Dev Builds » 20140127-1909

You are viewing an old NCM Stockfish dev build test. You may find the most recent dev build tests using Stockfish 15 as the baseline here.

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 06:38:51 1992610 2660 157 1097 1406 -128.31 ± 8.88
ncm-et-4 06:37:42 1969690 2642 146 1128 1368 -135.63 ± 9.02
ncm-et-5 01:45:57 2021494 716 47 313 356 -135.56 ± 17.85
ncm-et-6 01:46:03 2026554 719 39 293 387 -128.26 ± 16.86
ncm-et-7 01:46:00 1997410 705 42 295 368 -130.49 ± 17.42
ncm-et-8 01:46:05 2016729 718 46 312 360 -135.14 ± 17.73
ncm-et-9 02:22:58 2010546 967 50 411 506 -136.29 ± 14.79
ncm-et-10 06:38:35 1981484 2656 173 1121 1362 -129.72 ± 9.09
ncm-et-11 01:46:04 2004917 703 42 296 365 -131.47 ± 17.5
ncm-et-12 01:45:56 2017191 727 38 330 359 -147.87 ± 17.71
ncm-et-13 06:38:22 1989650 2693 168 1174 1351 -136.39 ± 9.14
ncm-et-14 01:46:04 2018964 715 31 313 371 -144.88 ± 17.23
ncm-et-15 06:38:16 1987231 2684 190 1134 1360 -127.65 ± 9.12
ncm-et-16 01:46:00 1965761 695 44 283 368 -124.55 ± 17.41
20000 1213 8500 10287 -132.68 ± 3.3

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
96061 ncm-et-3 2019-09-14 05:46 00:37:06 1978517 254 18 86 150 -95.34 ± 26.81
96060 ncm-et-13 2019-09-14 05:46 00:37:14 1970988 249 10 96 143 -125.14 ± 27.2
96059 ncm-et-15 2019-09-14 05:46 00:37:20 1973449 250 16 101 133 -123.02 ± 29.0
96058 ncm-et-4 2019-09-14 05:46 00:37:19 1970806 237 17 108 112 -140.61 ± 32.11
96057 ncm-et-9 2019-09-14 05:46 00:37:20 1980661 250 13 103 134 -130.94 ± 28.7
96056 ncm-et-10 2019-09-14 05:46 00:37:12 1972589 250 16 95 139 -113.68 ± 28.15
95569 ncm-et-10 2019-09-10 04:19 00:28:22 1974555 187 9 83 95 -145.42 ± 34.36
95568 ncm-et-3 2019-09-10 04:18 00:29:57 1982267 200 11 86 103 -136.97 ± 33.02
95567 ncm-et-15 2019-09-10 04:17 00:31:00 1972031 205 21 93 91 -127.45 ± 35.86
95566 ncm-et-13 2019-09-10 04:17 00:30:52 1980911 208 20 84 104 -110.48 ± 33.33
95565 ncm-et-4 2019-09-10 04:16 00:32:02 1972040 210 12 79 119 -114.86 ± 30.22
95564 ncm-et-10 2019-09-10 03:01 01:16:59 1956482 500 37 206 257 -122.24 ± 21.0
95563 ncm-et-15 2019-09-10 03:01 01:14:41 1975138 500 41 209 250 -121.46 ± 21.39
95562 ncm-et-3 2019-09-10 03:01 01:16:23 1983356 500 30 192 278 -116.78 ± 19.84
95561 ncm-et-13 2019-09-10 03:00 01:15:26 1977986 500 35 233 232 -145.54 ± 22.29
95560 ncm-et-4 2019-09-10 03:00 01:15:20 1946887 500 34 208 258 -126.18 ± 20.91
95559 ncm-et-15 2019-09-10 01:46 01:14:07 1972353 500 33 215 252 -132.54 ± 21.2
95558 ncm-et-13 2019-09-10 01:46 01:13:46 1977943 500 28 216 256 -137.37 ± 20.92
95557 ncm-et-3 2019-09-10 01:45 01:14:13 1975915 500 35 201 264 -119.89 ± 20.62
95556 ncm-et-10 2019-09-10 01:44 01:15:39 1958488 500 39 203 258 -118.33 ± 20.97
95555 ncm-et-4 2019-09-10 01:44 01:14:46 1973270 500 28 216 256 -137.37 ± 20.92
95554 ncm-et-15 2019-09-10 00:30 01:15:05 1979885 500 37 207 256 -123.02 ± 21.05
95553 ncm-et-13 2019-09-10 00:29 01:15:07 1969722 500 35 215 250 -130.94 ± 21.33
95552 ncm-et-4 2019-09-10 00:29 01:13:19 1967400 500 27 196 277 -122.24 ± 19.83
95551 ncm-et-3 2019-09-10 00:29 01:15:05 1975763 500 25 225 250 -147.19 ± 21.18
95550 ncm-et-10 2019-09-10 00:29 01:14:15 1966924 500 30 206 264 -127.76 ± 20.54
27332 ncm-et-16 2018-05-07 17:15 00:28:47 2019064 195 11 80 104 -128.49 ± 32.71
27331 ncm-et-11 2018-05-07 17:15 00:29:31 2012400 203 13 83 107 -124.92 ± 32.4
27330 ncm-et-4 2018-05-07 17:13 00:30:03 1947378 195 9 89 97 -151.45 ± 34.06
27329 ncm-et-7 2018-05-07 17:13 00:30:41 2025245 205 11 83 111 -127.45 ± 31.53
27328 ncm-et-9 2018-05-07 17:13 00:30:37 2025735 217 8 91 118 -140.01 ± 30.25
27327 ncm-et-3 2018-05-07 17:12 00:31:45 2025572 206 12 76 118 -111.63 ± 30.3
27326 ncm-et-5 2018-05-07 17:12 00:32:04 2022956 216 13 94 109 -136.97 ± 32.24
27325 ncm-et-14 2018-05-07 17:12 00:32:19 2018882 215 14 89 112 -126.51 ± 31.73
27324 ncm-et-6 2018-05-07 17:11 00:32:36 2027210 219 11 88 120 -127.6 ± 30.19
27323 ncm-et-10 2018-05-07 17:11 00:32:45 2018720 219 11 86 122 -123.99 ± 29.85
27322 ncm-et-12 2018-05-07 17:11 00:33:10 2015174 227 12 94 121 -131.44 ± 30.26
27321 ncm-et-8 2018-05-07 17:11 00:33:22 2024917 218 18 91 109 -121.01 ± 32.48
27320 ncm-et-15 2018-05-07 17:11 00:33:26 2018070 229 14 105 110 -146.11 ± 32.28
27319 ncm-et-13 2018-05-07 17:09 00:34:59 2025899 236 12 104 120 -143.0 ± 30.57
27318 ncm-et-4 2018-05-07 15:58 01:14:53 2010052 500 19 232 249 -158.08 ± 21.1
27317 ncm-et-9 2018-05-07 15:57 01:15:01 2025244 500 29 217 254 -137.37 ± 21.04
27316 ncm-et-13 2018-05-07 15:57 01:10:58 2024102 500 28 226 246 -145.54 ± 21.44
27315 ncm-et-7 2018-05-07 15:57 01:15:19 1969576 500 31 212 257 -131.74 ± 20.91
27314 ncm-et-8 2018-05-07 15:57 01:12:43 2008542 500 28 221 251 -141.44 ± 21.18
27313 ncm-et-16 2018-05-07 15:57 01:17:13 1912458 500 33 203 264 -123.02 ± 20.59
27312 ncm-et-6 2018-05-07 15:57 01:13:27 2025899 500 28 205 267 -128.55 ± 20.35
27311 ncm-et-11 2018-05-07 15:57 01:16:33 1997435 500 29 213 258 -134.15 ± 20.83
27310 ncm-et-3 2018-05-07 15:57 01:14:22 2026882 500 26 231 243 -151.35 ± 21.57
27309 ncm-et-5 2018-05-07 15:57 01:13:53 2020033 500 34 219 247 -134.95 ± 21.47
27308 ncm-et-14 2018-05-07 15:57 01:13:45 2019046 500 17 224 259 -153.02 ± 20.52
27307 ncm-et-15 2018-05-07 15:57 01:12:37 2019696 500 28 204 268 -127.76 ± 20.3
27306 ncm-et-12 2018-05-07 15:57 01:12:46 2019209 500 26 236 238 -155.54 ± 21.84
27305 ncm-et-10 2018-05-07 15:57 01:13:23 2022631 500 31 242 227 -156.39 ± 22.52

Commit

Commit ID 93f95cc936690a1b2637670bb22d138295eb5214
Author Uri Blass
Date 2014-01-27 19:09:51 UTC
Reduce VALUE_KNOWN_WIN to 10000 With some positions like 8/8/8/2p2K2/1pp5/br1p1b2/2p2r2/qqkqq3 w - - The eval score is higher than VALUE_INFINITE because is the sum of VALUE_KNOWN_WIN plus a big material advantage. This leads to an assert. Here are the steps to reproduce: Compile SF with debug=yes then do ./stockfish position fen 8/8/8/2p2K2/1pp5/br1p1b2/2p2r2/qqkqq3 w - - go depth 1 This patch fixes the issue in this case, but do exsist other positions for which the patch is not enough and we will need to limit the eval score to be sure not overflow the limit. Note that is not possible to increase the value of VALUE_INFINITE because should remain within int16_t type to be stored in a TT entry. bench: 7356053
Copyright 2011–2024 Next Chess Move LLC