Dev Builds » 20180406-2331

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 06:02:42 2001283 2386 1117 72 1197 +163.2 +/- 9.55
ncm-et-4 06:02:30 1998104 2390 1062 94 1234 +149.27 +/- 9.42
ncm-et-5 01:47:42 2022549 720 338 14 368 +168.4 +/- 17.05
ncm-et-6 01:47:46 2026309 724 346 22 356 +167.32 +/- 17.59
ncm-et-7 01:47:24 2025408 722 318 32 372 +145.59 +/- 17.23
ncm-et-8 01:47:16 1984807 709 324 24 361 +156.87 +/- 17.42
ncm-et-9 06:02:54 2001815 2408 1093 84 1231 +155.13 +/- 9.42
ncm-et-10 06:02:54 1983716 2359 1057 81 1221 +152.91 +/- 9.44
ncm-et-11 01:47:31 1991221 709 335 22 352 +164.7 +/- 17.67
ncm-et-12 01:47:36 2020349 702 325 19 358 +162.31 +/- 17.4
ncm-et-13 06:02:13 1959507 2361 1068 55 1238 +159.38 +/- 9.28
ncm-et-14 01:47:24 2018314 712 327 23 362 +158.49 +/- 17.38
ncm-et-15 06:02:20 1994296 2384 1091 72 1221 +158.69 +/- 9.43
ncm-et-16 01:47:39 2020023 714 334 26 354 +160.37 +/- 17.68
20000 9135 640 10225 +157.55 +/- 3.26

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
51420 ncm-et-10 2018-08-27 14:42 00:23:19 1982321 145 73 4 68 +179.84 +/- 40.66
51419 ncm-et-13 2018-08-27 14:41 00:24:28 1770435 139 56 2 81 +142.46 +/- 35.28
51418 ncm-et-3 2018-08-27 14:39 00:26:08 1988772 170 79 4 87 +164.58 +/- 35.33
51417 ncm-et-15 2018-08-27 14:39 00:26:01 1983733 173 79 7 87 +153.94 +/- 35.83
51416 ncm-et-4 2018-08-27 14:37 00:28:09 1987350 184 81 8 95 +145.84 +/- 34.19
51415 ncm-et-9 2018-08-27 14:36 00:29:26 1987667 189 80 5 104 +145.88 +/- 31.91
51414 ncm-et-10 2018-08-27 13:24 01:16:56 1980130 500 243 11 246 +174.55 +/- 21.04
51413 ncm-et-13 2018-08-27 13:22 01:17:40 1958430 500 237 12 251 +168.4 +/- 20.8
51412 ncm-et-3 2018-08-27 13:22 01:16:05 1988140 500 231 13 256 +162.35 +/- 20.56
51411 ncm-et-15 2018-08-27 13:22 01:16:03 1981538 500 236 12 252 +167.53 +/- 20.74
51410 ncm-et-4 2018-08-27 13:20 01:16:10 1986721 500 209 19 272 +138.99 +/- 19.9
51409 ncm-et-9 2018-08-27 13:18 01:16:48 1990667 500 229 23 248 +152.18 +/- 21.25
51408 ncm-et-10 2018-08-27 12:07 01:16:38 1981381 500 207 17 276 +138.99 +/- 19.64
51407 ncm-et-15 2018-08-27 12:05 01:15:49 1981539 500 212 19 269 +141.44 +/- 20.05
51406 ncm-et-13 2018-08-27 12:05 01:16:05 1987826 500 231 12 257 +163.21 +/- 20.47
51405 ncm-et-3 2018-08-27 12:05 01:16:10 1988138 500 246 17 237 +171.9 +/- 21.72
51404 ncm-et-9 2018-08-27 12:03 01:13:46 1987982 500 228 10 262 +162.35 +/- 20.14
51403 ncm-et-4 2018-08-27 12:03 01:15:54 1988612 500 229 15 256 +158.93 +/- 20.62
51402 ncm-et-15 2018-08-27 10:47 01:16:45 1981851 500 212 12 276 +147.19 +/- 19.49
51401 ncm-et-4 2018-08-27 10:47 01:14:47 1991298 500 221 20 259 +148.02 +/- 20.6
51400 ncm-et-13 2018-08-27 10:47 01:16:56 1990190 500 202 9 289 +141.44 +/- 18.72
51399 ncm-et-10 2018-08-27 10:47 01:18:18 1920865 500 215 20 265 +143.07 +/- 20.28
51398 ncm-et-3 2018-08-27 10:47 01:16:48 1990032 500 233 20 247 +158.08 +/- 21.24
51397 ncm-et-9 2018-08-27 10:47 01:15:11 1991140 500 230 21 249 +154.7 +/- 21.15
16220 ncm-et-4 2018-04-07 01:33 00:30:23 2022794 206 82 14 110 +119.15 +/- 31.94
16219 ncm-et-11 2018-04-07 01:32 00:30:49 2017097 209 93 7 109 +151.97 +/- 31.66
16218 ncm-et-12 2018-04-07 01:32 00:31:20 2021162 202 92 9 101 +151.72 +/- 33.33
16217 ncm-et-15 2018-04-07 01:32 00:31:26 2018720 211 110 4 97 +191.95 +/- 33.83
16216 ncm-et-14 2018-04-07 01:31 00:31:46 2017745 212 95 5 112 +157.46 +/- 30.93
16215 ncm-et-8 2018-04-07 01:31 00:31:35 1992426 209 98 7 104 +162.1 +/- 32.66
16214 ncm-et-13 2018-04-07 01:31 00:32:02 2025411 222 108 7 107 +170.57 +/- 32.29
16213 ncm-et-10 2018-04-07 01:31 00:32:17 2019370 214 98 10 106 +151.85 +/- 32.6
16212 ncm-et-6 2018-04-07 01:31 00:32:25 2026882 224 114 8 102 +178.65 +/- 33.37
16211 ncm-et-9 2018-04-07 01:31 00:32:31 2026391 219 90 4 125 +144.18 +/- 28.63
16210 ncm-et-3 2018-04-07 01:30 00:32:38 2025899 216 95 7 114 +150.27 +/- 30.86
16209 ncm-et-7 2018-04-07 01:30 00:32:47 2025572 222 94 11 117 +136.51 +/- 30.78
16208 ncm-et-16 2018-04-07 01:30 00:32:52 2020185 214 95 7 112 +151.85 +/- 31.19
16207 ncm-et-5 2018-04-07 01:30 00:33:10 2022956 220 95 5 120 +150.97 +/- 29.66
16206 ncm-et-13 2018-04-07 00:15 01:15:02 2024754 500 234 13 253 +164.93 +/- 20.72
16205 ncm-et-7 2018-04-07 00:15 01:14:37 2025244 500 224 21 255 +149.68 +/- 20.83
16204 ncm-et-14 2018-04-07 00:15 01:15:38 2018883 500 232 18 250 +158.93 +/- 21.03
16203 ncm-et-3 2018-04-07 00:15 01:14:53 2026718 500 233 11 256 +165.8 +/- 20.49
16202 ncm-et-11 2018-04-07 00:14 01:16:42 1965346 500 242 15 243 +170.15 +/- 21.33
16201 ncm-et-8 2018-04-07 00:14 01:15:41 1977188 500 226 17 257 +154.7 +/- 20.62
16200 ncm-et-12 2018-04-07 00:14 01:16:16 2019537 500 233 10 257 +166.66 +/- 20.41
16199 ncm-et-10 2018-04-07 00:14 01:15:26 2018232 500 221 19 260 +148.85 +/- 20.52
16198 ncm-et-4 2018-04-07 00:14 01:17:07 2011850 500 240 18 242 +165.8 +/- 21.46
16197 ncm-et-5 2018-04-07 00:14 01:14:32 2022143 500 243 9 248 +176.33 +/- 20.86
16196 ncm-et-16 2018-04-07 00:14 01:14:47 2019861 500 239 19 242 +164.07 +/- 21.49
16195 ncm-et-15 2018-04-07 00:14 01:16:16 2018396 500 242 18 240 +167.53 +/- 21.57
16194 ncm-et-9 2018-04-07 00:14 01:15:12 2027046 500 236 21 243 +159.78 +/- 21.48
16193 ncm-et-6 2018-04-07 00:14 01:15:21 2025737 500 232 14 254 +162.35 +/- 20.7

Commit

Commit ID 759d129e4760938ded462a05061b421e18990937
Author Mark Tenzer
Date 2018-04-06 23:31:23 UTC
Introduce Overload This patch applies a S(10, 5) bonus for every square that is: - Occupied by an enemy piece which is not a pawn - Attacked exactly once by our pieces - Defended exactly once by enemy pieces The idea is that these pieces must be defended. Their defenders have dramatically limited mobility, and they are vulnerable to our future attack. As with connectivity, there are probably many more tests to be run in this area. In particular: - I believe @snicolet's queen overload tests have demonstrated a potential need for a queen overload bonus above and beyond this one; however, the conditions for "overload" in this patch are different (excluding pieces we attack twice). My next test after this is (hopefully) merged will be to intersect the Bitboard I define here with the enemy's queen attacks and attempt to give additional bonus. - Perhaps we should exclude pieces attacked by pawns--can pawns really be overloaded? Should they have the same weight, or less? This didn't work with a previous version, but it could work with this one. - More generally, different pieces may need more or less bonus. We could change bonuses based on what type of enemy piece is being overloaded, what type of friendly piece is attacking, and/or what type of piece is being defended by the overloaded piece and attacked by us, or any intersection of these three. For example, here attacked/defended pawns are excluded, but they're not totally worthless targets, and could be added again with a smaller bonus. - This list is by no means exhaustive. STC: LLR: 2.96 (-2.94,2.94) [0.00,5.00] Total: 17439 W: 3599 L: 3390 D: 10450 http://tests.stockfishchess.org/tests/view/5ac78a2e0ebc59435923735e LTC: LLR: 2.95 (-2.94,2.94) [0.00,5.00] Total: 43304 W: 6533 L: 6256 D: 30515 http://tests.stockfishchess.org/tests/view/5ac7a1d80ebc59435923736f Closes https://github.com/official-stockfish/Stockfish/pull/1533 Bench: 5248871 ---------------- This is my first time opening a PR, so I apologize if there are errors. There are too many people to thank since I submitted my first test just over a month ago. Thank you all for the warm welcome and here is to more green patches! In particular, I would like to thank: - @crossbr, whose comment in a FishCooking thread first inspired me to consider the overloading of pieces other than queens, - @snicolet, whose queen overload tests inspired this one and served as the base of my first overload attempts, - @protonspring, whose connectivity tests inspired this one and who provided much of the feedback needed to take this from red to green, - @vondele, who kindly corrected me when I submitted a bad LTC test, - @Rocky640, who has helped me over and over again in the past month. Thank you all!
Copyright 2011–2024 Next Chess Move LLC