Dev Builds » 20141112-2116

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 1989024 2381 179 666 1536 -72.08 +/- 8.15
ncm-et-4 06:02:24 1965114 2359 175 662 1522 -72.77 +/- 8.18
ncm-et-5 01:47:03 2023529 708 47 216 445 -84.56 +/- 15.23
ncm-et-6 01:47:08 2026638 716 48 207 461 -78.46 +/- 14.84
ncm-et-7 01:46:59 2023937 725 51 215 459 -79.98 +/- 14.99
ncm-et-8 01:46:50 1995023 708 59 199 450 -69.62 +/- 15.2
ncm-et-9 06:02:56 1992311 2389 188 669 1532 -70.92 +/- 8.19
ncm-et-10 06:02:13 1981839 2374 185 679 1510 -73.37 +/- 8.27
ncm-et-11 01:47:05 2018558 713 66 207 440 -69.62 +/- 15.57
ncm-et-12 01:46:48 2021899 711 41 213 457 -85.75 +/- 14.84
ncm-et-13 06:02:49 1990277 2393 199 724 1470 -77.48 +/- 8.5
ncm-et-14 01:46:51 2018720 726 45 215 466 -82.89 +/- 14.74
ncm-et-15 06:02:47 1994005 2389 187 686 1516 -73.65 +/- 8.27
ncm-et-16 01:47:00 2016001 708 44 199 465 -77.31 +/- 14.62
20000 1514 5757 12729 -74.85 +/- 2.85

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
84254 ncm-et-4 2019-06-06 11:57 00:22:10 1931326 141 12 34 95 -54.66 +/- 32.4
84253 ncm-et-10 2019-06-06 11:55 00:24:12 1962920 150 9 50 91 -97.44 +/- 34.05
84252 ncm-et-3 2019-06-06 11:52 00:27:02 1961535 174 10 46 118 -72.94 +/- 28.56
84251 ncm-et-9 2019-06-06 11:52 00:27:19 1977007 178 15 54 109 -77.38 +/- 31.34
84250 ncm-et-15 2019-06-06 11:52 00:27:47 1989770 179 19 46 114 -52.81 +/- 30.47
84249 ncm-et-13 2019-06-06 11:52 00:27:47 1977014 178 12 52 114 -79.43 +/- 29.97
84248 ncm-et-4 2019-06-06 10:38 01:18:35 1922559 500 36 144 320 -76.25 +/- 17.9
84247 ncm-et-10 2019-06-06 10:37 01:16:43 1966635 500 45 156 299 -78.44 +/- 19.05
84246 ncm-et-3 2019-06-06 10:36 01:15:40 1965690 500 42 139 319 -68.27 +/- 18.04
84245 ncm-et-13 2019-06-06 10:35 01:15:21 1968471 500 39 151 310 -79.17 +/- 18.43
84244 ncm-et-9 2019-06-06 10:35 01:16:23 1989456 500 39 144 317 -74.06 +/- 18.09
84243 ncm-et-15 2019-06-06 10:35 01:16:10 1967545 500 41 138 321 -68.27 +/- 17.93
84242 ncm-et-10 2019-06-06 09:20 01:16:26 1962794 500 22 157 321 -96.19 +/- 17.56
84241 ncm-et-4 2019-06-06 09:19 01:17:05 1967234 500 37 147 316 -77.71 +/- 18.11
84240 ncm-et-13 2019-06-06 09:18 01:15:55 1971878 500 39 154 307 -81.37 +/- 18.58
84239 ncm-et-3 2019-06-06 09:18 01:16:40 1968313 500 36 139 325 -72.61 +/- 17.65
84238 ncm-et-9 2019-06-06 09:18 01:15:48 1966006 500 39 134 327 -66.82 +/- 17.6
84237 ncm-et-15 2019-06-06 09:16 01:17:23 1991990 500 39 160 301 -85.78 +/- 18.87
84236 ncm-et-10 2019-06-06 08:01 01:17:50 1958319 500 42 143 315 -71.16 +/- 18.23
84235 ncm-et-15 2019-06-06 08:01 01:14:23 1974520 500 39 137 324 -68.99 +/- 17.75
84234 ncm-et-4 2019-06-06 08:01 01:17:40 1961693 500 27 135 338 -76.25 +/- 16.84
84233 ncm-et-3 2019-06-06 08:01 01:16:19 1983355 500 34 147 319 -79.9 +/- 17.91
84232 ncm-et-13 2019-06-06 08:01 01:16:38 1975775 500 42 162 296 -85.04 +/- 19.15
84231 ncm-et-9 2019-06-06 08:00 01:16:12 1968948 500 43 132 325 -62.51 +/- 17.76
19548 ncm-et-12 2018-04-16 08:33 00:30:43 2021814 211 10 69 132 -99.81 +/- 27.75
19547 ncm-et-6 2018-04-16 08:33 00:30:55 2028522 216 16 67 133 -83.61 +/- 28.2
19546 ncm-et-15 2018-04-16 08:32 00:31:24 2020836 210 11 55 144 -73.89 +/- 25.61
19545 ncm-et-8 2018-04-16 08:32 00:31:22 2024101 208 19 53 136 -57.31 +/- 27.48
19544 ncm-et-3 2018-04-16 08:32 00:31:41 2024590 207 11 57 139 -78.52 +/- 26.36
19543 ncm-et-5 2018-04-16 08:32 00:32:06 2022630 208 13 62 133 -83.41 +/- 27.68
19542 ncm-et-9 2018-04-16 08:32 00:32:11 2026718 211 13 57 141 -73.53 +/- 26.37
19541 ncm-et-4 2018-04-16 08:31 00:32:15 1984265 218 21 59 138 -61.19 +/- 27.66
19540 ncm-et-11 2018-04-16 08:31 00:32:14 2019371 213 25 62 126 -60.97 +/- 29.67
19539 ncm-et-16 2018-04-16 08:31 00:32:30 2013119 208 13 71 124 -99.52 +/- 29.32
19538 ncm-et-10 2018-04-16 08:31 00:32:30 2020184 224 16 60 148 -69.14 +/- 26.01
19537 ncm-et-13 2018-04-16 08:31 00:33:00 2023446 215 20 66 129 -75.5 +/- 29.04
19536 ncm-et-7 2018-04-16 08:30 00:33:26 2025572 225 14 75 136 -96.61 +/- 27.88
19535 ncm-et-14 2018-04-16 08:29 00:34:05 2018557 226 16 74 136 -91.2 +/- 28.03
19534 ncm-et-12 2018-04-16 07:16 01:16:05 2021984 500 31 144 325 -79.9 +/- 17.56
19533 ncm-et-4 2018-04-16 07:16 01:14:39 2023612 500 42 143 315 -71.16 +/- 18.23
19532 ncm-et-16 2018-04-16 07:16 01:14:30 2018884 500 31 128 341 -68.27 +/- 16.77
19531 ncm-et-5 2018-04-16 07:16 01:14:57 2024428 500 34 154 312 -85.04 +/- 18.25
19530 ncm-et-15 2018-04-16 07:16 01:15:40 2019371 500 38 150 312 -79.17 +/- 18.32
19529 ncm-et-8 2018-04-16 07:15 01:15:28 1965946 500 40 146 314 -74.79 +/- 18.25
19528 ncm-et-7 2018-04-16 07:15 01:13:33 2022303 500 37 140 323 -72.61 +/- 17.77
19527 ncm-et-10 2018-04-16 07:15 01:14:32 2020184 500 51 113 336 -43.3 +/- 17.31
19526 ncm-et-14 2018-04-16 07:15 01:12:46 2018884 500 29 141 330 -79.17 +/- 17.28
19525 ncm-et-3 2018-04-16 07:15 01:15:20 2030662 500 46 138 316 -64.66 +/- 18.24
19524 ncm-et-13 2018-04-16 07:15 01:14:08 2025081 500 47 139 314 -64.66 +/- 18.35
19523 ncm-et-11 2018-04-16 07:15 01:14:51 2017746 500 41 145 314 -73.34 +/- 18.27
19522 ncm-et-6 2018-04-16 07:15 01:16:13 2024754 500 32 140 328 -76.25 +/- 17.44
19521 ncm-et-9 2018-04-16 07:15 01:15:03 2025735 500 39 148 313 -76.98 +/- 18.29

Commit

Commit ID 4739037f967ac3c818907e89cc88c7b97021d027
Author Gary Linscott
Date 2014-11-12 21:16:33 UTC
100% accurate PV display This gives SF accurate PVs, such that the evaluation of the leaf node in the PV matches the score backed up to the root (99% of the time. q-search will use the value stored in the hash table instead of the eval value sometimes). One drawback is that fail-high/low only get a minimal 2 move PV. It doesn't add any additional overhead to the non-PV codepath except an extra eight bytes to the SearchStack structure in multi-threaded searches. A core part of this is not pruning based on TT score in PV nodes. This was measured as not being a regression at multiple TCs, except for one exception, fast TC with huge hash, which is not realistic for longer searches. STC - 1 thread, 128 mb hash ELO: 1.42 +-3.1 (95%) LOS: 81.9% Total: 20000 W: 4078 L: 3996 D: 11926 STC - 3 thread, 128 mb hash ELO: -3.60 +-2.9 (95%) LOS: 0.8% Total: 20000 W: 3575 L: 3782 D: 12643 STC - 3 thread, 8 mb hash ELO: 0.12 +-2.9 (95%) LOS: 53.3% Total: 20000 W: 3654 L: 3647 D: 12699 LTC - 3 thread, 32mb hash ELO: 2.29 +-2.0 (95%) LOS: 98.8% Total: 35740 W: 5618 L: 5382 D: 24740 Bench: 6984058 Resolves #102
Copyright 2011–2024 Next Chess Move LLC