Dev Builds » 20180412-0737

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 06:03:37 2002034 2401 1151 72 1178 +168.14 +/- 9.66
ncm-et-4 06:04:01 1994740 2397 1092 71 1234 +158.06 +/- 9.37
ncm-et-5 01:48:13 2024265 718 321 20 377 +155.22 +/- 16.88
ncm-et-6 01:48:06 2025490 716 322 17 377 +158.07 +/- 16.82
ncm-et-7 01:47:14 2024020 712 332 21 359 +162.69 +/- 17.44
ncm-et-8 01:47:15 2016310 715 347 24 344 +169.16 +/- 17.98
ncm-et-9 06:03:37 2001881 2418 1139 74 1205 +164.26 +/- 9.53
ncm-et-10 06:03:27 1980800 2351 1086 76 1189 +159.62 +/- 9.59
ncm-et-11 01:47:24 2000734 715 341 19 355 +168.55 +/- 17.54
ncm-et-12 01:47:32 2017680 717 338 21 358 +164.98 +/- 17.49
ncm-et-13 06:04:10 1978491 2360 1084 70 1206 +159.64 +/- 9.49
ncm-et-14 01:47:55 2018234 710 295 23 392 +140.25 +/- 16.45
ncm-et-15 06:03:27 1985323 2364 1118 80 1166 +163.68 +/- 9.73
ncm-et-16 01:47:11 1998428 706 300 31 375 +139.41 +/- 17.08
20000 9266 619 10115 +160.79 +/- 3.28

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
51348 ncm-et-10 2018-08-26 19:20 00:22:09 1947974 138 60 3 75 +152.62 +/- 37.59
51347 ncm-et-13 2018-08-26 19:20 00:22:19 1986563 139 64 3 72 +163.57 +/- 38.74
51346 ncm-et-15 2018-08-26 19:19 00:23:12 1983262 152 85 3 64 +209.65 +/- 42.19
51345 ncm-et-4 2018-08-26 19:15 00:27:03 1989245 183 84 2 97 +167.57 +/- 32.83
51344 ncm-et-3 2018-08-26 19:13 00:28:27 1991930 190 90 7 93 +162.71 +/- 34.7
51343 ncm-et-9 2018-08-26 19:12 00:30:15 1990348 198 94 6 98 +165.99 +/- 33.61
51342 ncm-et-13 2018-08-26 18:02 01:16:20 1960440 500 235 13 252 +165.8 +/- 20.78
51341 ncm-et-15 2018-08-26 18:01 01:16:43 1980442 500 214 17 269 +144.72 +/- 20.0
51340 ncm-et-10 2018-08-26 17:59 01:20:09 1940687 500 244 14 242 +172.78 +/- 21.36
51339 ncm-et-4 2018-08-26 17:57 01:16:56 1990032 500 224 12 264 +157.24 +/- 20.11
51338 ncm-et-9 2018-08-26 17:57 01:14:04 1988927 500 234 15 251 +163.21 +/- 20.89
51337 ncm-et-3 2018-08-26 17:55 01:17:02 1987193 500 249 14 237 +177.22 +/- 21.64
51336 ncm-et-15 2018-08-26 16:42 01:17:27 1980285 500 228 15 257 +158.08 +/- 20.57
51335 ncm-et-13 2018-08-26 16:42 01:19:04 1951146 500 218 16 266 +148.85 +/- 20.13
51334 ncm-et-9 2018-08-26 16:41 01:14:52 1990664 500 241 16 243 +168.4 +/- 21.36
51333 ncm-et-4 2018-08-26 16:40 01:15:24 1989875 500 219 25 256 +142.26 +/- 20.86
51332 ncm-et-10 2018-08-26 16:40 01:17:42 1974524 500 230 15 255 +159.78 +/- 20.67
51331 ncm-et-3 2018-08-26 16:38 01:15:59 1990823 500 248 13 239 +177.22 +/- 21.49
51330 ncm-et-4 2018-08-26 15:23 01:16:35 1988138 500 243 12 245 +173.67 +/- 21.13
51329 ncm-et-3 2018-08-26 15:23 01:14:18 1991773 500 224 16 260 +153.86 +/- 20.44
51328 ncm-et-13 2018-08-26 15:23 01:18:13 1921002 500 241 12 247 +171.9 +/- 21.02
51327 ncm-et-15 2018-08-26 15:23 01:18:23 1927908 500 251 21 228 +172.78 +/- 22.32
51326 ncm-et-9 2018-08-26 15:23 01:16:48 1989875 500 240 19 241 +164.93 +/- 21.54
51325 ncm-et-10 2018-08-26 15:23 01:15:58 1980598 500 236 17 247 +163.21 +/- 21.16
18180 ncm-et-12 2018-04-12 14:44 00:31:09 2014036 217 103 3 111 +173.15 +/- 30.97
18179 ncm-et-16 2018-04-12 14:43 00:31:28 2020673 206 99 10 97 +160.65 +/- 34.31
18178 ncm-et-14 2018-04-12 14:43 00:31:44 2018883 210 91 5 114 +151.15 +/- 30.49
18177 ncm-et-8 2018-04-12 14:43 00:31:48 2009175 215 101 6 108 +164.87 +/- 31.88
18176 ncm-et-11 2018-04-12 14:43 00:31:56 2016123 215 95 5 115 +154.96 +/- 30.44
18175 ncm-et-10 2018-04-12 14:43 00:32:01 2020184 213 95 7 111 +152.66 +/- 31.36
18174 ncm-et-15 2018-04-12 14:43 00:32:05 2018883 212 108 6 98 +182.21 +/- 33.86
18173 ncm-et-3 2018-04-12 14:42 00:32:15 2024590 211 93 7 111 +150.34 +/- 31.32
18172 ncm-et-4 2018-04-12 14:42 00:32:25 1988687 214 98 5 111 +161.74 +/- 31.17
18171 ncm-et-7 2018-04-12 14:42 00:32:20 2023287 212 103 6 103 +171.7 +/- 32.82
18170 ncm-et-6 2018-04-12 14:42 00:32:43 2026718 216 95 4 117 +156.09 +/- 29.97
18169 ncm-et-9 2018-04-12 14:42 00:32:51 2025901 220 96 5 119 +152.87 +/- 29.83
18168 ncm-et-13 2018-04-12 14:42 00:33:02 2025736 221 108 5 108 +175.47 +/- 31.88
18167 ncm-et-5 2018-04-12 14:41 00:33:31 2025245 218 103 5 110 +168.2 +/- 31.44
18166 ncm-et-16 2018-04-12 13:26 01:15:43 1976184 500 201 21 278 +130.94 +/- 19.65
18165 ncm-et-8 2018-04-12 13:26 01:15:27 2023446 500 246 18 236 +171.02 +/- 21.8
18164 ncm-et-7 2018-04-12 13:26 01:14:54 2024754 500 229 15 256 +158.93 +/- 20.62
18163 ncm-et-10 2018-04-12 13:26 01:15:28 2020837 500 221 20 259 +148.02 +/- 20.6
18162 ncm-et-11 2018-04-12 13:26 01:15:28 1985345 500 246 14 240 +174.55 +/- 21.47
18161 ncm-et-12 2018-04-12 13:26 01:16:23 2021325 500 235 18 247 +161.49 +/- 21.19
18160 ncm-et-9 2018-04-12 13:26 01:14:47 2025572 500 234 13 253 +164.93 +/- 20.72
18159 ncm-et-15 2018-04-12 13:26 01:15:37 2021162 500 232 18 250 +158.93 +/- 21.03
18158 ncm-et-14 2018-04-12 13:26 01:16:11 2017585 500 204 18 278 +135.76 +/- 19.57
18157 ncm-et-3 2018-04-12 13:26 01:15:36 2025899 500 247 15 238 +174.55 +/- 21.61
18156 ncm-et-4 2018-04-12 13:26 01:15:38 2022468 500 224 15 261 +154.7 +/- 20.36
18155 ncm-et-6 2018-04-12 13:26 01:15:23 2024263 500 227 13 260 +158.93 +/- 20.35
18154 ncm-et-13 2018-04-12 13:25 01:15:12 2026063 500 218 21 261 +144.72 +/- 20.52
18153 ncm-et-5 2018-04-12 13:25 01:14:42 2023285 500 218 15 267 +149.68 +/- 20.04

Commit

Commit ID 6413d9b1f965ed73154fdffb36476c82a2c66c96
Author Marco Costalba
Date 2018-04-12 07:37:18 UTC
Further documentation and coding style on TB code This patch adds some documentation and code cleanup to tablebase code. It took me some time to understand the relation among the differrent structs, although I have rewrote them fully in the past. So I wrote some detailed documentation to avoid the same efforts for future readers. Also noteworthy is the use a standard hash table implementation with a more efficient 1D array instead of a 2D array. This reduces the average lookup steps of 90% (from 343 to 38 in a bench 128 1 16 run) and reduces also the table from 5K to 4K entries. I have tested on 5-men and no functional and no slowdown reported. It should be verified on 6-men that the new hash does not overflow. It is enough to run ./stockfish with 6-men available: if it does not assert at startup it means everything is ok with 6-men too. EDIT: verified for 6-men tablebase by Jörg Oster. Thanks! No functional change.
Copyright 2011–2024 Next Chess Move LLC