Dev Builds » 20180209-0941

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 14:10:25 2013454 5507 2518 219 2770 +154.48 ± 6.31
ncm-et-4 14:11:38 2013634 5498 2529 201 2768 +156.99 ± 6.3
ncm-et-5 05:57:35 2022925 2346 1048 82 1216 +152.08 ± 9.46
ncm-et-9 04:22:12 1989165 1684 776 62 846 +157.23 ± 11.41
ncm-et-10 04:22:19 1962486 1642 768 57 817 +161.07 ± 11.61
ncm-et-13 04:22:20 1987865 1667 737 66 864 +148.23 ± 11.26
ncm-et-15 04:22:23 1962679 1656 745 70 841 +150.35 ± 11.46
20000 9121 757 10122 +154.79 ± 3.3

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
53820 ncm-et-10 2018-09-15 08:52 00:22:12 1978879 142 76 2 64 +200.78 ± 41.7
53819 ncm-et-15 2018-09-15 08:50 00:24:24 1943941 156 65 4 87 +143.49 ± 34.79
53818 ncm-et-13 2018-09-15 08:48 00:25:50 1985620 167 59 9 99 +107.31 ± 32.74
53817 ncm-et-4 2018-09-15 08:47 00:27:22 1988296 175 74 5 96 +144.83 ± 33.31
53816 ncm-et-9 2018-09-15 08:45 00:28:20 1989402 184 86 9 89 +154.9 ± 35.74
53815 ncm-et-3 2018-09-15 08:45 00:29:00 1990349 176 78 5 93 +153.34 ± 34.1
53814 ncm-et-10 2018-09-15 07:31 01:20:10 1951919 500 223 23 254 +147.19 ± 20.93
53813 ncm-et-15 2018-09-15 07:29 01:19:08 1984362 500 235 21 244 +158.93 ± 21.42
53812 ncm-et-13 2018-09-15 07:29 01:17:54 1987826 500 222 20 258 +148.85 ± 20.65
53811 ncm-et-4 2018-09-15 07:27 01:18:33 1989402 500 215 22 263 +141.44 ± 20.43
53810 ncm-et-3 2018-09-15 07:27 01:16:49 1988927 500 224 27 249 +144.72 ± 21.27
53809 ncm-et-9 2018-09-15 07:26 01:18:33 1990982 500 227 21 252 +152.18 ± 20.99
53808 ncm-et-15 2018-09-15 06:11 01:17:37 1978885 500 223 21 256 +148.85 ± 20.78
53807 ncm-et-13 2018-09-15 06:10 01:18:24 1989087 500 244 17 239 +170.15 ± 21.6
53806 ncm-et-10 2018-09-15 06:08 01:21:17 1936669 500 231 20 249 +156.39 ± 21.13
53805 ncm-et-4 2018-09-15 06:08 01:17:56 1989085 500 225 19 256 +152.18 ± 20.73
53804 ncm-et-9 2018-09-15 06:07 01:17:42 1988296 500 242 16 242 +169.27 ± 21.41
53803 ncm-et-3 2018-09-15 06:07 01:19:15 1990664 500 230 24 246 +152.18 ± 21.37
53802 ncm-et-9 2018-09-15 04:48 01:17:37 1987981 500 221 16 263 +151.35 ± 20.28
53801 ncm-et-4 2018-09-15 04:48 01:18:31 1987667 500 232 21 247 +156.39 ± 21.26
53800 ncm-et-15 2018-09-15 04:48 01:21:14 1943530 500 222 24 254 +145.54 ± 20.95
53799 ncm-et-10 2018-09-15 04:48 01:18:40 1982478 500 238 12 250 +169.27 ± 20.85
53798 ncm-et-13 2018-09-15 04:48 01:20:12 1988928 500 212 20 268 +140.62 ± 20.13
53797 ncm-et-3 2018-09-15 04:48 01:17:06 1989874 500 212 21 267 +139.81 ± 20.21
1044 ncm-et-4 2018-02-10 10:07 00:49:18 2025736 323 159 12 152 +170.63 ± 27.22
1043 ncm-et-3 2018-02-10 10:06 00:51:06 2023773 331 141 11 179 +144.2 ± 24.5
1042 ncm-et-5 2018-02-10 10:03 00:54:05 2020185 346 141 15 190 +132.61 ± 23.86
1041 ncm-et-4 2018-02-10 08:49 01:17:35 2023446 500 238 12 250 +169.27 ± 20.85
1040 ncm-et-3 2018-02-10 08:48 01:16:51 2027046 500 229 17 254 +157.24 ± 20.78
1039 ncm-et-5 2018-02-10 08:46 01:15:17 2024100 500 232 14 254 +162.35 ± 20.7
1038 ncm-et-4 2018-02-10 07:32 01:16:02 2026719 500 238 23 239 +159.78 ± 21.74
1037 ncm-et-3 2018-02-10 07:30 01:16:50 2024427 500 247 21 232 +169.27 ± 22.09
1036 ncm-et-5 2018-02-10 07:29 01:16:12 2024917 500 222 21 257 +148.02 ± 20.72
1035 ncm-et-3 2018-02-10 06:13 01:15:34 2026882 500 242 15 243 +170.15 ± 21.33
1034 ncm-et-4 2018-02-10 06:13 01:17:44 2026065 500 210 19 271 +139.81 ± 19.95
1033 ncm-et-5 2018-02-10 06:12 01:16:14 2022141 500 223 22 255 +148.02 ± 20.85
1032 ncm-et-3 2018-02-10 04:55 01:17:06 2025900 500 241 20 239 +164.93 ± 21.68
1031 ncm-et-4 2018-02-10 04:55 01:16:51 2025900 500 236 15 249 +164.93 ± 21.0
1030 ncm-et-5 2018-02-10 04:55 01:15:47 2023282 500 230 10 260 +164.07 ± 20.25
1007 ncm-et-3 2018-02-09 18:18 01:16:48 2024263 500 212 22 266 +138.99 ± 20.28
1006 ncm-et-4 2018-02-09 18:15 01:17:33 2026390 500 240 25 235 +159.78 ± 21.99
1005 ncm-et-3 2018-02-09 17:01 01:15:57 2025408 500 215 25 260 +138.99 ± 20.65
1004 ncm-et-4 2018-02-09 16:57 01:16:55 2026554 500 226 16 258 +155.54 ± 20.54
1003 ncm-et-3 2018-02-09 15:41 01:18:03 2023937 500 247 11 242 +178.11 ± 21.26
1002 ncm-et-4 2018-02-09 15:39 01:17:18 2028358 500 236 12 252 +167.53 ± 20.74

Commit

Commit ID d71adc5bd979fd42ff9bbb3d2257e188aac86be9
Author Leonid Pechenik
Date 2018-02-09 09:41:32 UTC
Retire "Extra thinking before accepting draw PVs" This patch simplifies the time management code, removing the extra thinking time for moves with draw PV and increasing thinking time for all moves proportionally by around 4%. Last time when the time management was carefully tuned was 1.5-2 years ago. As new patches were getting added, time management was drifting out of optimum. This happens because when search becomes more precise pv and score are becoming more stable, there are less fail lows, best move is picked earlier and there are less best move changes. All this factors are entering in time management, and average time per move is decreasing with more and more good patches. For individual patches such effect is small (except some) and may be up or down, but when there are many of them, effect is more substantial. The same way benchmark with more and more patches is slowly drifting down on average. So my understanding that back in October adding more think time for draw PV showed positive Elo because time management was not well tuned, there was more time available, and think_hard patch applied this additional time to moves with draw PV, while just retuning back to optimum would recover Elo anyway. It is possible that absence of contempt also helped, as SF9 is showing less 0.0 scores than the October version. Anyway, to me it seems that proper place to deal with draw PV is search, and contempt sounds as much better solution. In time management there is little additional elo, and if some code is not helping like removed here, it is better to discard it. It is simpler to find genuine improvement if code is clean. • Passed STC: LLR: 2.95 (-2.94,2.94) [-3.00,1.00] Total: 20487 W: 4558 L: 4434 D: 11495 http://tests.stockfishchess.org/tests/view/5a7706ec0ebc5902971a9854 • Passed LTC: LLR: 2.96 (-2.94,2.94) [-3.00,1.00] Total: 41960 W: 7145 L: 7058 D: 27757 http://tests.stockfishchess.org/tests/view/5a778c830ebc5902971a9895 • Passed an additional non-regression [-5..0] test at the time control of 60sec for the game (sudden death) with disabled draw adjudication: LLR: 2.95 (-2.94,2.94) [-5.00,0.00] Total: 8438 W: 1675 L: 1586 D: 5177 http://tests.stockfishchess.org/tests/view/5a7c3d8d0ebc5902971a9ac0 • Passed an additional non-regression [-5..0] test at the time control of 1sec+1sec per move with disabled draw adjudication: LLR: 2.97 (-2.94,2.94) [-5.00,0.00] Total: 27664 W: 5575 L: 5574 D: 16515 http://tests.stockfishchess.org/tests/view/5a7c3e820ebc5902971a9ac3 This is a functional change for the time management code. Bench: 4983414
Copyright 2011–2024 Next Chess Move LLC