Dev Builds » 20170101-0956

Use this dev build

NCM plays each Stockfish dev build 20,000 times against Stockfish 7. This yields an approximate Elo difference and establishes confidence in the strength of the dev builds.

Summary

Host Duration Avg Base NPS Games Wins Losses Draws Elo
ncm-et-3 06:08:42 1986219 2461 746 102 1613 +93.08 +/- 7.74
ncm-et-4 06:08:45 1978276 2477 757 116 1604 +92.0 +/- 7.84
ncm-et-5 01:45:18 2024263 716 208 27 481 +89.78 +/- 13.99
ncm-et-6 01:45:07 2026803 724 195 37 492 +77.06 +/- 13.89
ncm-et-7 01:45:06 1977447 721 229 27 465 +100.01 +/- 14.48
ncm-et-8 01:45:14 1979507 703 213 36 454 +89.4 +/- 14.8
ncm-et-9 06:08:40 1984627 2487 761 103 1623 +94.16 +/- 7.73
ncm-et-10 06:08:54 1984898 2472 723 115 1634 +87.24 +/- 7.69
ncm-et-11 01:45:05 2018965 718 204 28 486 +86.94 +/- 13.88
ncm-et-12 01:44:36 1989094 699 195 42 462 +77.3 +/- 14.61
ncm-et-13 06:08:36 1994135 2514 761 104 1649 +92.95 +/- 7.66
ncm-et-14 01:45:14 2018316 710 210 22 478 +94.24 +/- 13.93
ncm-et-15 05:07:41 1810295 1890 546 94 1250 +84.73 +/- 8.82
ncm-et-16 01:44:34 2000549 708 213 30 465 +91.89 +/- 14.43
20000 5961 883 13156 +90.19 +/- 2.71

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
63607 ncm-et-15 2018-12-23 15:17 00:33:26 1555087 177 44 9 124 +69.62 +/- 27.24
63606 ncm-et-3 2018-12-23 15:14 00:36:58 1963378 248 83 9 156 +106.92 +/- 25.26
63605 ncm-et-4 2018-12-23 15:11 00:39:28 1930839 256 82 11 163 +98.95 +/- 24.72
63604 ncm-et-9 2018-12-23 15:11 00:39:51 1968016 270 88 10 172 +103.31 +/- 23.95
63603 ncm-et-10 2018-12-23 15:10 00:40:47 1967547 265 93 7 165 +116.98 +/- 24.44
63602 ncm-et-13 2018-12-23 15:08 00:42:38 1986282 284 76 10 198 +82.24 +/- 21.34
63601 ncm-et-3 2018-12-23 13:58 01:14:27 1963688 500 142 22 336 +85.04 +/- 16.82
63600 ncm-et-9 2018-12-23 13:55 01:14:51 1960462 500 150 21 329 +91.71 +/- 17.14
63599 ncm-et-10 2018-12-23 13:55 01:14:04 1962787 500 151 23 326 +90.97 +/- 17.34
63598 ncm-et-4 2018-12-23 13:54 01:16:16 1964917 500 148 25 327 +87.26 +/- 17.34
63597 ncm-et-13 2018-12-23 13:54 01:13:18 1974684 500 152 26 322 +89.48 +/- 17.6
63596 ncm-et-15 2018-12-23 13:44 01:31:30 1550662 500 148 19 333 +91.71 +/- 16.89
63595 ncm-et-3 2018-12-23 12:42 01:14:33 1976217 500 149 25 326 +88.0 +/- 17.39
63594 ncm-et-10 2018-12-23 12:40 01:13:17 1977689 500 147 25 328 +86.52 +/- 17.29
63593 ncm-et-4 2018-12-23 12:39 01:13:26 1963686 500 156 21 323 +96.19 +/- 17.43
63592 ncm-et-13 2018-12-23 12:39 01:13:33 1973897 500 153 23 324 +92.46 +/- 17.44
63591 ncm-et-9 2018-12-23 12:39 01:14:56 1966017 500 159 23 318 +96.94 +/- 17.73
63590 ncm-et-15 2018-12-23 12:26 01:17:43 1911212 500 161 28 311 +94.7 +/- 18.19
63589 ncm-et-13 2018-12-23 11:24 01:14:00 1980443 500 158 16 326 +101.46 +/- 17.14
63588 ncm-et-9 2018-12-23 11:24 01:13:54 1965232 500 156 15 329 +100.71 +/- 16.97
63587 ncm-et-3 2018-12-23 11:24 01:17:38 1963382 500 142 23 335 +84.31 +/- 16.89
63586 ncm-et-10 2018-12-23 11:24 01:15:39 1960019 500 148 30 322 +83.57 +/- 17.69
63585 ncm-et-4 2018-12-23 11:24 01:14:28 1961689 500 145 22 333 +87.26 +/- 16.97
5620 ncm-et-12 2018-03-06 21:40 00:30:28 1988783 199 54 19 126 +61.75 +/- 28.94
5619 ncm-et-8 2018-03-06 21:39 00:31:08 1943300 203 52 5 146 +81.93 +/- 24.12
5618 ncm-et-14 2018-03-06 21:39 00:31:11 2018883 210 58 8 144 +84.34 +/- 25.34
5617 ncm-et-10 2018-03-06 21:39 00:31:10 2020021 207 51 8 148 +73.24 +/- 24.37
5616 ncm-et-9 2018-03-06 21:39 00:31:31 2024263 217 58 10 149 +78.14 +/- 25.04
5615 ncm-et-15 2018-03-06 21:39 00:31:23 2018070 213 66 16 131 +83.11 +/- 28.43
5614 ncm-et-16 2018-03-06 21:39 00:31:16 1979446 208 66 10 132 +95.9 +/- 27.62
5613 ncm-et-3 2018-03-06 21:39 00:31:45 2026226 213 72 5 136 +113.12 +/- 26.6
5612 ncm-et-4 2018-03-06 21:38 00:32:10 2023936 221 91 11 119 +131.74 +/- 30.41
5611 ncm-et-6 2018-03-06 21:38 00:32:04 2027379 224 55 14 155 +64.32 +/- 24.71
5610 ncm-et-5 2018-03-06 21:38 00:32:14 2024755 216 68 5 143 +104.37 +/- 25.52
5609 ncm-et-11 2018-03-06 21:38 00:32:19 2019209 218 66 7 145 +96.43 +/- 25.51
5608 ncm-et-7 2018-03-06 21:38 00:32:17 1953482 221 79 7 135 +117.47 +/- 27.34
5607 ncm-et-13 2018-03-06 21:37 00:33:05 2024263 230 73 8 149 +100.94 +/- 25.52
5606 ncm-et-16 2018-03-06 20:24 01:13:18 2021652 500 147 20 333 +90.22 +/- 16.92
5605 ncm-et-4 2018-03-06 20:24 01:12:57 2024590 500 135 26 339 +76.98 +/- 16.77
5604 ncm-et-11 2018-03-06 20:24 01:12:46 2018721 500 138 21 341 +82.83 +/- 16.54
5603 ncm-et-12 2018-03-06 20:24 01:14:08 1989406 500 141 23 336 +83.57 +/- 16.84
5602 ncm-et-9 2018-03-06 20:24 01:13:37 2023773 500 150 24 326 +89.48 +/- 17.36
5601 ncm-et-7 2018-03-06 20:24 01:12:49 2001412 500 150 20 330 +92.46 +/- 17.06
5600 ncm-et-10 2018-03-06 20:24 01:13:57 2021325 500 133 22 345 +78.44 +/- 16.37
5599 ncm-et-3 2018-03-06 20:24 01:13:21 2024427 500 158 18 324 +99.95 +/- 17.3
5598 ncm-et-6 2018-03-06 20:24 01:13:03 2026227 500 140 23 337 +82.83 +/- 16.79
5597 ncm-et-13 2018-03-06 20:24 01:12:02 2025245 500 149 21 330 +90.97 +/- 17.09
5596 ncm-et-15 2018-03-06 20:24 01:13:39 2016447 500 127 22 351 +74.06 +/- 16.07
5595 ncm-et-14 2018-03-06 20:24 01:14:03 2017749 500 152 14 334 +98.44 +/- 16.69
5594 ncm-et-5 2018-03-06 20:24 01:13:04 2023772 500 140 22 338 +83.57 +/- 16.72
5593 ncm-et-8 2018-03-06 20:24 01:14:06 2015714 500 161 31 308 +92.46 +/- 18.39

Commit

Commit ID 881a9dfb0a8fec3b1472791e2d98415e4a9a182a
Author Sergei Antonov
Date 2017-01-01 09:56:46 UTC
Threefold repetition detection Implement a threefold repetition detection. Below are the examples of problems fixed by this change. Loosing move in a drawn position. position fen 8/k7/3p4/p2P1p2/P2P1P2/8/8/K7 w - - 0 1 moves a1a2 a7a8 a2a1 The old code suggested a loosing move "bestmove a8a7", the new code suggests "bestmove a8b7" leading to a draw. Incorrect evaluation (happened in a real game in TCEC Season 9). position fen 4rbkr/1q3pp1/b3pn2/7p/1pN5/1P1BBP1P/P1R2QP1/3R2K1 w - - 5 31 moves e3d4 h8h6 d4e3 The old code evaluated it as "cp 0", the new code evaluation is around "cp -50" which is adequate. Brings 0.5-1 ELO gain. Passes [-3.00,1.00]. STC: http://tests.stockfishchess.org/tests/view/584ece040ebc5903140c5aea LLR: 2.96 (-2.94,2.94) [-3.00,1.00] Total: 47744 W: 8537 L: 8461 D: 30746 LTC: http://tests.stockfishchess.org/tests/view/584f134d0ebc5903140c5b37 LLR: 2.96 (-2.94,2.94) [-3.00,1.00] Total: 36775 W: 4739 L: 4639 D: 27397 Patch has been rewritten into current form for simplification and logic slightly changed so that return a draw score if the position repeats once earlier but after or at the root, or repeats twice strictly before the root. In its original form, repetition at root was not returned as an immediate draw. After retestimng testing both version with SPRT[-3, 1], both passed succesfully, but this version was chosen becuase more natural. There is an argument about MultiPV in which an extended draw at root may be sensible. See discussion here: https://github.com/official-stockfish/Stockfish/pull/925 For documentation, current version passed both at STC and LTC: STC LLR: 2.96 (-2.94,2.94) [-3.00,1.00] Total: 51562 W: 9314 L: 9245 D: 33003 LTC LLR: 2.96 (-2.94,2.94) [-3.00,1.00] Total: 115663 W: 14904 L: 14906 D: 85853 bench: 5468995
Copyright 2011–2024 Next Chess Move LLC