Dev Builds » 20180209-1807

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 13:31:38 2013580 5273 2407 192 2674 +155.57 ± 6.41
ncm-et-4 13:33:08 2014310 5284 2455 192 2637 +159.05 ± 6.47
ncm-et-5 07:08:30 2024454 2794 1308 93 1393 +161.86 ± 8.88
ncm-et-9 04:20:40 1988336 1679 782 68 829 +157.77 ± 11.58
ncm-et-10 04:20:00 1962224 1640 727 62 851 +149.47 ± 11.33
ncm-et-13 04:20:24 1989717 1679 773 52 854 +159.54 ± 11.3
ncm-et-15 04:20:22 1979195 1651 755 66 830 +154.42 ± 11.54
20000 9207 725 10068 +157.28 ± 3.3

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
53796 ncm-et-10 2018-09-15 04:24 00:22:45 1959085 140 49 6 85 +110.27 ± 34.87
53795 ncm-et-15 2018-09-15 04:24 00:23:25 1969810 151 70 4 77 +162.82 ± 37.69
53794 ncm-et-3 2018-09-15 04:20 00:26:54 1989085 170 76 14 80 +132.83 ± 38.14
53793 ncm-et-9 2018-09-15 04:19 00:27:49 1986878 179 82 8 89 +152.77 ± 35.55
53792 ncm-et-13 2018-09-15 04:19 00:28:06 1988928 179 85 4 90 +169.5 ± 34.8
53791 ncm-et-4 2018-09-15 04:19 00:28:30 1990822 181 88 5 88 +172.15 ± 35.51
53790 ncm-et-10 2018-09-15 03:05 01:18:22 1984362 500 223 22 255 +148.02 ± 20.85
53789 ncm-et-15 2018-09-15 03:03 01:19:36 1982792 500 238 19 243 +163.21 ± 21.43
53788 ncm-et-9 2018-09-15 03:02 01:16:34 1990665 500 236 17 247 +163.21 ± 21.16
53787 ncm-et-3 2018-09-15 03:01 01:18:40 1988927 500 235 22 243 +158.08 ± 21.5
53786 ncm-et-4 2018-09-15 03:00 01:17:18 1990191 500 239 17 244 +165.8 ± 21.33
53785 ncm-et-13 2018-09-15 03:00 01:17:56 1990034 500 237 15 248 +165.8 ± 21.05
53784 ncm-et-10 2018-09-15 01:47 01:16:58 1976252 500 230 18 252 +157.24 ± 20.92
53783 ncm-et-9 2018-09-15 01:43 01:17:18 1987666 500 228 23 249 +151.35 ± 21.19
53782 ncm-et-3 2018-09-15 01:43 01:16:33 1990823 500 227 20 253 +153.02 ± 20.91
53781 ncm-et-15 2018-09-15 01:43 01:19:29 1984204 500 216 16 268 +147.19 ± 20.02
53780 ncm-et-13 2018-09-15 01:42 01:16:26 1989243 500 229 15 256 +158.93 ± 20.62
53779 ncm-et-4 2018-09-15 01:42 01:17:22 1991613 500 230 19 251 +156.39 ± 21.0
53778 ncm-et-15 2018-09-15 00:24 01:17:52 1979976 500 231 27 242 +150.51 ± 21.64
53777 ncm-et-10 2018-09-15 00:24 01:21:55 1929197 500 225 16 259 +154.7 ± 20.49
53776 ncm-et-13 2018-09-15 00:23 01:17:56 1990664 500 222 18 260 +150.51 ± 20.49
53775 ncm-et-4 2018-09-15 00:23 01:17:25 1988612 500 252 13 235 +180.8 ± 21.72
53774 ncm-et-9 2018-09-15 00:23 01:18:59 1988138 500 236 20 244 +160.64 ± 21.4
53773 ncm-et-3 2018-09-15 00:23 01:18:39 1989716 500 227 26 247 +148.02 ± 21.36
1029 ncm-et-4 2018-02-10 04:38 00:16:10 2022957 103 42 2 59 +142.4 ± 41.83
1028 ncm-et-3 2018-02-10 04:37 00:16:40 2025408 103 44 3 56 +146.39 ± 43.86
1027 ncm-et-5 2018-02-10 04:08 00:45:50 2024590 294 139 6 149 +169.44 ± 26.91
1026 ncm-et-4 2018-02-10 03:20 01:16:41 2028686 500 229 17 254 +157.24 ± 20.78
1025 ncm-et-3 2018-02-10 03:19 01:16:56 2024754 500 229 14 257 +159.78 ± 20.54
1024 ncm-et-5 2018-02-10 02:51 01:16:10 2025572 500 229 22 249 +153.02 ± 21.17
1023 ncm-et-3 2018-02-10 02:02 01:16:23 2025572 500 241 14 245 +170.15 ± 21.19
1022 ncm-et-4 2018-02-10 02:00 01:19:13 2028030 500 210 23 267 +136.56 ± 20.25
1021 ncm-et-5 2018-02-10 01:33 01:16:34 2024426 500 256 16 228 +181.7 ± 22.21
1020 ncm-et-3 2018-02-10 00:44 01:16:20 2024754 500 213 19 268 +142.26 ± 20.1
1019 ncm-et-4 2018-02-10 00:41 01:17:21 2024263 500 234 17 249 +161.49 ± 21.05
1018 ncm-et-5 2018-02-10 00:16 01:16:10 2024590 500 229 17 254 +157.24 ± 20.78
1017 ncm-et-3 2018-02-09 23:27 01:16:12 2026718 500 231 19 250 +157.24 ± 21.05
1016 ncm-et-4 2018-02-09 23:24 01:16:10 2028358 500 221 24 255 +144.72 ± 20.89
1015 ncm-et-5 2018-02-09 22:58 01:16:36 2024427 500 232 16 252 +160.64 ± 20.86
1014 ncm-et-3 2018-02-09 22:09 01:16:24 2027210 500 218 10 272 +153.86 ± 19.62
1013 ncm-et-4 2018-02-09 22:06 01:16:45 2026064 500 235 20 245 +159.78 ± 21.34
1012 ncm-et-5 2018-02-09 21:40 01:17:10 2023119 500 223 16 261 +153.02 ± 20.39
1011 ncm-et-3 2018-02-09 20:52 01:16:05 2026391 500 239 17 244 +165.8 ± 21.33
1010 ncm-et-4 2018-02-09 20:50 01:15:18 2025245 500 236 17 247 +163.21 ± 21.16
1009 ncm-et-3 2018-02-09 19:35 01:15:52 2023609 500 227 14 259 +158.08 ± 20.43
1008 ncm-et-4 2018-02-09 19:34 01:14:55 2026883 500 239 18 243 +164.93 ± 21.41

Commit

Commit ID cb1324312d051269700b74cb59759a12530d0b7a
Author Stefano Cardanobile
Date 2018-02-09 18:07:19 UTC
Introduce dynamic contempt Make contempt dependent on the current score of the root position. The idea is that we now use a linear formula like the following to decide on the contempt to use during a search : contempt = x + y * eval where x is the base contempt set by the user in the "Contempt" UCI option, and y * eval is the dynamic part which adapts itself to the estimation of the evaluation of the root position returned by the search. In this patch, we use x = 18 centipawns by default, and the y * eval correction can go from -20 centipawns if the root eval is less than -2.0 pawns, up to +20 centipawns when the root eval is more than 2.0 pawns. To summarize, the new contempt goes from -0.02 to 0.38 pawns, depending if Stockfish is losing or winning, with an average value of 0.18 pawns by default. STC: LLR: 2.95 (-2.94,2.94) [0.00,5.00] Total: 110052 W: 24614 L: 23938 D: 61500 http://tests.stockfishchess.org/tests/view/5a72e6020ebc590f2c86ea20 LTC: LLR: 2.97 (-2.94,2.94) [0.00,5.00] Total: 16470 W: 2896 L: 2705 D: 10869 http://tests.stockfishchess.org/tests/view/5a76c5b90ebc5902971a9830 A second match at LTC was organised against the current master: ELO: 1.45 +-2.9 (95%) LOS: 84.0% Total: 19369 W: 3350 L: 3269 D: 12750 http://tests.stockfishchess.org/tests/view/5a7acf980ebc5902971a9a2e Finally, we checked that there is no apparent problem with multithreading, despite the fact that some threads might have a slightly different contempt level that the main thread. Match of this version against master, both using 5 threads, time control 30+0.3: ELO: 2.18 +-3.2 (95%) LOS: 90.8% Total: 14840 W: 2502 L: 2409 D: 9929 http://tests.stockfishchess.org/tests/view/5a7bf3e80ebc5902971a9aa2 Include suggestions from Marco Costalba, Aram Tumanian, Ronald de Man, etc. Bench: 5207156
Copyright 2011–2024 Next Chess Move LLC