Dev Builds » 20180327-2231

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 06:02:13 2001636 2388 1073 59 1256 +157.5 +/- 9.22
ncm-et-4 06:03:59 1997167 2392 1126 94 1172 +160.4 +/- 9.74
ncm-et-5 01:49:53 2023283 736 335 26 375 +155.48 +/- 17.1
ncm-et-6 01:47:50 2024755 711 332 21 358 +162.96 +/- 17.47
ncm-et-7 01:43:34 1992537 678 289 20 369 +145.85 +/- 16.97
ncm-et-8 01:47:21 2001618 713 324 13 376 +162.43 +/- 16.76
ncm-et-9 06:05:27 2000972 2417 1097 70 1250 +157.62 +/- 9.3
ncm-et-10 06:03:35 1980624 2367 1105 71 1191 +162.71 +/- 9.57
ncm-et-11 01:47:32 2009976 702 329 15 358 +167.22 +/- 17.32
ncm-et-12 01:48:08 2021080 726 359 19 348 +176.47 +/- 17.8
ncm-et-13 06:04:50 1985123 2383 1132 75 1176 +165.61 +/- 9.67
ncm-et-14 01:48:52 2020350 725 326 26 373 +152.93 +/- 17.13
ncm-et-15 06:02:47 1971551 2360 1079 70 1211 +158.74 +/- 9.46
ncm-et-16 01:49:02 1960272 702 294 17 391 +144.96 +/- 16.33
20000 9200 596 10204 +159.87 +/- 3.26

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
51954 ncm-et-10 2018-08-31 23:11 00:22:14 1979977 144 56 7 81 +123.13 +/- 36.58
51953 ncm-et-15 2018-08-31 23:10 00:23:12 1950417 148 70 7 71 +157.95 +/- 40.08
51952 ncm-et-13 2018-08-31 23:10 00:23:47 1988770 151 62 6 83 +135.3 +/- 36.12
51951 ncm-et-4 2018-08-31 23:05 00:28:04 1989243 176 80 4 92 +160.56 +/- 34.2
51950 ncm-et-9 2018-08-31 23:05 00:28:30 1987823 187 85 2 100 +165.73 +/- 32.26
51949 ncm-et-3 2018-08-31 23:03 00:30:00 1991614 194 85 7 102 +148.04 +/- 32.75
51948 ncm-et-10 2018-08-31 21:51 01:18:28 1941518 500 252 15 233 +179.0 +/- 21.89
51947 ncm-et-15 2018-08-31 21:51 01:17:43 1952960 500 249 13 238 +178.11 +/- 21.55
51946 ncm-et-13 2018-08-31 21:51 01:17:36 1950326 500 248 15 237 +175.44 +/- 21.67
51945 ncm-et-4 2018-08-31 21:48 01:16:13 1990981 500 229 26 245 +149.68 +/- 21.46
51944 ncm-et-9 2018-08-31 21:48 01:15:33 1987037 500 232 17 251 +159.78 +/- 20.95
51943 ncm-et-3 2018-08-31 21:48 01:14:29 1989874 500 222 14 264 +153.86 +/- 20.17
51942 ncm-et-15 2018-08-31 20:33 01:17:12 1942321 500 220 11 269 +154.7 +/- 19.81
51941 ncm-et-10 2018-08-31 20:32 01:18:13 1943564 500 233 13 254 +164.07 +/- 20.67
51940 ncm-et-13 2018-08-31 20:32 01:17:44 1956658 500 263 9 228 +194.57 +/- 22.0
51939 ncm-et-3 2018-08-31 20:32 01:14:45 1989875 500 206 14 280 +140.62 +/- 19.35
51938 ncm-et-4 2018-08-31 20:31 01:15:46 1989719 500 250 14 236 +178.11 +/- 21.69
51937 ncm-et-9 2018-08-31 20:31 01:16:00 1989085 500 228 14 258 +158.93 +/- 20.49
51936 ncm-et-9 2018-08-31 19:15 01:15:16 1988454 500 230 12 258 +162.35 +/- 20.42
51935 ncm-et-13 2018-08-31 19:14 01:16:35 1964006 500 226 19 255 +153.02 +/- 20.78
51934 ncm-et-10 2018-08-31 19:14 01:16:40 1980598 500 234 17 249 +161.49 +/- 21.05
51933 ncm-et-4 2018-08-31 19:14 01:15:45 1989876 500 232 22 246 +155.54 +/- 21.33
51932 ncm-et-15 2018-08-31 19:14 01:17:44 1946330 500 214 13 273 +148.02 +/- 19.67
51931 ncm-et-3 2018-08-31 19:14 01:16:32 1988454 500 243 11 246 +174.55 +/- 21.04
12763 ncm-et-7 2018-03-28 01:03 00:27:44 1987988 178 70 2 106 +139.82 +/- 30.5
12762 ncm-et-3 2018-03-28 01:00 00:30:02 2026554 194 84 3 107 +154.5 +/- 31.13
12761 ncm-et-11 2018-03-28 01:00 00:30:46 2017100 202 98 3 101 +177.35 +/- 32.64
12760 ncm-et-16 2018-03-28 01:00 00:30:46 2022630 202 83 4 115 +143.52 +/- 29.92
12759 ncm-et-8 2018-03-28 00:58 00:32:20 1982727 213 94 2 117 +160.61 +/- 29.59
12758 ncm-et-15 2018-03-28 00:58 00:32:43 2018722 212 101 7 104 +165.54 +/- 32.71
12757 ncm-et-6 2018-03-28 00:57 00:32:58 2025900 211 102 2 107 +178.97 +/- 31.46
12756 ncm-et-4 2018-03-28 00:57 00:33:04 1998433 216 104 10 102 +162.0 +/- 33.41
12755 ncm-et-10 2018-03-28 00:57 00:33:25 2019371 223 97 4 122 +154.3 +/- 29.26
12754 ncm-et-14 2018-03-28 00:56 00:34:03 2022306 225 102 8 115 +154.61 +/- 30.96
12753 ncm-et-12 2018-03-28 00:56 00:33:46 2020347 226 113 6 107 +178.76 +/- 32.26
12752 ncm-et-9 2018-03-28 00:56 00:34:33 2025736 230 99 9 122 +143.61 +/- 29.93
12751 ncm-et-5 2018-03-28 00:55 00:35:39 2022303 236 99 7 130 +143.0 +/- 28.59
12750 ncm-et-13 2018-03-28 00:55 00:35:38 2024753 232 104 7 121 +154.74 +/- 29.97
12749 ncm-et-7 2018-03-27 23:46 01:15:50 1997087 500 219 18 263 +148.02 +/- 20.34
12748 ncm-et-3 2018-03-27 23:43 01:16:25 2023447 500 233 10 257 +166.66 +/- 20.41
12747 ncm-et-15 2018-03-27 23:42 01:14:13 2018558 500 225 19 256 +152.18 +/- 20.73
12746 ncm-et-8 2018-03-27 23:42 01:15:01 2020510 500 230 11 259 +163.21 +/- 20.33
12745 ncm-et-11 2018-03-27 23:42 01:16:46 2002853 500 231 12 257 +163.21 +/- 20.47
12744 ncm-et-6 2018-03-27 23:42 01:14:52 2023610 500 230 19 251 +156.39 +/- 21.0
12743 ncm-et-10 2018-03-27 23:41 01:14:35 2018721 500 233 15 252 +162.35 +/- 20.84
12742 ncm-et-4 2018-03-27 23:41 01:15:07 2024755 500 231 18 251 +158.08 +/- 20.97
12741 ncm-et-12 2018-03-27 23:41 01:14:22 2021814 500 246 13 241 +175.44 +/- 21.38
12740 ncm-et-14 2018-03-27 23:41 01:14:49 2018395 500 224 18 258 +152.18 +/- 20.6
12739 ncm-et-16 2018-03-27 23:40 01:18:16 1897914 500 211 13 276 +145.54 +/- 19.52
12738 ncm-et-13 2018-03-27 23:40 01:13:30 2026227 500 229 19 252 +155.54 +/- 20.94
12737 ncm-et-5 2018-03-27 23:40 01:14:14 2024264 500 236 19 245 +161.49 +/- 21.32
12736 ncm-et-9 2018-03-27 23:39 01:15:35 2027702 500 223 16 261 +153.02 +/- 20.39

Commit

Commit ID 9953bfff621b287f5e08139f51d6663c2620db1e
Author Jerry Donald Watson
Date 2018-03-27 22:31:47 UTC
Change reduction/improving interaction Adjust criterion for applying extra reduction if not improving. We now add an extra ply of reduction if r > 1.0, instead of the previous condition Reductions[NonPV][imp][d][mc] >= 2. Why does this work? Previously, reductions when not improving had a discontinuity as the depth and/or move count increases due to the Reductions[NonPV][imp][d][mc] >= 2 condition. Hence, values of r such that 0.5 < r < 1.5 would be mapped to a reduction of 1, while 1.5 < r < 2.5 would be mapped to a reduction of 3. This patch allows values of r satisfying 1.0 < r < 1.5 to be mapped to a reduction of 2, making the reduction formula more continuous. STC: LLR: 2.96 (-2.94,2.94) [0.00,5.00] Total: 35908 W: 7382 L: 7087 D: 21439 http://tests.stockfishchess.org/tests/view/5aba723a0ebc5902a4743e8f LTC: LLR: 2.96 (-2.94,2.94) [0.00,5.00] Total: 23087 W: 3584 L: 3378 D: 16125 http://tests.stockfishchess.org/tests/view/5aba89070ebc5902a4743ea9 Ideas for future work: - We could look at retuning the LMR formula. - We could look at adjusting the reductions in PV nodes if not improving. Bench: 5326261
Copyright 2011–2024 Next Chess Move LLC