Dev Builds » 20181120-0700

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 09:00:05 1963554 3473 1941 79 1453 +208.02 ± 8.8
ncm-et-4 08:50:32 1940265 3350 1926 67 1357 +217.31 ± 9.11
ncm-et-9 08:07:41 1776051 2756 1533 146 1077 +192.36 ± 10.37
ncm-et-10 09:01:19 1967302 3497 1961 86 1450 +208.03 ± 8.82
ncm-et-13 08:58:32 1976966 3478 1931 94 1453 +204.16 ± 8.82
ncm-et-15 08:54:59 1969041 3446 1946 81 1419 +210.5 ± 8.92
20000 11238 553 8209 +207.1 ± 3.72

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
59696 ncm-et-4 2018-11-20 21:00 00:55:02 1965071 350 210 3 137 +236.21 ± 28.61
59695 ncm-et-9 2018-11-20 21:00 00:55:05 1765771 306 138 28 140 +130.73 ± 28.83
59694 ncm-et-3 2018-11-20 20:51 01:04:03 1962303 419 247 8 164 +225.18 ± 26.34
59693 ncm-et-15 2018-11-20 20:46 01:09:41 1961233 446 270 13 163 +228.2 ± 26.65
59692 ncm-et-13 2018-11-20 20:43 01:11:55 1972649 478 263 20 195 +194.75 ± 24.33
59691 ncm-et-10 2018-11-20 20:38 01:16:38 1962794 497 289 16 192 +214.5 ± 24.51
59690 ncm-et-4 2018-11-20 19:38 01:21:02 1902724 500 294 9 197 +224.97 ± 23.99
59689 ncm-et-3 2018-11-20 19:33 01:17:06 1964456 500 265 9 226 +196.45 ± 22.12
59688 ncm-et-9 2018-11-20 19:30 01:28:30 1810246 500 245 34 221 +156.39 ± 22.89
59687 ncm-et-15 2018-11-20 19:27 01:17:23 1970574 500 265 11 224 +194.57 ± 22.31
59686 ncm-et-13 2018-11-20 19:25 01:17:35 1969561 500 284 10 206 +213.85 ± 23.41
59685 ncm-et-3 2018-11-20 19:24 00:08:13 1964304 54 30 2 22 +199.54 ± 74.1
59684 ncm-et-10 2018-11-20 19:19 01:17:57 1967079 500 267 12 221 +195.51 ± 22.52
59683 ncm-et-4 2018-11-20 18:15 01:21:40 1871839 500 287 7 206 +219.87 ± 23.31
59682 ncm-et-15 2018-11-20 18:07 01:19:26 1966329 500 284 15 201 +208.93 ± 23.88
59681 ncm-et-13 2018-11-20 18:06 01:17:38 1970805 500 271 12 217 +199.29 ± 22.77
59680 ncm-et-3 2018-11-20 18:04 01:18:59 1961691 500 265 10 225 +195.51 ± 22.21
59679 ncm-et-9 2018-11-20 18:03 01:26:47 1777692 500 288 23 189 +205.04 ± 24.81
59678 ncm-et-10 2018-11-20 18:02 01:16:22 1960767 500 296 12 192 +223.94 ± 24.43
59677 ncm-et-4 2018-11-20 16:55 01:19:40 1945408 500 283 16 201 +206.98 ± 23.9
59676 ncm-et-15 2018-11-20 16:48 01:17:23 1970752 500 277 11 212 +206.01 ± 23.05
59675 ncm-et-13 2018-11-20 16:47 01:17:30 1974519 500 269 14 217 +195.51 ± 22.82
59674 ncm-et-3 2018-11-20 16:45 01:17:57 1965851 500 291 12 197 +218.86 ± 24.08
59673 ncm-et-10 2018-11-20 16:44 01:17:10 1967855 500 269 18 213 +191.78 ± 23.16
59672 ncm-et-9 2018-11-20 16:32 01:29:11 1745702 500 308 20 172 +228.08 ± 26.05
59671 ncm-et-4 2018-11-20 15:36 01:17:49 1971592 500 289 9 202 +219.87 ± 23.65
59670 ncm-et-15 2018-11-20 15:31 01:16:01 1970788 500 274 11 215 +203.11 ± 22.86
59669 ncm-et-13 2018-11-20 15:29 01:16:59 1974369 500 288 13 199 +214.85 ± 23.97
59668 ncm-et-3 2018-11-20 15:26 01:18:03 1963689 500 278 12 210 +206.01 ± 23.21
59667 ncm-et-10 2018-11-20 15:25 01:17:17 1982848 500 281 11 208 +209.91 ± 23.31
59666 ncm-et-9 2018-11-20 15:02 01:29:34 1768866 500 292 16 192 +215.85 ± 24.52
59665 ncm-et-4 2018-11-20 14:18 01:16:38 1962304 500 294 8 198 +226.0 ± 23.89
59664 ncm-et-15 2018-11-20 14:13 01:17:07 1979755 500 287 10 203 +216.85 ± 23.61
59663 ncm-et-13 2018-11-20 14:10 01:17:56 1996262 500 282 11 207 +210.89 ± 23.38
59662 ncm-et-10 2018-11-20 14:08 01:16:35 1967762 500 281 10 209 +210.89 ± 23.22
59661 ncm-et-3 2018-11-20 14:06 01:18:18 1963993 500 270 9 221 +201.19 ± 22.42
59660 ncm-et-9 2018-11-20 13:42 01:18:34 1788034 450 262 25 163 +203.43 ± 26.77
59659 ncm-et-4 2018-11-20 12:58 01:18:41 1962919 500 269 15 216 +194.57 ± 22.91
59658 ncm-et-15 2018-11-20 12:54 01:17:58 1963861 500 289 10 201 +218.86 ± 23.75
59657 ncm-et-13 2018-11-20 12:50 01:18:59 1980598 500 274 14 212 +200.24 ± 23.14
59656 ncm-et-3 2018-11-20 12:48 01:17:26 1962148 500 295 17 188 +217.85 ± 24.82
59655 ncm-et-10 2018-11-20 12:47 01:19:20 1962014 500 278 7 215 +210.89 ± 22.72

Commit

Commit ID 4b88bea4fcd5c4cb931610f97460615dd71ff35b
Author Joost VandeVondele
Date 2018-11-20 07:00:19 UTC
Force time check on TB probe in search. Because of aggressive time management and optimistic assumptions about move overhead, it's still very easy to get Stockfish to forfeit on time when we hit an endgame and have Syzygy EGTB on a spinning drive. The latency from serving a few thousand EGTB probes (~10ms each), of which there can currently be up to 4000 outstanding before a time check, will easily overwhelm the default Move Overhead of 30ms. This problem was first raised by Gian-Carlo Pascutto and some solutions and improvements were discussed in the following pull requests: https://github.com/official-stockfish/Stockfish/pull/1471 https://github.com/official-stockfish/Stockfish/pull/1623 https://github.com/official-stockfish/Stockfish/pull/1783 This patch is a minimal change proposed by Marco Costalba to lower the impact of the bug. We now force a check of the clock right after each tablebase read. No functional change.
Copyright 2011–2024 Next Chess Move LLC