Dev Builds » 20260726-1043

Use this dev build

NCM plays each Stockfish dev build 20,000 times against Stockfish 15. This yields an approximate Elo difference and establishes confidence in the strength of the dev builds.

Summary

Host Duration Avg Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo
ncm-dbt-01 06:42:12 582284 4000 1820 314 1866 +137.58 ± 4.36 0 18 496 1448 38 +325.67 ± 15.27
ncm-dbt-02 06:43:37 590311 4004 1797 296 1911 +136.92 ± 4.41 1 21 493 1450 37 +323.99 ± 15.33
ncm-dbt-03 06:42:03 576777 4000 1822 328 1850 +136.36 ± 4.4 0 21 501 1441 37 +321.56 ± 15.2
ncm-dbt-04 06:43:35 567638 4000 1788 288 1924 +136.97 ± 4.32 0 17 499 1451 33 +325.29 ± 15.22
ncm-dbt-05 06:43:42 602841 3996 1788 272 1936 +138.74 ± 4.44 0 25 473 1459 41 +328.88 ± 15.67
20000 9015 1498 9487 +137.31 ± 1.96 1 102 2462 7249 186 +325.07 ± 6.85

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
480856 ncm-dbt-02 590739 4 1 0 3 +88.51 ± 137.63 0 0 1 1 0 +190.5 ± 634.21
480855 ncm-dbt-05 597117 496 232 32 232 +148.52 ± 12.99 0 3 51 185 9 +354.64 ± 48.54
480854 ncm-dbt-04 567918 500 218 31 251 +136.56 ± 11.97 0 1 65 180 4 +324.17 ± 42.47
480853 ncm-dbt-03 577233 500 236 39 225 +144.71 ± 12.68 0 2 57 183 8 +342.85 ± 45.71
480852 ncm-dbt-01 584243 500 230 46 224 +134.15 ± 12.06 0 1 68 177 4 +315.35 ± 41.44
480851 ncm-dbt-02 590310 500 227 38 235 +138.18 ± 13.12 0 3 63 176 8 +318.25 ± 43.39
480850 ncm-dbt-03 577889 500 227 43 230 +134.15 ± 12.27 0 2 66 178 4 +315.35 ± 42.24
480849 ncm-dbt-05 598524 500 217 35 248 +132.54 ± 12.91 0 4 65 176 5 +306.84 ± 42.73
480848 ncm-dbt-04 566651 500 230 36 234 +142.25 ± 12.58 0 1 62 179 8 +333.32 ± 43.57
480847 ncm-dbt-01 583908 500 235 42 223 +141.44 ± 12.41 0 1 62 180 7 +333.32 ± 43.57
480846 ncm-dbt-02 586223 500 232 43 225 +138.18 ± 13.5 1 1 65 174 9 +318.25 ± 42.59
480845 ncm-dbt-03 578465 500 221 33 246 +137.37 ± 12.56 0 3 61 181 5 +324.17 ± 44.15
480844 ncm-dbt-05 604083 500 228 30 242 +145.54 ± 11.79 0 3 49 195 3 +363.2 ± 49.57
480843 ncm-dbt-04 567482 500 215 32 253 +133.34 ± 12.29 0 2 67 177 4 +312.48 ± 41.9
480842 ncm-dbt-01 581485 500 230 39 231 +139.81 ± 11.62 0 1 60 186 3 +339.63 ± 44.35
480841 ncm-dbt-02 591642 500 227 33 240 +142.25 ± 11.95 0 1 59 185 5 +342.85 ± 44.76
480840 ncm-dbt-03 575802 500 234 35 231 +146.36 ± 12.61 0 4 49 191 6 +356.21 ± 49.54
480839 ncm-dbt-04 569430 500 227 45 228 +132.54 ± 12.91 0 6 59 182 3 +312.48 ± 44.93
480838 ncm-dbt-05 606698 500 211 31 258 +130.94 ± 13.14 0 4 68 172 6 +298.62 ± 41.71
480837 ncm-dbt-01 581319 500 226 46 228 +130.94 ± 13.14 0 4 68 172 6 +298.62 ± 41.71
480836 ncm-dbt-02 593239 500 225 29 246 +143.89 ± 11.65 0 2 53 192 3 +356.21 ± 47.52
480835 ncm-dbt-03 574297 500 231 55 214 +127.76 ± 13.02 0 4 71 170 5 +290.66 ± 40.76
480834 ncm-dbt-04 568673 500 217 37 246 +130.94 ± 11.96 0 2 68 178 2 +309.64 ± 41.56
480833 ncm-dbt-05 603008 500 219 37 244 +132.54 ± 12.32 0 2 68 176 4 +309.64 ± 41.56
480832 ncm-dbt-01 581319 500 227 28 245 +146.36 ± 11.96 0 3 49 194 4 +363.2 ± 49.57
480831 ncm-dbt-02 590868 500 217 36 247 +131.74 ± 11.94 0 3 64 182 1 +315.35 ± 43.03
480830 ncm-dbt-03 571994 500 234 44 222 +138.99 ± 11.88 0 1 62 183 4 +333.32 ± 43.57
480829 ncm-dbt-05 601891 500 231 34 235 +144.71 ± 13.48 0 6 49 187 8 +342.85 ± 49.35
480828 ncm-dbt-04 564919 500 228 37 235 +139.81 ± 11.84 0 1 61 184 4 +336.46 ± 43.96
480827 ncm-dbt-01 581777 500 218 36 246 +132.54 ± 12.32 0 3 65 179 3 +312.48 ± 42.67
480826 ncm-dbt-02 588302 500 225 38 237 +136.56 ± 12.59 0 4 59 183 4 +324.17 ± 44.97
480825 ncm-dbt-03 579785 500 227 33 240 +142.25 ± 11.95 0 1 59 185 5 +342.85 ± 44.76
480824 ncm-dbt-05 604847 500 232 31 237 +148.02 ± 11.19 0 1 50 196 3 +374.11 ± 48.94
480823 ncm-dbt-04 567997 500 225 35 240 +138.99 ± 11.22 0 0 62 186 2 +339.63 ± 43.41
480822 ncm-dbt-02 592116 500 221 36 243 +134.95 ± 11.83 0 2 63 183 2 +324.17 ± 43.32
480821 ncm-dbt-01 582444 500 232 41 227 +139.81 ± 13.07 0 4 58 181 7 +327.18 ± 45.37
480820 ncm-dbt-03 578753 500 212 46 242 +119.89 ± 12.23 0 4 76 170 0 +277.93 ± 39.29
480819 ncm-dbt-05 606562 500 218 42 240 +127.76 ± 12.26 0 2 73 172 3 +295.94 ± 39.99
480818 ncm-dbt-04 568037 500 228 35 237 +141.44 ± 12.81 0 4 55 185 6 +336.46 ± 46.65
480817 ncm-dbt-01 581777 500 222 36 242 +135.76 ± 12.01 0 1 66 179 4 +321.19 ± 42.12
480816 ncm-dbt-02 589368 500 222 43 235 +130.14 ± 13.16 0 5 66 174 5 +298.62 ± 42.41

Commit

Commit ID 43d4d0ef8bf905b0516422048500ec643333d1ca
Author Olaf Bernstein
Date 2026-07-26 10:43:40 UTC
Optimize RVV affine transforms This PR optimize the RVV implementations of the affine transforms. For the dense transform it adds special cases, when the entire inner loop fits into a vector register, which gave a small speedup. The sparse transform was optimized to use ChunkSize=1, because RVV currently doesn't have a dpbusd like instruction and it makes sure we don't operate on zero inputs. SpacemiT X100 benchmarks of the changes: ```py # pyshbench orig XXX 30 dense: +0.0137x sparse +0.2898x both: +0.3204x ``` ```py # stockfish bench X100: orig: 170061 nodes/second both: 206581 nodes/second A100: orig: 112999 nodes/second both: 128794 nodes/second ``` closes https://github.com/official-stockfish/Stockfish/pull/6967 No functional change Co-authored-by: wCgiLKcH <111658515+43eayl5@users.noreply.github.com>
Copyright 2011–2026 Next Chess Move LLC