Dev Builds » 20191123-2334

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 08:25:28 1974141 3327 2038 52 1237 +239.16 +/- 9.57
ncm-et-4 08:24:51 1974890 3346 2021 57 1268 +233.83 +/- 9.45
ncm-et-9 08:26:21 1967547 3334 2044 61 1229 +238.0 +/- 9.62
ncm-et-10 08:26:36 1963547 3315 2003 61 1251 +233.23 +/- 9.52
ncm-et-13 08:24:12 1967731 3314 2014 76 1224 +232.68 +/- 9.66
ncm-et-15 08:26:41 1971761 3364 2044 57 1263 +235.8 +/- 9.47
20000 12164 364 7472 +235.45 +/- 3.9

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
105188 ncm-et-10 2019-11-25 06:14 00:47:33 1973633 315 192 8 115 +232.33 +/- 31.74
105187 ncm-et-13 2019-11-25 06:13 00:48:26 1968318 314 201 5 108 +254.28 +/- 32.71
105186 ncm-et-3 2019-11-25 06:12 00:49:17 1982731 327 189 6 132 +219.68 +/- 29.31
105185 ncm-et-9 2019-11-25 06:10 00:50:50 1970797 334 200 6 128 +230.6 +/- 29.87
105184 ncm-et-4 2019-11-25 06:10 00:51:40 1970959 346 198 6 142 +217.3 +/- 28.19
105183 ncm-et-15 2019-11-25 06:07 00:54:16 1977322 364 238 6 120 +261.87 +/- 31.07
105182 ncm-et-10 2019-11-25 04:57 01:15:35 1963840 500 302 15 183 +227.04 +/- 25.15
105181 ncm-et-13 2019-11-25 04:56 01:15:51 1971435 500 299 8 193 +231.21 +/- 24.24
105180 ncm-et-3 2019-11-25 04:56 01:15:12 1973122 500 318 10 172 +249.64 +/- 25.91
105179 ncm-et-4 2019-11-25 04:54 01:14:40 1979608 500 282 9 209 +212.86 +/- 23.18
105178 ncm-et-9 2019-11-25 04:52 01:17:21 1963532 500 303 5 192 +238.66 +/- 24.2
105177 ncm-et-15 2019-11-25 04:52 01:14:20 1976753 500 300 10 190 +230.16 +/- 24.52
105176 ncm-et-10 2019-11-25 03:39 01:17:12 1953587 500 302 9 189 +233.32 +/- 24.56
105175 ncm-et-3 2019-11-25 03:38 01:16:20 1971409 500 307 9 184 +238.66 +/- 24.94
105174 ncm-et-13 2019-11-25 03:38 01:17:02 1962302 500 282 14 204 +207.95 +/- 23.65
105173 ncm-et-4 2019-11-25 03:37 01:15:38 1964320 500 310 7 183 +244.1 +/- 24.95
105172 ncm-et-9 2019-11-25 03:36 01:15:04 1972805 500 305 9 186 +236.51 +/- 24.79
105171 ncm-et-15 2019-11-25 03:35 01:15:29 1968334 500 287 5 208 +221.9 +/- 23.1
105170 ncm-et-10 2019-11-25 02:23 01:15:09 1960926 500 311 6 183 +246.3 +/- 24.91
105169 ncm-et-3 2019-11-25 02:22 01:15:42 1977474 500 298 7 195 +231.21 +/- 24.06
105168 ncm-et-4 2019-11-25 02:21 01:15:00 1978762 500 306 10 184 +236.51 +/- 24.97
105167 ncm-et-13 2019-11-25 02:20 01:16:43 1970943 500 321 12 167 +250.77 +/- 26.37
105166 ncm-et-15 2019-11-25 02:20 01:14:11 1972340 500 312 11 177 +241.91 +/- 25.53
105165 ncm-et-9 2019-11-25 02:19 01:16:04 1982643 500 305 10 185 +235.45 +/- 24.89
105164 ncm-et-3 2019-11-25 01:05 01:15:52 1971886 500 313 8 179 +246.3 +/- 25.29
105163 ncm-et-13 2019-11-25 01:04 01:15:08 1965085 500 298 9 193 +229.12 +/- 24.27
105162 ncm-et-4 2019-11-25 01:03 01:16:50 1987626 500 313 8 179 +246.3 +/- 25.29
105161 ncm-et-10 2019-11-25 01:03 01:18:27 1952829 500 283 9 208 +213.85 +/- 23.25
105160 ncm-et-15 2019-11-25 01:03 01:16:19 1969087 500 308 7 185 +241.91 +/- 24.8
105159 ncm-et-9 2019-11-25 01:02 01:15:45 1947644 500 316 7 177 +250.77 +/- 25.42
105158 ncm-et-13 2019-11-24 23:48 01:15:05 1971877 500 295 13 192 +221.9 +/- 24.45
105157 ncm-et-3 2019-11-24 23:46 01:17:08 1968467 500 313 7 180 +247.41 +/- 25.18
105156 ncm-et-4 2019-11-24 23:46 01:15:52 1978501 500 302 10 188 +232.26 +/- 24.67
105155 ncm-et-10 2019-11-24 23:46 01:16:11 1977395 500 306 5 189 +241.91 +/- 24.42
105154 ncm-et-9 2019-11-24 23:45 01:15:46 1964774 500 303 15 182 +228.08 +/- 25.23
105153 ncm-et-15 2019-11-24 23:45 01:16:19 1968161 500 299 11 190 +228.08 +/- 24.55
105152 ncm-et-13 2019-11-24 22:31 01:15:57 1964160 500 318 15 167 +244.1 +/- 26.41
105151 ncm-et-4 2019-11-24 22:30 01:15:11 1964457 500 310 7 183 +244.1 +/- 24.95
105150 ncm-et-3 2019-11-24 22:29 01:15:57 1973900 500 300 5 195 +235.45 +/- 23.98
105149 ncm-et-9 2019-11-24 22:29 01:15:31 1970636 500 312 9 179 +244.1 +/- 25.32
105148 ncm-et-10 2019-11-24 22:28 01:16:29 1962624 500 307 9 184 +238.66 +/- 24.94
105147 ncm-et-15 2019-11-24 22:28 01:15:47 1970331 500 300 7 193 +233.32 +/- 24.21

Commit

Commit ID 87ed9facf16bc9d0af9daf8680801759ce1d8662
Author 31m059
Date 2019-11-23 23:34:43 UTC
King danger: retire attacked-by-bishop defense In a recent commit, "Introduce king flank defenders," a term was introduced by Michael Chaly (@Vizvezdenec) to reduce king danger based on king defenders, i.e., friendly attacks on our King Flank and Camp. This is a powerful idea and broadly applicable to all of our pieces. An earlier, but narrower, version of a similar idea was already coded into king danger, with a term reducing king danger simply if we had a bishop and king attacking the same square -- there is also a similar term for knights, but roughly three times larger. I had attempted to tweak this term's coefficient fairly recently, in a series of tests in early September which increased this coefficient. All failed STC with significantly negative scores. Now that the king flank defenders term has been introduced, it appears that the bishop-defense term can be simplified away without compensation or significant Elo loss. Where do we go from here? This PR is a natural follow-up to "Introduce king flank defenders," which proposed simplification with existing and overlapping terms, such as this one. That PR also mentioned that the coefficient it introduced appeared arbitrary, so perhaps this PR can facilitate a tweak to increase king flank defenders' coefficient. Additionally, this pull request is extremely similar to https://github.com/official-stockfish/Stockfish/pull/1821, which was (coincidentally) merged a year ago, to the day (November 23, 2018). That patch also simplified away a linear king danger tropism term, which was soon after replaced with a quadratic term by @Vizvezdenec (which would not have passed without the simplification). @Vizvezdenec, again by coincidence, has recently been trying to implement a quadratic term, this time for defenders rather than attackers. This history of this evaluation code suggests that this simplification might be enough to help a patch for quadratic king-flank defenders pass. Bench: 4959670 STC: LLR: 2.94 (-2.94,2.94) [-3.00,1.00] Total: 22209 W: 4920 L: 4800 D: 12489 https://tests.stockfishchess.org/tests/view/5dd444d914339111b9b6bed7 LTC: LLR: 2.95 (-2.94,2.94) [-3.00,1.00] Total: 152107 W: 24658 L: 24743 D: 102706 https://tests.stockfishchess.org/tests/view/5dd4be31f531e81cf278ea9d Interesting discussion on Github about this pull request: https://github.com/official-stockfish/Stockfish/pull/2424 --- This pull request was opened less than one week before the holiday of Thanksgiving here in the United States. In keeping with the holiday tradition of expressing gratitude, I would like to thank our generous CPU donors, talented forum contributors, innovative developers, speedy fishtest approvers, and especially our hardworking server maintainers (@ppigazzini and @tomtor). Thank you all for a year of great Stockfish progress!
Copyright 2011–2024 Next Chess Move LLC