Dev Builds » 20210619-0957

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 06:48:28 580595 4000 441 1612 1947 -104.78 ± 5.05 49 1149 726 76 0 -220.38 ± 12.62
ncm-dbt-02 06:48:36 586642 4000 468 1628 1904 -103.73 ± 4.99 42 1151 732 75 0 -219.36 ± 12.57
ncm-dbt-03 06:48:45 585168 4000 473 1606 1921 -101.18 ± 5.05 48 1115 759 78 0 -211.13 ± 12.33
ncm-dbt-04 06:48:59 566724 4000 452 1614 1934 -103.92 ± 5.25 52 1161 684 103 0 -217.35 ± 13.03
ncm-dbt-05 06:48:48 582911 4000 451 1620 1929 -104.59 ± 5.1 59 1124 744 73 0 -217.35 ± 12.46
20000 2285 8080 9635 -103.64 ± 2.28 250 5700 3645 405 0 -217.1 ± 5.63

Test Detail

ID Host Base NPS Games WLD Standard Elo Ptnml(0-2) Gamepair Elo CLI PGN
445261 ncm-dbt-03 587155 500 57 204 239 -105.25 ± 14.96 9 140 90 11 0 -215.85 ± 36.11 ↓
445260 ncm-dbt-04 564370 500 60 204 236 -102.97 ± 15.09 7 144 85 14 0 -213.85 ± 37.18 ↓
445259 ncm-dbt-02 587155 500 60 212 228 -109.07 ± 14.21 7 146 89 8 0 -230.16 ± 36.26 ↓
445258 ncm-dbt-05 583656 500 54 210 236 -112.14 ± 14.38 9 145 89 7 0 -234.38 ± 36.22 ↓
445257 ncm-dbt-01 579868 500 68 189 243 -85.78 ± 14.51 4 128 103 15 0 -176.33 ± 33.67 ↓
445256 ncm-dbt-01 578177 500 60 203 237 -102.22 ± 14.63 8 137 95 10 0 -209.91 ± 35.06 ↓
445255 ncm-dbt-04 567759 500 56 194 250 -98.44 ± 14.29 4 142 92 12 0 -207.95 ± 35.71 ↓
445254 ncm-dbt-03 585590 500 58 198 244 -99.95 ± 14.0 3 145 91 11 0 -213.85 ± 35.9 ↓
445253 ncm-dbt-02 587792 500 58 201 241 -102.22 ± 13.86 3 147 90 10 0 -219.87 ± 36.1 ↓
445252 ncm-dbt-05 586097 500 55 212 233 -112.91 ± 13.89 7 149 88 6 0 -240.82 ± 36.39 ↓
445251 ncm-dbt-01 577192 500 37 199 264 -116.77 ± 13.54 6 155 84 5 0 -254.16 ± 37.27 ↓
445250 ncm-dbt-04 564880 500 48 201 251 -109.83 ± 14.84 6 154 77 13 0 -234.38 ± 39.07 ↓
445249 ncm-dbt-03 582277 500 63 202 235 -99.2 ± 14.75 8 134 97 11 0 -202.15 ± 34.7 ↓
445248 ncm-dbt-05 584243 500 59 192 249 -94.69 ± 14.39 9 123 110 8 0 -189.0 ± 32.17 ↓
445247 ncm-dbt-02 585590 500 56 212 232 -112.14 ± 14.06 4 158 78 10 0 -245.2 ± 38.89 ↓
445246 ncm-dbt-05 585084 500 56 208 236 -109.07 ± 14.99 9 145 85 11 0 -226.0 ± 37.21 ↓
445245 ncm-dbt-01 579702 500 61 206 233 -103.73 ± 14.95 8 141 89 12 0 -213.85 ± 36.34 ↓
445244 ncm-dbt-04 565863 500 61 203 236 -101.46 ± 15.22 8 140 88 14 0 -207.95 ± 36.54 ↓
445243 ncm-dbt-03 582987 500 52 209 239 -112.91 ± 14.22 11 139 96 4 0 -232.26 ± 34.5 ↓
445242 ncm-dbt-02 585590 500 58 203 239 -103.73 ± 13.71 4 145 93 8 0 -221.9 ± 35.4 ↓
445241 ncm-dbt-01 582318 500 65 198 237 -94.69 ± 14.69 6 134 97 13 0 -194.57 ± 34.74 ↓
445240 ncm-dbt-05 579083 500 54 211 235 -112.91 ± 14.38 8 149 85 8 0 -238.66 ± 37.17 ↓
445239 ncm-dbt-04 567799 500 63 206 231 -102.22 ± 14.48 6 142 91 11 0 -213.85 ± 35.9 ↓
445238 ncm-dbt-03 586689 500 57 208 235 -108.3 ± 14.83 8 146 85 11 0 -226.0 ± 37.21 ↓
445237 ncm-dbt-02 586054 500 56 195 249 -99.2 ± 14.3 5 140 94 11 0 -207.95 ± 35.29 ↓
445236 ncm-dbt-03 584495 500 59 196 245 -97.69 ± 14.43 4 142 91 13 0 -206.01 ± 35.92 ↓
445235 ncm-dbt-01 582360 500 48 214 238 -119.89 ± 13.51 6 159 80 5 0 -263.42 ± 38.27 ↓
445234 ncm-dbt-04 568752 500 55 191 254 -96.94 ± 14.72 3 146 85 16 0 -206.01 ± 37.14 ↓
445233 ncm-dbt-02 587834 500 53 210 237 -112.91 ± 14.69 9 148 84 9 0 -236.51 ± 37.42 ↓
445232 ncm-dbt-05 580365 500 67 192 241 -88.74 ± 14.28 3 133 100 14 0 -185.33 ± 34.19 ↓
445231 ncm-dbt-03 584748 500 56 183 261 -90.22 ± 12.57 0 133 111 6 0 -194.57 ± 31.84 ↓
445230 ncm-dbt-01 584159 500 48 198 254 -107.54 ± 14.21 6 147 88 9 0 -228.08 ± 36.51 ↓
445229 ncm-dbt-04 567799 500 52 204 244 -109.07 ± 15.14 9 146 83 12 0 -226.0 ± 37.66 ↓
445228 ncm-dbt-05 581485 500 52 211 237 -114.45 ± 15.31 12 145 83 10 0 -234.38 ± 37.67 ↓
445227 ncm-dbt-02 585717 500 71 200 229 -91.71 ± 13.88 4 131 105 10 0 -190.85 ± 33.16 ↓
445226 ncm-dbt-03 587409 500 71 206 223 -96.19 ± 14.26 5 136 98 11 0 -200.24 ± 34.5 ↓
445225 ncm-dbt-02 587409 500 56 195 249 -99.2 ± 14.15 6 136 99 9 0 -206.01 ± 34.24 ↓
445224 ncm-dbt-05 583279 500 54 184 262 -92.45 ± 13.43 2 135 104 9 0 -196.45 ± 33.3 ↓
445223 ncm-dbt-04 566572 500 57 211 232 -110.6 ± 14.99 9 147 83 11 0 -230.16 ± 37.67 ↓
445222 ncm-dbt-01 580987 500 54 205 241 -108.3 ± 13.73 5 148 90 7 0 -232.26 ± 35.99 ↓

Commit

Commit ID 2e745956c0478627c4822104fc48a079c2f15e77
Author Tomasz Sobczyk
Date 2021-06-19 09:57:01 UTC
Change trace with NNUE eval support This patch adds some more output to the `eval` command. It adds a board display with estimated piece values (method is remove-piece, evaluate, put-piece), and splits the NNUE evaluation with (psqt,layers) for each bucket for the NNUE net. Example: ``` ./stockfish position fen 3Qb1k1/1r2ppb1/pN1n2q1/Pp1Pp1Pr/4P2p/4BP2/4B1R1/1R5K b - - 11 40 eval Contributing terms for the classical eval: +------------+-------------+-------------+-------------+ | Term | White | Black | Total | | | MG EG | MG EG | MG EG | +------------+-------------+-------------+-------------+ | Material | ---- ---- | ---- ---- | -0.73 -1.55 | | Imbalance | ---- ---- | ---- ---- | -0.21 -0.17 | | Pawns | 0.35 -0.00 | 0.19 -0.26 | 0.16 0.25 | | Knights | 0.04 -0.08 | 0.12 -0.01 | -0.08 -0.07 | | Bishops | -0.34 -0.87 | -0.17 -0.61 | -0.17 -0.26 | | Rooks | 0.12 0.00 | 0.08 0.00 | 0.04 0.00 | | Queens | 0.00 0.00 | -0.27 -0.07 | 0.27 0.07 | | Mobility | 0.84 1.76 | 0.01 0.66 | 0.83 1.10 | |King safety | -0.99 -0.17 | -0.72 -0.10 | -0.27 -0.07 | | Threats | 0.27 0.27 | 0.73 0.86 | -0.46 -0.59 | | Passed | 0.00 0.00 | 0.79 0.82 | -0.79 -0.82 | | Space | 0.61 0.00 | 0.24 0.00 | 0.37 0.00 | | Winnable | ---- ---- | ---- ---- | 0.00 -0.03 | +------------+-------------+-------------+-------------+ | Total | ---- ---- | ---- ---- | -1.03 -2.14 | +------------+-------------+-------------+-------------+ NNUE derived piece values: +-------+-------+-------+-------+-------+-------+-------+-------+ | | | | Q | b | | k | | | | | | +12.4 | -1.62 | | | | +-------+-------+-------+-------+-------+-------+-------+-------+ | | r | | | p | p | b | | | | -3.89 | | | -0.84 | -1.19 | -3.32 | | +-------+-------+-------+-------+-------+-------+-------+-------+ | p | N | | n | | | q | | | -1.81 | +3.71 | | -4.82 | | | -5.04 | | +-------+-------+-------+-------+-------+-------+-------+-------+ | P | p | | P | p | | P | r | | +1.16 | -0.91 | | +0.55 | +0.12 | | +0.50 | -4.02 | +-------+-------+-------+-------+-------+-------+-------+-------+ | | | | | P | | | p | | | | | | +2.33 | | | +1.17 | +-------+-------+-------+-------+-------+-------+-------+-------+ | | | | | B | P | | | | | | | | +4.79 | +1.54 | | | +-------+-------+-------+-------+-------+-------+-------+-------+ | | | | | B | | R | | | | | | | +4.54 | | +6.03 | | +-------+-------+-------+-------+-------+-------+-------+-------+ | | R | | | | | | K | | | +4.81 | | | | | | | +-------+-------+-------+-------+-------+-------+-------+-------+ NNUE network contributions (Black to move) +------------+------------+------------+------------+ | Bucket | Material | Positional | Total | | | (PSQT) | (Layers) | | +------------+------------+------------+------------+ | 0 | + 0.32 | - 1.46 | - 1.13 | | 1 | + 0.25 | - 0.68 | - 0.43 | | 2 | + 0.46 | - 1.72 | - 1.25 | | 3 | + 0.55 | - 1.80 | - 1.25 | | 4 | + 0.48 | - 1.77 | - 1.29 | | 5 | + 0.40 | - 2.00 | - 1.60 | | 6 | + 0.57 | - 2.12 | - 1.54 | <-- this bucket is used | 7 | + 3.38 | - 2.00 | + 1.37 | +------------+------------+------------+------------+ Classical evaluation -1.00 (white side) NNUE evaluation +1.54 (white side) Final evaluation +2.38 (white side) [with scaled NNUE, hybrid, ...] ``` Also renames the export_net() function to save_eval() while there. closes https://github.com/official-stockfish/Stockfish/pull/3562 No functional change
Copyright 2011–2025 Next Chess Move LLC
Feedback & Support Terms of Service Privacy Policy