]
2020‑02‑27master vs Stockfish 11
Bench: 4923286
Weak queen protection
[differences] 32
Elo: 1.33 ±1.3
Ptnml: 327, 5308, 18486, 5567, 312
nElo: 2.80 ±2.8
PairsRatio: 1.04
[raw statistics]
Elo: 0.51 ±1.4
Ptnml: 118, 3072, 13560, 3133, 117
nElo: 1.21 ±3.4
PairsRatio: 1.02
[raw statistics]
2020‑03‑20master vs Stockfish 11
Bench: 5398277
Adjust singular extension search depth
[differences] 48
Elo: 2.94 ±1.3
Ptnml: 351, 5099, 18580, 5631, 339
nElo: 6.21 ±2.8
PairsRatio: 1.10
[raw statistics]
2020‑04‑07master vs Stockfish 11
Bench: 4417023
Introduce capture history pruning
[differences] 63
Elo: 5.74 ±1.3
Ptnml: 274, 5058, 18460, 5818, 390
nElo: 12.11 ±2.8
PairsRatio: 1.16
[raw statistics]
Elo: 6.49 ±1.5
Ptnml: 113, 2854, 13369, 3501, 163
nElo: 15.05 ±3.4
PairsRatio: 1.23
[raw statistics]
2020‑04‑16master vs Stockfish 11
Bench: 4958027
Remove one condition in probcut TTmove
[differences] 76
Elo: 11.33 ±1.3
Ptnml: 281, 4681, 18282, 6313, 443
nElo: 23.65 ±2.8
PairsRatio: 1.36
[raw statistics]
2020‑05‑02master vs Stockfish 11
Bench: 4247490
Fishtest Tuning Framework
[differences] 84
Elo: 15.21 ±1.3
Ptnml: 236, 4370, 18388, 6545, 461
nElo: 32.08 ±2.8
PairsRatio: 1.52
[raw statistics]
Elo: 16.71 ±1.5
Ptnml: 105, 2457, 13100, 4087, 251
nElo: 37.85 ±3.4
PairsRatio: 1.69
[raw statistics]
2020‑05‑21master vs Stockfish 11
Bench: 4778956
Tweak knight mobility
[differences] 100
Elo: 15.97 ±1.3
Ptnml: 230, 4544, 17963, 6766, 497
nElo: 33.08 ±2.8
PairsRatio: 1.52
[raw statistics]
2020‑06‑06master vs Stockfish 11
Bench: 4582693
Use lowply-history also on low depths
[differences] 117
Elo: 19.45 ±1.4
Ptnml: 250, 4360, 17761, 7042, 587
nElo: 39.70 ±2.8
PairsRatio: 1.65
[raw statistics]
Elo: 23.70 ±1.5
Ptnml: 88, 2142, 12987, 4524, 259
nElo: 53.99 ±3.4
PairsRatio: 2.14
[raw statistics]
2020‑06‑13master vs Stockfish 11
Bench: 4246971
Tuned values for search constants
[differences] 127
Elo: 20.91 ±1.4
Ptnml: 195, 4263, 17878, 7069, 595
nElo: 43.14 ±2.8
PairsRatio: 1.72
[raw statistics]
Elo: 24.86 ±1.5
Ptnml: 81, 2088, 13016, 4523, 292
nElo: 56.58 ±3.4
PairsRatio: 2.22
[raw statistics]
2020‑06‑29master vs Stockfish 11
Bench: 4523573
Tweak single queen endgame scaling
[differences] 148
Elo: 25.67 ±1.3
Ptnml: 192, 3878, 17888, 7397, 645
nElo: 53.10 ±2.8
PairsRatio: 1.98
[raw statistics]
2020‑07‑17master vs Stockfish 11
Bench: 4578298
Do not overwrite valuable TT data
[differences] 163
Elo: 26.44 ±1.3
Ptnml: 192, 3800, 17928, 7418, 662
nElo: 54.75 ±2.8
PairsRatio: 2.02
[raw statistics]
Elo: 30.71 ±1.5
Ptnml: 63, 1923, 12759, 4935, 320
nElo: 69.50 ±3.4
PairsRatio: 2.65
[raw statistics]
2020‑07‑31master vs Stockfish 11
Bench: 4746616
Tweak cutnode reduction
[differences] 167
Elo: 25.49 ±1.4
Ptnml: 203, 3910, 17861, 7342, 684
nElo: 52.38 ±2.8
PairsRatio: 1.95
[raw statistics]
Elo: 32.39 ±1.5
Ptnml: 69, 1829, 12779, 4961, 362
nElo: 73.03 ±3.4
PairsRatio: 2.80
[raw statistics]
2020‑08‑06master vs Stockfish 11
Bench: 4746616
Add NNUE evaluation
[differences] 168
Elo: 83.42 ±1.7
Ptnml: 172, 2656, 12724, 11761, 2687
nElo: 144.72 ±3.0
PairsRatio: 5.11
[raw statistics]
Elo: 86.10 ±1.9
Ptnml: 36, 1192, 9342, 7881, 1549
nElo: 163.20 ±3.6
PairsRatio: 7.68
[raw statistics]
2020‑08‑07Add NNUE evaluation [[https://stockfishchess.org/images/logo/icon_128x128.png|width=20px]]
Bench: 4746616
[differences] 168
2020‑08‑08master vs Stockfish 11
Bench: 4084753
LMR search tweak
[differences] 185
Elo: 106.20 ±1.7
Ptnml: 67, 1716, 11867, 13060, 3290
nElo: 189.91 ±3.2
PairsRatio: 9.17
[raw statistics]
2020‑08‑11master vs Stockfish 11
Bench: 4290577
This commit enables a mixed bench
[differences] 205
Elo: 125.60 ±1.7
Ptnml: 48, 1240, 10613, 14070, 4029
nElo: 224.82 ±3.3
PairsRatio: 14.05
[raw statistics]
Elo: 111.78 ±1.9
Ptnml: 11, 591, 8286, 9168, 1944
nElo: 217.93 ±3.8
PairsRatio: 18.46
[raw statistics]
2020‑08‑18master vs Stockfish 11
Bench: 4026216
Fix Makefile typo
[differences] 226
Elo: 121.54 ±1.7
Ptnml: 52, 1373, 10930, 13640, 4005
nElo: 215.22 ±3.3
PairsRatio: 12.38
[raw statistics]
Elo: 111.40 ±1.9
Ptnml: 20, 700, 8128, 9161, 1991
nElo: 213.20 ±3.8
PairsRatio: 15.49
[raw statistics]
2020‑08‑30master vs Stockfish 11
Bench: 3736029
Update parameters in classical evaluation
[differences] 255
Elo: 130.96 ±1.7
Ptnml: 44, 1161, 10305, 14128, 4362
nElo: 232.58 ±3.4
PairsRatio: 15.34
[raw statistics]
2020‑09‑02master vs Stockfish 11
Bench: 3624569
Stockfish 12
[differences] 262
Elo: 133.65 ±1.7
Ptnml: 32, 1088, 10158, 14286, 4436
nElo: 238.67 ±3.4
PairsRatio: 16.72
[raw statistics]
Elo: 117.62 ±1.9
Ptnml: 10, 562, 8016, 9195, 2217
nElo: 224.93 ±3.8
PairsRatio: 19.95
[raw statistics]
2020‑09‑02Stockfish 12 [[https://stockfishchess.org/images/logo/icon_128x128.png|width=20px]] [[https://github.githubassets.com/images/icons/emoji/unicode/1f4c8.png|width=20px]]
Bench: 3624569
[differences] 262
Stockfish 13 Development (02/09/20 - 18/02/21)
DateVersion1 Thread8 Threads
2020‑09‑08master vs Stockfish 12
Bench: 4161067
Double probability of using classical eval
[differences] 10
Elo: 6.49 ±1.0
Ptnml: 85, 3319, 22112, 4359, 125
nElo: 17.26 ±2.8
PairsRatio: 1.32
[raw statistics]
2020‑09‑21master vs Stockfish 12
Bench: 3973739
Add large page support for NNUE weights
[differences] 21
Elo: 13.88 ±1.0
Ptnml: 49, 2767, 22106, 4896, 182
nElo: 37.07 ±2.8
PairsRatio: 1.80
[raw statistics]
Elo: 8.97 ±1.1
Ptnml: 15, 1474, 16020, 2445, 46
nElo: 27.99 ±3.4
PairsRatio: 1.67
[raw statistics]
2020‑09‑28master vs Stockfish 12
Bench: 3776081
Include pawns in NNUE scaling
[differences] 33
Elo: 23.15 ±1.0
Ptnml: 38, 1945, 22217, 5587, 213
nElo: 63.29 ±2.7
PairsRatio: 2.92
[raw statistics]
2020‑10‑18master vs Stockfish 12
Bench: 4066972
Do more reductions for late quiet moves
[differences] 41
Elo: 24.09 ±1.0
Ptnml: 43, 2023, 21889, 5827, 218
nElo: 64.60 ±2.7
PairsRatio: 2.93
[raw statistics]
Elo: 21.17 ±1.1
Ptnml: 12, 908, 15789, 3216, 75
nElo: 65.43 ±3.2
PairsRatio: 3.58
[raw statistics]
2020‑11‑01master vs Stockfish 12
Bench: 3517795
Update default net to nn-cb26f10b1fd9.nnue
[differences] 48
Elo: 28.02 ±1.1
Ptnml: 63, 2079, 21107, 6469, 282
nElo: 71.59 ±2.7
PairsRatio: 3.15
[raw statistics]
2020‑11‑15master vs Stockfish 12
Bench: 3597730
Rook Mobility Tweak
[differences] 60
Elo: 29.99 ±1.0
Ptnml: 32, 1667, 21658, 6389, 254
nElo: 80.38 ±2.7
PairsRatio: 3.91
[raw statistics]
Elo: 25.49 ±1.1
Ptnml: 10, 799, 15532, 3570, 89
nElo: 77.27 ±3.2
PairsRatio: 4.52
[raw statistics]
2020‑11‑29master vs Stockfish 12
Bench: 3561701
Update default net to nn-62ef826d1a6d.nnue
[differences] 72
Elo: 30.61 ±1.0
Ptnml: 19, 1645, 21655, 6407, 274
nElo: 82.09 ±2.7
PairsRatio: 4.02
[raw statistics]
2020‑12‑14master vs Stockfish 12
Bench: 4050630
Increase reduction in case of stable best move
[differences] 79
Elo: 32.09 ±1.0
Ptnml: 33, 1581, 21474, 6651, 261
nElo: 85.57 ±2.7
PairsRatio: 4.28
[raw statistics]
Elo: 27.50 ±1.1
Ptnml: 10, 706, 15480, 3723, 81
nElo: 83.75 ±3.1
PairsRatio: 5.31
[raw statistics]
2020‑12‑31master vs Stockfish 12
Bench: 4109336
WeakUnopposed penalty for backwards
[differences] 89
Elo: 33.71 ±1.0
Ptnml: 23, 1475, 21475, 6730, 297
nElo: 90.01 ±2.6
PairsRatio: 4.69
[raw statistics]
2021‑01‑13master vs Stockfish 12
Bench: 4287509
Optimize generate_moves
[differences] 101
Elo: 32.40 ±1.0
Ptnml: 31, 1487, 21588, 6660, 234
nElo: 87.55 ±2.7
PairsRatio: 4.54
[raw statistics]
Elo: 28.29 ±1.1
Ptnml: 10, 671, 15468, 3761, 90
nElo: 86.06 ±3.1
PairsRatio: 5.65
[raw statistics]
2021‑02‑15master vs Stockfish 12
Bench: 3766422
Small trivial clean-ups, February 2021
[differences] 121
Elo: 36.03 ±1.1
Ptnml: 29, 1395, 21210, 7079, 287
nElo: 95.63 ±2.7
PairsRatio: 5.17
[raw statistics]
Elo: 29.08 ±1.1
Ptnml: 6, 626, 15476, 3806, 86
nElo: 89.08 ±3.1
PairsRatio: 6.16
[raw statistics]
2021‑02‑18Stockfish 13 [[https://stockfishchess.org/images/logo/icon_128x128.png|width=20px]] [[https://github.githubassets.com/images/icons/emoji/unicode/1f4c8.png|width=20px]]
Bench: 3766422
[differences] 123
Stockfish 14 Development (18/02/21 - 02/07/21)
DateVersion1 Thread8 Threads
2021‑02‑26master vs Stockfish 13
Bench: 5037279
Introduce DistanceFromPV
[differences] 4
Elo: 1.34 ±0.8
Ptnml: 28, 2367, 24980, 2596, 29
nElo: 4.55 ±2.8
PairsRatio: 1.10
[raw statistics]
2021‑03‑24master vs Stockfish 13
Bench: 4339126
Small cleanups (march 2021)
[differences] 18
Elo: 0.61 ±0.8
Ptnml: 18, 2457, 24938, 2575, 12
nElo: 2.09 ±2.8
PairsRatio: 1.05
[raw statistics]
2021‑04‑15master vs Stockfish 13
Bench: 4503918
Use classical eval for Bishop vs Pawns
[differences] 29
Elo: 1.47 ±0.8
Ptnml: 39, 2384, 24886, 2667, 24
nElo: 4.93 ±2.8
PairsRatio: 1.11
[raw statistics]
Elo: 0.89 ±0.9
Ptnml: 11, 1279, 17318, 1381, 11
nElo: 3.38 ±3.4
PairsRatio: 1.08
[raw statistics]
2021‑05‑22master vs Stockfish 13
Bench: 3856635
Sometimes change the balance
[differences] 62
Elo: 10.12 ±1.0
Ptnml: 60, 2454, 23277, 4096, 113
nElo: 29.34 ±2.8
PairsRatio: 1.67
[raw statistics]
2021‑06‑14master vs Stockfish 13
Bench: 4877339
Update default net to nn-8e47cf062333.nnue
[differences] 90
Elo: 21.80 ±1.1
Ptnml: 67, 2216, 21852, 5620, 245
nElo: 57.47 ±2.7
PairsRatio: 2.57
[raw statistics]
Elo: 16.96 ±1.1
Ptnml: 15, 1083, 15923, 2894, 85
nElo: 52.38 ±3.3
PairsRatio: 2.71
[raw statistics]
2021‑06‑18master vs Stockfish 13
Bench: 4900906
Make net nn-50144f835024.nnue the default
[differences] 100
Elo: 25.56 ±1.0
Ptnml: 30, 1814, 22084, 5864, 208
nElo: 69.95 ±2.7
PairsRatio: 3.29
[raw statistics]
2021‑06‑29master vs Stockfish 13
Bench: 4770936
Update Top CPU Contributors
[differences] 113
Elo: 30.27 ±1.0
Ptnml: 14, 1443, 22127, 6146, 270
nElo: 83.66 ±2.6
PairsRatio: 4.40
[raw statistics]
Elo: 22.62 ±1.1
Ptnml: 7, 755, 15977, 3153, 108
nElo: 71.15 ±3.1
PairsRatio: 4.28
[raw statistics]
2021‑07‑02Stockfish 14 [[https://stockfishchess.org/images/logo/icon_128x128.png|width=20px]] [[https://github.githubassets.com/images/icons/emoji/unicode/1f4c8.png|width=20px]]
Bench: 4770936
[differences] 114
Stockfish 15 Development (02/07/21 - 18/04/22)
DateVersion1 Thread8 Threads1 Thread (UHO)
2021‑07‑26master vs Stockfish 14
Bench: 5124774
Update default net to nn-26abeed38351.nnue
[differences] 17
Elo: 2.74 ±0.8
Ptnml: 24, 2000, 25476, 2478, 22
nElo: 9.85 ±2.8
PairsRatio: 1.24
[raw statistics]
2021‑08‑15master vs Stockfish 14
Bench: 5189338
New NNUE architecture and net
[differences] 26
Elo: 9.31 ±0.9
Ptnml: 24, 2044, 24321, 3522, 89
nElo: 29.61 ±2.7
PairsRatio: 1.75
[raw statistics]
Elo: 6.08 ±0.9
Ptnml: 5, 1063, 17183, 1725, 24
nElo: 22.66 ±3.4
PairsRatio: 1.64
[raw statistics]
2021‑08‑31master vs Stockfish 14
Bench: 5600615
Update default net to nn-735bba95dec0.nnue
[differences] 39
Elo: 15.04 ±0.9
Ptnml: 17, 1601, 24272, 3990, 120
nElo: 47.86 ±2.7
PairsRatio: 2.54
[raw statistics]
2021‑09‑15master vs Stockfish 14
Bench: 6658747
Update default net to nn-13406b1dcbe0.nnue
[differences] 46
Elo: 16.64 ±0.9
Ptnml: 21, 1878, 23452, 4506, 143
nElo: 49.53 ±2.7
PairsRatio: 2.45
[raw statistics]
Elo: 12.39 ±0.9
Ptnml: 1, 766, 17084, 2104, 45
nElo: 45.59 ±3.2
PairsRatio: 2.80
[raw statistics]
2021‑10‑06master vs Stockfish 14
Bench: 6261865
Capping stat bonus at 2000
[differences] 57
Elo: 18.68 ±0.9
Ptnml: 28, 1496, 23821, 4535, 120
nElo: 57.70 ±2.7
PairsRatio: 3.05
[raw statistics]
Elo: 11.74 ±0.9
Ptnml: 4, 769, 17137, 2052, 38
nElo: 43.60 ±3.2
PairsRatio: 2.70
[raw statistics]
2021‑10‑18master vs Stockfish 14
Bench: 5005810
Simplify probCutCount away
[differences] 67
Elo: 17.69 ±0.9
Ptnml: 13, 1577, 23891, 4382, 137
nElo: 54.78 ±2.7
PairsRatio: 2.84
[raw statistics]
2021‑10‑23master vs Stockfish 14
Bench: 6334068
Adjust ButterflyHistory decay parameter
[differences] 78
Elo: 17.87 ±0.9
Ptnml: 11, 1377, 24244, 4253, 115
nElo: 57.35 ±2.6
PairsRatio: 3.15
[raw statistics]
Elo: 10.27 ±0.9
Ptnml: 3, 696, 17438, 1842, 21
nElo: 40.55 ±3.2
PairsRatio: 2.67
[raw statistics]
2021‑10‑28Stockfish 14.1 [[https://stockfishchess.org/images/logo/icon_128x128.png|width=20px]]
Bench: 6334068
[differences] 80
2021‑11‑05master vs Stockfish 14
Bench: 6719976
Tweak initial aspiration window
[differences] 89
Elo: 18.71 ±0.9
Ptnml: 16, 1293, 24255, 4319, 117
nElo: 60.14 ±2.6
PairsRatio: 3.39
[raw statistics]
2021‑11‑23master vs Stockfish 14
Bench: 7334766
Less futility pruning
[differences] 98
Elo: 19.35 ±0.9
Ptnml: 19, 1405, 23923, 4524, 129
nElo: 60.39 ±2.6
PairsRatio: 3.27
[raw statistics]
Elo: 13.28 ±0.9
Ptnml: 3, 696, 17112, 2148, 41
nElo: 49.26 ±3.2
PairsRatio: 3.13
[raw statistics]
2021‑11‑28master vs Stockfish 14
Bench: 6302543
Refine futility pruning for parent nodes
[differences] 103
Elo: 24.44 ±0.9
Ptnml: 9, 1054, 23776, 5037, 124
nElo: 76.95 ±2.5
PairsRatio: 4.86
[raw statistics]
2021‑12‑07master vs Stockfish 14
Bench: 4667742
Update default net to nn-63376713ba63.nnue
[differences] 118
Elo: 26.99 ±0.9
Ptnml: 11, 998, 23473, 5365, 153
nElo: 83.13 ±2.5
PairsRatio: 5.47
[raw statistics]
Elo: 17.80 ±0.9
Ptnml: 6, 529, 16919, 2504, 42
nElo: 64.76 ±3.1
PairsRatio: 4.76
[raw statistics]
2021‑12‑14master vs Stockfish 14
Bench: 4735679
Remove NNUE scaling term
[differences] 126
Elo: 26.45 ±0.9
Ptnml: 9, 962, 23656, 5207, 166
nElo: 82.28 ±2.5
PairsRatio: 5.53
[raw statistics]
2021‑12‑22master vs Stockfish 14
Bench: 4633875
Update default net to nn-ac07bd334b62.nnue
[differences] 139
Elo: 28.65 ±0.9
Ptnml: 11, 1036, 23172, 5567, 214
nElo: 85.61 ±2.5
PairsRatio: 5.52
[raw statistics]
Elo: 17.66 ±0.9
Ptnml: 2, 520, 16976, 2448, 54
nElo: 64.60 ±3.0
PairsRatio: 4.79
[raw statistics]
2022‑01‑10master vs Stockfish 14
Bench: 4572746
Adjust pruning constants
[differences] 148
Elo: 30.51 ±1.0
Ptnml: 13, 1065, 22809, 5879, 234
nElo: 89.03 ±2.5
PairsRatio: 5.67
[raw statistics]
2022‑01‑29master vs Stockfish 14
Bench: 4637392
Do stats updates after LMR for captures
[differences] 159
Elo: 32.07 ±1.0
Ptnml: 10, 1009, 22681, 6049, 251
nElo: 93.05 ±2.5
PairsRatio: 6.18
[raw statistics]
Elo: 22.31 ±1.0
Ptnml: 3, 496, 16495, 2945, 61
nElo: 76.78 ±3.0
PairsRatio: 6.02
[raw statistics]
2022‑02‑10master vs Stockfish 14
Bench: 4919707
Update architecture to "SFNNv4"
[differences] 166
Elo: 34.88 ±1.0
Ptnml: 13, 884, 22430, 6432, 241
nElo: 100.85 ±2.5
PairsRatio: 7.44
[raw statistics]
Elo: 24.37 ±1.0
Ptnml: 2, 416, 16441, 3061, 80
nElo: 83.29 ±2.9
PairsRatio: 7.51
[raw statistics]
2022‑02‑17master vs Stockfish 14
Bench: 6318903
Tune search at very long time control
[differences] 168
Elo: 33.07 ±1.0
Ptnml: 17, 984, 22522, 6243, 234
nElo: 95.49 ±2.5
PairsRatio: 6.47
[raw statistics]
Elo: 23.83 ±1.0
Ptnml: 4, 410, 16496, 3023, 67
nElo: 82.31 ±2.9
PairsRatio: 7.46
[raw statistics]
2022‑03‑19master vs Stockfish 14
Bench: 7044203
Remove ttPv tree shrinking
[differences] 180
Elo: 35.01 ±1.0
Ptnml: 11, 801, 22566, 6395, 227
nElo: 102.62 ±2.5
PairsRatio: 8.16
[raw statistics]
2022‑04‑17master vs Stockfish 14
Bench: 8129754
Decrease LMR at PV nodes with low depth
[differences] 189
Elo: 36.69 ±1.0
Ptnml: 5, 711, 22489, 6557, 238
nElo: 107.69 ±2.4
PairsRatio: 9.49
[raw statistics]
Elo: 26.09 ±1.0
Ptnml: 1, 341, 16379, 3217, 62
nElo: 89.84 ±2.8
PairsRatio: 9.59
[raw statistics]
Elo: 91.22 ±1.3
Ptnml: 17, 1335, 12388, 15750, 510
nElo: 205.62 ±3.6
PairsRatio: 12.03
[raw statistics]
2022‑04‑18Stockfish 15 [[https://stockfishchess.org/images/logo/icon_128x128.png|width=20px]] [[https://github.githubassets.com/images/icons/emoji/unicode/1f4c8.png|width=20px]]
Bench: 8129754
[differences] 190
Stockfish 16 Development (18/04/22 - 29/06/23)
DateVersion1 Thread8 Threads1 Thread (UHO)
2022‑05‑14master vs Stockfish 15
Bench: 6481017
SE depth scaling using the previous depth
[differences] 9
Elo: 4.54 ±0.8
Ptnml: 18, 1850, 25507, 2580, 45
nElo: 16.29 ±2.8
PairsRatio: 1.41
[raw statistics]
Elo: 3.12 ±0.8
Ptnml: 5, 803, 18029, 1154, 9
nElo: 13.92 ±3.4
PairsRatio: 1.44
[raw statistics]
2022‑06‑16master vs Stockfish 15
Bench: 5845802
Simplify away condition in ttSave in probCut
[differences] 25
Elo: 5.72 ±0.8
Ptnml: 31, 1838, 25288, 2799, 44
nElo: 19.99 ±2.8
PairsRatio: 1.52
[raw statistics]
2022‑07‑13master vs Stockfish 15
Bench: 5905619
Update default net to nn-ad9b42354671.nnue
[differences] 34
Elo: 5.70 ±0.8
Ptnml: 42, 1867, 25207, 2832, 52
nElo: 19.67 ±2.8
PairsRatio: 1.51
[raw statistics]
Elo: 4.86 ±0.8
Ptnml: 3, 734, 17975, 1276, 12
nElo: 21.46 ±3.3
PairsRatio: 1.75
[raw statistics]
2022‑08‑12master vs Stockfish 15
Bench: 5868987
Remove an unneeded randomization of evals
[differences] 43
Elo: 7.18 ±0.8
Ptnml: 48, 1784, 25112, 2993, 63
nElo: 24.44 ±2.8
PairsRatio: 1.67
[raw statistics]
Elo: 4.60 ±0.8
Ptnml: 5, 737, 17987, 1265, 6
nElo: 20.43 ±3.4
PairsRatio: 1.71
[raw statistics]
Elo: 14.61 ±1.3
Ptnml: 149, 5043, 17073, 7608, 127
nElo: 30.73 ±2.8
PairsRatio: 1.49
[raw statistics]
2022‑09‑07master vs Stockfish 15
Bench: 5609606
VLTC tuning
[differences] 52
Elo: 6.87 ±0.8
Ptnml: 40, 1796, 25154, 2958, 52
nElo: 23.61 ±2.8
PairsRatio: 1.64
[raw statistics]
Elo: 5.45 ±0.8
Ptnml: 3, 732, 17913, 1339, 13
nElo: 23.68 ±3.3
PairsRatio: 1.84
[raw statistics]
2022‑10‑05master vs Stockfish 15
Bench: 4114228
Revert "Mix alpha and statScore for reduction"
[differences] 66
Elo: 6.86 ±0.8
Ptnml: 21, 1580, 25632, 2727, 40
nElo: 25.04 ±2.7
PairsRatio: 1.73
[raw statistics]
Elo: 5.98 ±0.8
Ptnml: 6, 643, 18015, 1329, 7
nElo: 26.72 ±3.3
PairsRatio: 2.06
[raw statistics]
2022‑10‑30master vs Stockfish 15
Bench: 4271738
Adjust reduction less at medium depths
[differences] 81
Elo: 8.52 ±0.8
Ptnml: 23, 1469, 25573, 2884, 51
nElo: 30.84 ±2.7
PairsRatio: 1.97
[raw statistics]
Elo: 5.91 ±0.7
Ptnml: 6, 589, 18129, 1271, 5
nElo: 27.24 ±3.3
PairsRatio: 2.14
[raw statistics]
Elo: 17.17 ±1.8
Ptnml: 38, 2430, 8572, 3933, 27
nElo: 36.91 ±4.0
PairsRatio: 1.60
[raw statistics]
2022‑12‑02master vs Stockfish 15
Bench: 3467381
Fix bestThread selection
[differences] 97
Elo: 7.46 ±0.7
Ptnml: 21, 1389, 25902, 2657, 31
nElo: 28.19 ±2.7
PairsRatio: 1.91
[raw statistics]
Elo: 5.97 ±0.7
Ptnml: 3, 573, 18164, 1254, 6
nElo: 27.83 ±3.3
PairsRatio: 2.19
[raw statistics]
Elo: 17.04 ±1.3
Ptnml: 73, 4844, 17208, 7820, 55
nElo: 36.73 ±2.8
PairsRatio: 1.60
[raw statistics]
2022‑12‑04Stockfish 15.1 [[https://stockfishchess.org/images/logo/icon_128x128.png|width=20px]]
Bench: 3467381
[differences] 98
2022‑12‑19master vs Stockfish 15
Bench: 3727508
Sometimes do a reduced search if LMR is skipped
[differences] 119
Elo: 8.55 ±0.8
Ptnml: 16, 1420, 25675, 2850, 39
nElo: 31.50 ±2.7
PairsRatio: 2.01
[raw statistics]
2023‑01‑01master vs Stockfish 15
Bench: 4015511
Update default net to nn-60fa44e376d9.nnue
[differences] 126
Elo: 10.00 ±0.8
Ptnml: 15, 1301, 25667, 2976, 41
nElo: 36.92 ±2.7
PairsRatio: 2.29
[raw statistics]
Elo: 7.10 ±0.7
Ptnml: 6, 529, 18118, 1336, 11
nElo: 32.56 ±3.2
PairsRatio: 2.52
[raw statistics]
Elo: 23.55 ±1.3
Ptnml: 59, 4354, 17132, 8377, 78
nElo: 51.05 ±2.8
PairsRatio: 1.92
[raw statistics]
2023‑01‑23master vs Stockfish 15
Bench: 3941848
Update default net to nn-bc24c101ada0.nnue
[differences] 143
Elo: 11.37 ±0.7
Ptnml: 9, 1127, 25786, 3048, 30
nElo: 42.95 ±2.7
PairsRatio: 2.71
[raw statistics]
Elo: 9.09 ±0.8
Ptnml: 2, 484, 17984, 1526, 4
nElo: 40.84 ±3.2
PairsRatio: 3.15
[raw statistics]
Elo: 29.49 ±1.3
Ptnml: 43, 3893, 17099, 8870, 95
nElo: 64.58 ±2.9
PairsRatio: 2.28
[raw statistics]
2023‑02‑09master vs Stockfish 15
Bench: 3841998
Update default net to nn-1337b1adec5b.nnue
[differences] 162
Elo: 12.76 ±0.8
Ptnml: 9, 1089, 25635, 3224, 43
nElo: 47.33 ±2.6
PairsRatio: 2.98
[raw statistics]
Elo: 8.49 ±0.8
Ptnml: 2, 502, 18018, 1473, 5
nElo: 38.38 ±3.2
PairsRatio: 2.93
[raw statistics]
Elo: 32.71 ±1.3
Ptnml: 44, 3760, 16821, 9269, 106
nElo: 71.26 ±2.9
PairsRatio: 2.46
[raw statistics]
2023‑02‑18master vs Stockfish 15
Bench: 4283297
Remove one reduction call
[differences] 168
Elo: 12.58 ±0.8
Ptnml: 11, 1127, 25591, 3221, 50
nElo: 46.26 ±2.6
PairsRatio: 2.87
[raw statistics]
Elo: 31.91 ±1.3
Ptnml: 47, 3828, 16804, 9225, 96
nElo: 69.38 ±2.9
PairsRatio: 2.41
[raw statistics]
2023‑02‑24master vs Stockfish 15
Bench: 4705194
Search tuning at very long time control
[differences] 174
Elo: 11.69 ±0.7
Ptnml: 10, 1151, 25692, 3105, 42
nElo: 43.51 ±2.6
PairsRatio: 2.71
[raw statistics]
Elo: 9.37 ±0.8
Ptnml: 1, 455, 18013, 1526, 5
nElo: 42.48 ±3.1
PairsRatio: 3.36
[raw statistics]
Elo: 32.42 ±1.3
Ptnml: 44, 3804, 16761, 9307, 84
nElo: 70.59 ±2.9
PairsRatio: 2.44
[raw statistics]
2023‑03‑19master vs Stockfish 15
Bench: 4980082
Remove 'si' StateInfo variable/parameter.
[differences] 196
Elo: 13.36 ±0.8
Ptnml: 9, 1098, 25532, 3300, 61
nElo: 48.75 ±2.6
PairsRatio: 3.04
[raw statistics]
Elo: 10.10 ±0.8
Ptnml: 0, 460, 17934, 1589, 17
nElo: 44.63 ±3.1
PairsRatio: 3.49
[raw statistics]
Elo: 35.27 ±1.3
Ptnml: 48, 3610, 16657, 9594, 91
nElo: 76.91 ±2.9
PairsRatio: 2.65
[raw statistics]
2023‑04‑01master vs Stockfish 15
Bench: 4380438
Decrease Depth more for positions not in TT.
[differences] 211
Elo: 14.42 ±0.8
Ptnml: 7, 1031, 25480, 3431, 51
nElo: 52.68 ±2.6
PairsRatio: 3.35
[raw statistics]
Elo: 9.90 ±0.8
Ptnml: 2, 450, 17964, 1575, 9
nElo: 44.20 ±3.1
PairsRatio: 3.50
[raw statistics]
Elo: 37.37 ±1.3
Ptnml: 48, 3485, 16555, 9815, 97
nElo: 81.54 ±2.9
PairsRatio: 2.81
[raw statistics]
2023‑04‑22master vs Stockfish 15
Bench: 3548023
Less reduction for tt move.
[differences] 235
Elo: 14.60 ±0.8
Ptnml: 9, 967, 25569, 3405, 50
nElo: 53.91 ±2.6
PairsRatio: 3.54
[raw statistics]
Elo: 11.32 ±0.8
Ptnml: 0, 386, 17939, 1661, 14
nElo: 50.39 ±3.0
PairsRatio: 4.34
[raw statistics]
Elo: 38.07 ±1.3
Ptnml: 27, 3392, 16687, 9793, 101
nElo: 83.87 ±2.9
PairsRatio: 2.89
[raw statistics]
2023‑05‑07master vs Stockfish 15
Bench: 3808503
Refine deeper post-lmr searches
[differences] 244
Elo: 14.36 ±0.8
Ptnml: 11, 1004, 25521, 3423, 41
nElo: 52.83 ±2.6
PairsRatio: 3.41
[raw statistics]
Elo: 10.63 ±0.7
Ptnml: 0, 379, 18029, 1581, 11
nElo: 48.41 ±3.0
PairsRatio: 4.20
[raw statistics]
Elo: 39.62 ±1.3
Ptnml: 34, 3309, 16568, 9988, 101
nElo: 87.17 ±2.9
PairsRatio: 3.02
[raw statistics]
2023‑06‑04master vs Stockfish 15
Bench: 2551691
Move internal iterative reduction before probcut
[differences] 265
Elo: 15.01 ±0.8
Ptnml: 9, 1037, 25355, 3552, 47
nElo: 54.25 ±2.6
PairsRatio: 3.44
[raw statistics]
Elo: 11.87 ±0.8
Ptnml: 3, 362, 17908, 1720, 7
nElo: 52.68 ±3.0
PairsRatio: 4.73
[raw statistics]
Elo: 38.10 ±1.3
Ptnml: 48, 3387, 16634, 9826, 105
nElo: 83.49 ±2.9
PairsRatio: 2.89
[raw statistics]
2023‑06‑12master vs Stockfish 15
Bench: 2370027
Use block sparse input for the first layer.
[differences] 274
Elo: 17.57 ±0.8
Ptnml: 10, 881, 25223, 3840, 46
nElo: 63.12 ±2.5
PairsRatio: 4.36
[raw statistics]
Elo: 13.95 ±0.8
Ptnml: 1, 290, 17828, 1865, 16
nElo: 60.93 ±2.8
PairsRatio: 6.46
[raw statistics]
Elo: 44.18 ±1.3
Ptnml: 45, 3031, 16345, 10449, 130
nElo: 97.20 ±3.0
PairsRatio: 3.44
[raw statistics]
2023‑06‑22master vs Stockfish 15
Bench: 2593605
Update default net to nn-5af11540bbfe.nnue
[differences] 289
Elo: 18.30 ±0.8
Ptnml: 3, 812, 25265, 3864, 56
nElo: 66.19 ±2.5
PairsRatio: 4.81
[raw statistics]
Elo: 14.33 ±0.8
Ptnml: 3, 297, 17767, 1914, 19
nElo: 61.59 ±2.9
PairsRatio: 6.44
[raw statistics]
Elo: 47.03 ±1.3
Ptnml: 26, 2938, 16102, 10805, 129
nElo: 103.71 ±3.0
PairsRatio: 3.69
[raw statistics]
2023‑06‑29Stockfish 16 [[https://stockfishchess.org/images/logo/icon_128x128.png|width=20px]] [[https://github.githubassets.com/images/icons/emoji/unicode/1f4c8.png|width=20px]]
Bench: 2593605
[differences] 290

There are several pages on the web run by chess engine fans. Some collect progress of Stockfish over previous stable versions and development builds. Some compare Stockfish progress to other chess engines. Here is a collection of some useful links in that regard.