Dev Builds » 20190627-0746

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 08:37:16 1974026 3380 2017 67 1296 +228.56 +/- 9.36
ncm-et-4 08:34:34 1955913 3316 2007 64 1245 +233.29 +/- 9.55
ncm-et-9 08:35:57 1970444 3314 1912 66 1336 +218.37 +/- 9.19
ncm-et-10 08:32:24 1965833 3303 1962 79 1262 +225.02 +/- 9.51
ncm-et-13 08:35:14 1972410 3322 1947 77 1298 +221.35 +/- 9.36
ncm-et-15 08:35:08 1974155 3365 1989 81 1295 +223.44 +/- 9.38
20000 11834 434 7732 +224.97 +/- 3.83

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
86882 ncm-et-10 2019-06-27 21:57 00:45:44 1980391 303 188 4 111 +244.79 +/- 32.09
86881 ncm-et-4 2019-06-27 21:54 00:48:57 1961843 316 193 8 115 +233.03 +/- 31.74
86880 ncm-et-9 2019-06-27 21:53 00:49:49 1964152 314 192 9 113 +231.63 +/- 32.08
86879 ncm-et-13 2019-06-27 21:53 00:49:48 1970647 322 182 6 134 +213.15 +/- 29.02
86878 ncm-et-15 2019-06-27 21:48 00:54:55 1980473 365 215 6 144 +226.31 +/- 28.06
86877 ncm-et-3 2019-06-27 21:45 00:57:35 1978289 380 222 9 149 +220.14 +/- 27.72
86876 ncm-et-10 2019-06-27 20:38 01:18:24 1955261 500 294 13 193 +220.88 +/- 24.38
86875 ncm-et-4 2019-06-27 20:35 01:17:13 1964765 500 324 8 168 +258.75 +/- 26.19
86874 ncm-et-13 2019-06-27 20:34 01:17:29 1967084 500 297 7 196 +230.16 +/- 23.99
86873 ncm-et-9 2019-06-27 20:34 01:17:42 1975155 500 282 10 208 +211.87 +/- 23.28
86872 ncm-et-15 2019-06-27 20:31 01:15:49 1975136 500 295 16 189 +218.86 +/- 24.73
86871 ncm-et-3 2019-06-27 20:29 01:15:08 1964147 500 297 7 196 +230.16 +/- 23.99
86870 ncm-et-10 2019-06-27 19:20 01:16:20 1967616 500 299 10 191 +229.12 +/- 24.45
86869 ncm-et-13 2019-06-27 19:17 01:16:38 1972038 500 283 12 205 +210.89 +/- 23.54
86868 ncm-et-9 2019-06-27 19:15 01:17:44 1968938 500 283 8 209 +214.85 +/- 23.15
86867 ncm-et-4 2019-06-27 19:15 01:18:54 1942752 500 283 11 206 +211.87 +/- 23.44
86866 ncm-et-15 2019-06-27 19:14 01:15:57 1969556 500 290 6 204 +223.94 +/- 23.4
86865 ncm-et-3 2019-06-27 19:11 01:17:10 1970014 500 299 6 195 +233.32 +/- 24.03
86864 ncm-et-10 2019-06-27 18:02 01:17:24 1964161 500 298 9 193 +229.12 +/- 24.27
86863 ncm-et-13 2019-06-27 17:59 01:16:57 1967849 500 299 10 191 +229.12 +/- 24.45
86862 ncm-et-4 2019-06-27 17:58 01:16:03 1964000 500 297 12 191 +224.97 +/- 24.5
86861 ncm-et-9 2019-06-27 17:57 01:17:37 1967102 500 298 11 191 +227.04 +/- 24.48
86860 ncm-et-15 2019-06-27 17:55 01:17:26 1970955 500 301 12 187 +229.12 +/- 24.79
86859 ncm-et-3 2019-06-27 17:54 01:16:16 1977135 500 298 8 194 +230.16 +/- 24.17
86858 ncm-et-10 2019-06-27 16:44 01:16:51 1969239 500 298 19 183 +218.86 +/- 25.2
86857 ncm-et-4 2019-06-27 16:40 01:16:44 1972678 500 290 9 201 +220.88 +/- 23.72
86856 ncm-et-13 2019-06-27 16:38 01:19:04 1969550 500 293 16 191 +216.85 +/- 24.59
86855 ncm-et-15 2019-06-27 16:38 01:16:19 1968481 500 279 17 204 +202.15 +/- 23.72
86854 ncm-et-9 2019-06-27 16:37 01:18:13 1978956 500 289 10 201 +218.86 +/- 23.75
86853 ncm-et-3 2019-06-27 16:35 01:17:23 1972540 500 320 9 171 +253.02 +/- 25.97
86852 ncm-et-10 2019-06-27 15:24 01:19:10 1958010 500 287 12 201 +214.85 +/- 23.8
86851 ncm-et-4 2019-06-27 15:22 01:17:41 1962766 500 309 8 183 +241.91 +/- 24.98
86850 ncm-et-9 2019-06-27 15:20 01:16:12 1972650 500 294 8 198 +226.0 +/- 23.89
86849 ncm-et-15 2019-06-27 15:20 01:17:02 1979192 500 317 10 173 +248.53 +/- 25.83
86848 ncm-et-13 2019-06-27 15:19 01:17:51 1980912 500 305 13 182 +232.26 +/- 25.19
86847 ncm-et-3 2019-06-27 15:18 01:16:17 1990522 500 289 16 195 +212.86 +/- 24.31
86846 ncm-et-10 2019-06-27 14:04 01:18:31 1966156 500 298 12 190 +226.0 +/- 24.57
86845 ncm-et-4 2019-06-27 14:02 01:19:02 1922587 500 311 8 181 +244.1 +/- 25.14
86844 ncm-et-15 2019-06-27 14:01 01:17:40 1975297 500 292 14 194 +217.85 +/- 24.34
86843 ncm-et-13 2019-06-27 14:01 01:17:27 1978793 500 288 13 199 +214.85 +/- 23.97
86842 ncm-et-9 2019-06-27 14:01 01:18:40 1966157 500 274 10 216 +204.07 +/- 22.77
86841 ncm-et-3 2019-06-27 13:59 01:17:27 1965536 500 292 12 196 +219.87 +/- 24.15

Commit

Commit ID d889bb47185e33012b45cab63359952247bc86e2
Author protonspring
Date 2019-06-27 07:46:36 UTC
Bonus for double attacks on unsupported pawns This is a functional change that rewards double attacks on an unsupported pawns. STC (non-functional difference) LLR: 2.96 (-2.94,2.94) [0.50,4.50] Total: 83276 W: 18981 L: 18398 D: 45897 http://tests.stockfishchess.org/tests/view/5d0970500ebc5925cf0a77d4 LTC (incomplete looping version) LLR: 0.50 (-2.94,2.94) [0.00,3.50] Total: 82999 W: 14244 L: 13978 D: 54777 http://tests.stockfishchess.org/tests/view/5d0a8d480ebc5925cf0a8d58 LTC (completed non-looping version). LLR: 2.96 (-2.94,2.94) [0.00,3.50] Total: 223381 W: 38323 L: 37512 D: 147546 http://tests.stockfishchess.org/tests/view/5d0e80510ebc5925cf0ad320 Closes https://github.com/official-stockfish/Stockfish/pull/2205 Bench 3633546 ---------------------------------- Comments by Alain SAVARD: interesting result ! I would have expected that search would resolve such positions correctly on the very next move. This is not a very common pattern, and when it happens, it will quickly disappear. So I'm quite surprised that it passed LTC. I would be even more surprised if this would resist a simplification. Anyway, let's try to imagine a few cases. a) If you have White d5 f5 against Black e6, and White to move last move by Black was probably a capture on e6 and White is about to recapture on e6 b) If you have White d5 f5 against e6, and Black to move last move by White was possibly a capture on d5 or f5 or the pawn on e6 was pinned or could not move for some reason. and white wants to blast open the position and just pushed d4-d5 or f4-f5 Some possible follow-ups a) Motif is so rare that the popcount() can be safely replaced with a bool() But this would not pass a SPRT[0,4], So try a simplification with bool() and also without the & ~theirAttacks b) If it works, we probably can simply have this in the loop if (lever) score += S(0, 20); c) remove all this and tweak something in search for pawn captures (priority, SEE, extension,..)
Copyright 2011–2024 Next Chess Move LLC