Dev Builds » 20170826-0753

You are viewing an old NCM Stockfish dev build test. You may find the most recent dev build tests using Stockfish 15 as the baseline here.

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 06:13:01 2001777 2535 890 83 1562 +114.59 ± 8.0
ncm-et-4 06:13:39 1996169 2522 874 95 1553 +110.94 ± 8.06
ncm-et-5 02:02:29 2022549 833 321 26 486 +128.61 ± 14.57
ncm-et-6 02:02:13 2026228 839 319 27 493 +126.19 ± 14.45
ncm-et-7 02:02:23 2022804 834 293 37 504 +110.2 ± 14.3
ncm-et-8 02:02:17 1987220 814 308 30 476 +123.62 ± 14.78
ncm-et-9 06:13:34 2000635 2524 927 97 1500 +118.66 ± 8.29
ncm-et-10 06:13:21 1965694 2490 915 102 1473 +117.75 ± 8.39
ncm-et-11 02:02:28 2012769 839 286 27 526 +110.87 ± 13.7
ncm-et-12 02:02:28 1968391 811 311 25 475 +128.02 ± 14.72
ncm-et-13 06:13:33 1982279 2495 893 72 1530 +118.74 ± 8.07
ncm-et-14 02:02:16 2019615 826 312 25 489 +125.96 ± 14.45
ncm-et-15 04:11:24 1948319 1638 588 62 988 +115.66 ± 10.17
20000 7237 708 12055 +117.73 ± 2.91

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
55815 ncm-et-15 2018-09-30 08:56 00:21:00 1978728 138 46 2 90 +114.78 ± 32.13
55814 ncm-et-13 2018-09-30 08:55 00:21:49 1988614 148 46 6 96 +96.29 ± 31.97
55813 ncm-et-10 2018-09-30 08:53 00:23:36 1882432 147 67 6 74 +153.43 ± 38.89
55812 ncm-et-4 2018-09-30 08:50 00:26:59 1987350 177 62 5 110 +116.01 ± 30.04
55811 ncm-et-9 2018-09-30 08:48 00:28:26 1988457 187 85 5 97 +158.85 ± 33.45
55810 ncm-et-3 2018-09-30 08:47 00:29:38 1991297 203 70 6 127 +113.4 ± 27.91
55809 ncm-et-15 2018-09-30 07:39 01:16:11 1911319 500 173 17 310 +112.14 ± 17.96
55808 ncm-et-10 2018-09-30 07:38 01:14:02 1981381 500 174 20 306 +110.6 ± 18.24
55807 ncm-et-13 2018-09-30 07:36 01:17:49 1936269 500 197 14 289 +133.34 ± 18.9
55806 ncm-et-4 2018-09-30 07:35 01:13:54 1987825 500 187 22 291 +119.11 ± 19.03
55805 ncm-et-9 2018-09-30 07:33 01:14:44 1988456 500 171 16 313 +111.37 ± 17.78
55804 ncm-et-3 2018-09-30 07:32 01:14:17 1989717 500 186 21 293 +119.11 ± 18.9
55803 ncm-et-15 2018-09-30 06:21 01:16:49 1949918 500 195 20 285 +126.97 ± 19.27
55802 ncm-et-10 2018-09-30 06:20 01:16:57 1941336 500 184 19 297 +119.11 ± 18.65
55801 ncm-et-13 2018-09-30 06:19 01:16:16 1945200 500 174 14 312 +115.23 ± 17.77
55800 ncm-et-4 2018-09-30 06:19 01:15:12 1989716 500 165 20 315 +103.73 ± 17.8
55799 ncm-et-9 2018-09-30 06:18 01:13:43 1988139 500 175 30 295 +103.73 ± 19.01
55798 ncm-et-3 2018-09-30 06:18 01:13:20 1989558 500 179 13 308 +119.89 ± 17.93
55797 ncm-et-3 2018-09-30 05:03 01:13:40 1988457 500 159 16 325 +102.22 ± 17.19
55796 ncm-et-13 2018-09-30 05:03 01:15:18 1973597 500 172 13 315 +114.45 ± 17.59
55795 ncm-et-4 2018-09-30 05:02 01:15:10 1988140 500 183 20 297 +117.55 ± 18.68
55794 ncm-et-10 2018-09-30 05:02 01:16:33 1948813 500 191 19 290 +124.6 ± 19.0
55793 ncm-et-15 2018-09-30 05:02 01:17:24 1953311 500 174 23 303 +108.3 ± 18.46
55792 ncm-et-9 2018-09-30 05:02 01:14:19 1988927 500 184 20 296 +118.33 ± 18.73
4410 ncm-et-8 2018-03-03 02:16 00:47:33 1952300 314 127 9 178 +137.29 ± 24.23
4409 ncm-et-12 2018-03-03 02:16 00:47:36 1917737 311 112 7 192 +122.09 ± 22.63
4408 ncm-et-14 2018-03-03 02:14 00:48:56 2020186 326 125 7 194 +131.73 ± 22.76
4407 ncm-et-7 2018-03-03 02:14 00:48:59 2024263 334 127 11 196 +125.9 ± 22.96
4406 ncm-et-9 2018-03-03 02:14 00:49:15 2023609 337 122 12 203 +117.71 ± 22.44
4405 ncm-et-5 2018-03-03 02:14 00:49:22 2022141 333 124 13 196 +120.41 ± 23.05
4404 ncm-et-3 2018-03-03 02:14 00:49:06 2026228 332 136 8 188 +141.25 ± 23.49
4403 ncm-et-6 2018-03-03 02:14 00:49:19 2026554 339 123 8 208 +122.72 ± 21.81
4402 ncm-et-11 2018-03-03 02:14 00:49:47 2020022 339 104 14 221 +94.5 ± 20.99
4401 ncm-et-10 2018-03-03 02:13 00:49:58 2020347 343 120 13 210 +112.12 ± 21.99
4400 ncm-et-4 2018-03-03 02:13 00:50:31 2026226 345 114 15 216 +102.58 ± 21.61
4399 ncm-et-13 2018-03-03 02:12 00:51:16 2024263 347 115 12 220 +106.33 ± 21.17
4398 ncm-et-14 2018-03-03 01:00 01:13:20 2019045 500 187 18 295 +122.24 ± 18.72
4397 ncm-et-8 2018-03-03 01:00 01:14:44 2022140 500 181 21 298 +115.23 ± 18.66
4396 ncm-et-13 2018-03-03 01:00 01:11:05 2025735 500 189 13 298 +127.76 ± 18.42
4395 ncm-et-9 2018-03-03 01:00 01:13:07 2026226 500 190 14 296 +127.76 ± 18.55
4394 ncm-et-4 2018-03-03 01:00 01:11:53 1997758 500 163 13 324 +107.54 ± 17.15
4393 ncm-et-3 2018-03-03 01:00 01:13:00 2025408 500 160 19 321 +100.71 ± 17.48
4392 ncm-et-6 2018-03-03 01:00 01:12:54 2025902 500 196 19 285 +128.55 ± 19.25
4391 ncm-et-7 2018-03-03 01:00 01:13:24 2021345 500 166 26 308 +99.95 ± 18.29
4390 ncm-et-5 2018-03-03 01:00 01:13:07 2022958 500 197 13 290 +134.15 ± 18.82
4389 ncm-et-11 2018-03-03 01:00 01:12:41 2005517 500 182 13 305 +122.24 ± 18.08
4388 ncm-et-12 2018-03-03 01:00 01:14:52 2019046 500 199 18 283 +131.74 ± 19.32
4387 ncm-et-10 2018-03-03 01:00 01:12:15 2019859 500 179 25 296 +110.6 ± 18.85

Commit

Commit ID d5f883ab29d43b35746ff605cf13c3722df56041
Author Joost VandeVondele
Date 2017-08-26 07:53:34 UTC
Improve multi-threaded mate finding If any thread found a 'mate in x' stop the search. Previously only mainThread would do so. Requires the bestThread selection to be adjusted to always prefer mate scores, even if the search depth is less. I've tried to collect some data for this patch. On 30 cores, mate finding seems 5-30% faster on average. It is not so easy to get numbers for this, as the time to find a mate fluctuates significantly with multi-threaded runs, so it is an average over 100 searches for the same position. Furthermore, hash size and position make a difference as well. Bench: 5965302
Copyright 2011–2024 Next Chess Move LLC