Dev Builds » 20180513-0520

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 06:04:40 2001895 2399 1132 59 1208 +167.21 +/- 9.47
ncm-et-4 06:04:42 1990444 2405 1111 67 1227 +161.53 +/- 9.4
ncm-et-5 01:47:55 2020029 699 339 15 345 +174.34 +/- 17.74
ncm-et-6 01:48:09 2026554 719 341 17 361 +168.67 +/- 17.33
ncm-et-7 01:47:50 2025899 729 334 24 371 +157.76 +/- 17.17
ncm-et-8 01:47:42 2006754 707 344 22 341 +170.78 +/- 18.02
ncm-et-9 06:04:15 2000477 2393 1139 68 1186 +167.34 +/- 9.6
ncm-et-10 06:04:26 1980120 2387 1146 72 1169 +168.37 +/- 9.7
ncm-et-11 01:48:04 2019939 712 323 19 370 +158.49 +/- 17.06
ncm-et-12 01:47:52 2011043 709 344 18 347 +172.7 +/- 17.76
ncm-et-13 06:04:34 1964763 2346 1089 85 1172 +158.92 +/- 9.7
ncm-et-14 01:47:55 2021651 724 333 20 371 +160.77 +/- 17.09
ncm-et-15 06:04:21 1976292 2366 1135 69 1162 +168.64 +/- 9.72
ncm-et-16 01:47:57 2020837 705 319 22 364 +156.08 +/- 17.28
20000 9429 577 9994 +165.19 +/- 3.3

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
50676 ncm-et-13 2018-08-21 18:16 00:20:38 1926985 121 58 6 57 +159.68 +/- 44.94
50675 ncm-et-15 2018-08-21 18:11 00:25:49 1939641 157 70 6 81 +150.36 +/- 36.96
50674 ncm-et-9 2018-08-21 18:10 00:26:29 1985776 179 82 6 91 +157.48 +/- 34.82
50673 ncm-et-10 2018-08-21 18:09 00:28:01 1935020 167 83 4 80 +178.58 +/- 37.24
50672 ncm-et-4 2018-08-21 18:08 00:28:47 1989401 186 85 1 100 +169.11 +/- 32.03
50671 ncm-et-3 2018-08-21 18:07 00:29:51 1989402 190 82 4 104 +151.57 +/- 31.8
50670 ncm-et-13 2018-08-21 16:57 01:17:45 1958613 500 256 18 226 +179.9 +/- 22.38
50669 ncm-et-15 2018-08-21 16:54 01:16:02 1982165 500 235 13 252 +165.8 +/- 20.78
50668 ncm-et-9 2018-08-21 16:54 01:15:41 1989401 500 233 17 250 +160.64 +/- 21.0
50667 ncm-et-10 2018-08-21 16:52 01:16:13 1980911 500 254 14 232 +181.7 +/- 21.92
50666 ncm-et-3 2018-08-21 16:51 01:15:01 1992088 500 240 12 248 +171.02 +/- 20.96
50665 ncm-et-4 2018-08-21 16:51 01:16:38 1988769 500 235 21 244 +158.93 +/- 21.42
50664 ncm-et-13 2018-08-21 15:36 01:20:08 1886979 500 212 14 274 +145.54 +/- 19.65
50663 ncm-et-9 2018-08-21 15:36 01:16:53 1988296 500 242 8 250 +176.33 +/- 20.71
50662 ncm-et-15 2018-08-21 15:34 01:18:52 1946476 500 232 16 252 +160.64 +/- 20.86
50661 ncm-et-10 2018-08-21 15:34 01:17:01 1941759 500 242 16 242 +169.27 +/- 21.41
50660 ncm-et-4 2018-08-21 15:34 01:15:56 1989401 500 235 14 251 +164.93 +/- 20.86
50659 ncm-et-3 2018-08-21 15:33 01:16:56 1990508 500 234 17 249 +161.49 +/- 21.05
50658 ncm-et-15 2018-08-21 14:17 01:15:51 1953329 500 257 13 230 +185.33 +/- 22.01
50657 ncm-et-9 2018-08-21 14:17 01:17:21 1989558 500 233 16 251 +161.49 +/- 20.92
50656 ncm-et-13 2018-08-21 14:17 01:18:01 1964697 500 243 14 243 +171.9 +/- 21.3
50655 ncm-et-10 2018-08-21 14:17 01:15:25 1982008 500 233 15 252 +162.35 +/- 20.84
50654 ncm-et-3 2018-08-21 14:17 01:14:54 1990191 500 239 8 253 +173.67 +/- 20.55
50653 ncm-et-4 2018-08-21 14:17 01:15:24 1989559 500 229 17 254 +157.24 +/- 20.78
29910 ncm-et-8 2018-05-14 18:59 00:30:43 2024100 207 104 9 94 +172.32 +/- 34.91
29909 ncm-et-16 2018-05-14 18:59 00:30:50 2022303 205 81 5 119 +135.25 +/- 29.38
29908 ncm-et-5 2018-05-14 18:58 00:30:53 2020512 199 102 7 90 +180.53 +/- 35.57
29907 ncm-et-3 2018-05-14 18:58 00:31:26 2024100 209 92 5 112 +153.97 +/- 30.85
29906 ncm-et-12 2018-05-14 18:58 00:31:20 2020673 209 98 7 104 +162.1 +/- 32.66
29905 ncm-et-11 2018-05-14 18:57 00:32:05 2019858 212 92 4 116 +153.48 +/- 30.04
29904 ncm-et-9 2018-05-14 18:57 00:32:02 2025572 214 103 6 105 +169.83 +/- 32.45
29903 ncm-et-15 2018-05-14 18:57 00:32:02 2016610 209 114 4 91 +203.26 +/- 35.16
29902 ncm-et-4 2018-05-14 18:56 00:33:02 2000421 219 86 3 130 +138.59 +/- 27.63
29901 ncm-et-10 2018-05-14 18:56 00:33:20 2022467 220 92 6 122 +143.45 +/- 29.42
29900 ncm-et-14 2018-05-14 18:56 00:33:44 2022140 224 98 6 120 +151.65 +/- 29.87
29899 ncm-et-7 2018-05-14 18:56 00:33:33 2025736 229 105 9 115 +155.21 +/- 31.11
29898 ncm-et-6 2018-05-14 18:55 00:33:56 2026718 219 103 6 110 +165.33 +/- 31.57
29897 ncm-et-13 2018-05-14 18:55 00:34:29 2025245 225 102 10 113 +150.88 +/- 31.47
29896 ncm-et-8 2018-05-14 17:41 01:16:59 1989409 500 240 13 247 +170.15 +/- 21.05
29895 ncm-et-15 2018-05-14 17:40 01:15:45 2019534 500 227 17 256 +155.54 +/- 20.68
29894 ncm-et-14 2018-05-14 17:40 01:14:11 2021162 500 235 14 251 +164.93 +/- 20.86
29893 ncm-et-9 2018-05-14 17:40 01:15:49 2024264 500 246 15 239 +173.67 +/- 21.55
29892 ncm-et-3 2018-05-14 17:40 01:16:32 2025081 500 245 13 242 +174.55 +/- 21.33
29891 ncm-et-10 2018-05-14 17:40 01:14:26 2018558 500 242 17 241 +168.4 +/- 21.49
29890 ncm-et-16 2018-05-14 17:40 01:17:07 2019371 500 238 17 245 +164.93 +/- 21.27
29889 ncm-et-4 2018-05-14 17:40 01:14:55 1985116 500 241 11 248 +172.78 +/- 20.93
29888 ncm-et-13 2018-05-14 17:40 01:13:33 2026063 500 218 23 259 +143.07 +/- 20.66
29887 ncm-et-5 2018-05-14 17:40 01:17:02 2019547 500 237 8 255 +171.9 +/- 20.44
29886 ncm-et-12 2018-05-14 17:40 01:16:32 2001414 500 246 11 243 +177.22 +/- 21.21
29885 ncm-et-11 2018-05-14 17:40 01:15:59 2020021 500 231 15 254 +160.64 +/- 20.73
29884 ncm-et-7 2018-05-14 17:40 01:14:17 2026063 500 229 15 256 +158.93 +/- 20.62
29883 ncm-et-6 2018-05-14 17:40 01:14:13 2026390 500 238 11 251 +170.15 +/- 20.76

Commit

Commit ID bf01bd07551dd4e2d68d82b305cc5d89a4c23bb0
Author Stéphane Nicolet
Date 2018-05-13 05:20:39 UTC
Tweak kingAttackersCount and KingAttackWeights Use the whole kingRing for pawn attackers instead of only the squares directly around the king. This tends to give quite a lot more kingAttackersCount, so to compensate and to avoid raising the king danger too fast we lower the values in the KingAttackWeights array a little bit. STC: LLR: 2.95 (-2.94,2.94) [0.00,4.00] Total: 51892 W: 10723 L: 10369 D: 30800 http://tests.stockfishchess.org/tests/view/5af6d4dd0ebc5968e652428e LTC: LLR: 2.96 (-2.94,2.94) [0.00,4.00] Total: 24536 W: 3737 L: 3515 D: 17284 http://tests.stockfishchess.org/tests/view/5af709890ebc5968e65242ac Credits to user @xoroshiro for the idea of using the kingRing for pawn attackers. How to continue? It seems that the KingAttackWeights[] array stores values which are quite Elo-sensitive, yet they have not been tuned with SPSA recently. There might be easy Elo points to get there. Closes https://github.com/official-stockfish/Stockfish/pull/1597 Bench: 5282815
Copyright 2011–2024 Next Chess Move LLC