| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Linmiao Xu
Date: Sat Jul 15 09:16:09 2023 +0200 Timestamp: 1689405369 Reintroduce nnue eval pawn count multipliers again With separate multipliers for nnue eval and optimism scaling. This patch used 4 out of 7 params tuned with spsa at 30+0.3 using this tuning config: Value LazyThreshold1 = Value(3622); Value LazyThreshold2 = Value(1962); int psqThresh = 2048; int nnueNpmBase = 945; int nnuePcMult = 0; int optNpmBase = 150; int optPcMult = 0; TUNE(SetRange(3322, 3922), LazyThreshold1); TUNE(SetRange(1662, 2262), LazyThreshold2); TUNE(SetRange(1748, 2348), psqThresh); TUNE(SetRange(745, 1145), nnueNpmBase); TUNE(SetRange(-16, 16), nnuePcMult); TUNE(SetRange(0, 300), optNpmBase); TUNE(SetRange(-16, 16), optPcMult); Passed STC: https://tests.stockfishchess.org/tests/view/64a5a9b402cd07745c60ed07 LLR: 2.95 (-2.94,2.94) <0.00,2.00> Total: 173632 W: 44417 L: 43903 D: 85312 Elo +1.03 Ptnml(0-2): 547, 20025, 45068, 20719, 457 Passed LTC: https://tests.stockfishchess.org/tests/view/64a972a302cd07745c6136af LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 277644 W: 70955 L: 70147 D: 136542 Elo +1.01 Ptnml(0-2): 193, 29902, 77787, 30784, 156 closes https://github.com/official-stockfish/Stockfish/pull/4681 bench 1556301 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Joost VandeVondele
Date: Sat Jul 15 09:15:16 2023 +0200 Timestamp: 1689405316 Build and test more binaries in CI use a fixed compiler on Linux and Windows (right now gcc 11). build avxvnni on Windows (Linux needs updated core utils) build x86-32 on Linux (Windows needs other mingw) fix a Makefile issue where a failed PGOBENCH would not stop the build reuse the WINE_PATH for SDE as we do for QEMU use WINE_PATH variable also for the signature verify the bench for each of the binaries do not build x86-64-avx2 on macos closes https://github.com/official-stockfish/Stockfish/pull/4682 No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: FauziAkram
Date: Sat Jul 15 09:13:02 2023 +0200 Timestamp: 1689405182 Use more expressive names for bonus1 and bonus2 closes https://github.com/official-stockfish/Stockfish/pull/4683 No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Joost VandeVondele
Date: Thu Jul 13 08:21:17 2023 +0200 Timestamp: 1689229277 Generate binaries for more advanced architectures use intel's Software Development Emulator (SDE) in the actions that build the binaries. This allows for building on Windows and Linux binaries for - x86-64-avx512 - x86-64-vnni256 - x86-64-vnni512 (x86-64-avxvnni needs more recent gcc in the actions) also build x86-64-avx2 on macos. closes https://github.com/official-stockfish/Stockfish/pull/4679 No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: mstembera
Date: Thu Jul 13 08:20:33 2023 +0200 Timestamp: 1689229233 More simplifications and cleanup in affine_transform_sparse_input.h closes https://github.com/official-stockfish/Stockfish/pull/4677 No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Sebastian Buchwald
Date: Thu Jul 13 08:19:37 2023 +0200 Timestamp: 1689229177 Cleanup code after removal of classical evaluation This includes the following changes: - Remove declaration of removed global variable - Adapt string that mentions removed UCI option closes https://github.com/official-stockfish/Stockfish/pull/4675 No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Joost VandeVondele
Date: Tue Jul 11 22:56:49 2023 +0200 Timestamp: 1689109009 Remove classical evaluation since the introduction of NNUE (first released with Stockfish 12), we have maintained the classical evaluation as part of SF in frozen form. The idea that this code could lead to further inputs to the NN or search did not materialize. Now, after five releases, this PR removes the classical evaluation from SF. Even though this evaluation is probably the best of its class, it has become unimportant for the engine's strength, and there is little need to maintain this code (roughly 25% of SF) going forward, or to expend resources on trying to improve its integration in the NNUE eval. Indeed, it had still a very limited use in the current SF, namely for the evaluation of positions that are nearly decided based on material difference, where the speed of the classical evaluation outweights its inaccuracies. This impact on strength is small, roughly 2Elo, and probably decreasing in importance as the TC grows. Potentially, removal of this code could lead to the development of techniques to have faster, but less accurate NN evaluation, for certain positions. STC https://tests.stockfishchess.org/tests/view/64a320173ee09aa549c52157 Elo: -2.35 ± 1.1 (95%) LOS: 0.0% Total: 100000 W: 24916 L: 25592 D: 49492 Elo -2.35 Ptnml(0-2): 287, 12123, 25841, 11477, 272 nElo: -4.62 ± 2.2 (95%) PairsRatio: 0.95 LTC https://tests.stockfishchess.org/tests/view/64a320293ee09aa549c5215b Elo: -1.74 ± 1.0 (95%) LOS: 0.0% Total: 100000 W: 25010 L: 25512 D: 49478 Elo -1.74 Ptnml(0-2): 44, 11069, 28270, 10579, 38 nElo: -3.72 ± 2.2 (95%) PairsRatio: 0.96 VLTC SMP https://tests.stockfishchess.org/tests/view/64a3207c3ee09aa549c52168 Elo: -1.70 ± 0.9 (95%) LOS: 0.0% Total: 100000 W: 25673 L: 26162 D: 48165 Elo -1.70 Ptnml(0-2): 8, 9455, 31569, 8954, 14 nElo: -3.95 ± 2.2 (95%) PairsRatio: 0.95 closes https://github.com/official-stockfish/Stockfish/pull/4674 Bench: 1444646 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Muzhen Gaming
Date: Tue Jul 11 22:55:00 2023 +0200 Timestamp: 1689108900 Simplify PvNode reduction Simplification STC: https://tests.stockfishchess.org/tests/view/64a415803ee09aa549c539c3 LLR: 2.93 (-2.94,2.94) <-1.75,0.25> Total: 37856 W: 9719 L: 9504 D: 18633 Elo +1.97 Ptnml(0-2): 98, 4277, 9977, 4464, 112 Simplification LTC: https://tests.stockfishchess.org/tests/view/64a5ffe202cd07745c60f360 LLR: 2.96 (-2.94,2.94) <-1.75,0.25> Total: 55878 W: 14323 L: 14138 D: 27417 Elo +1.15 Ptnml(0-2): 21, 5993, 15732, 6166, 27 closes https://github.com/official-stockfish/Stockfish/pull/4673 Bench: 2604965 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Joost VandeVondele
Date: Tue Jul 11 22:53:15 2023 +0200 Timestamp: 1689108795 Fix CI output closes https://github.com/official-stockfish/Stockfish/pull/4669 No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Linmiao Xu
Date: Thu Jul 6 23:03:58 2023 +0200 Timestamp: 1688677438 Update default net to nn-c38c3d8d3920.nnue This was a later epoch from the same experiment that led to the previous master net. After training, it was prepared the same way: 1. greedy permuting L1 weights with https://github.com/official-stockfish/Stockfish/pull/4620 2. leb128 compression with https://github.com/glinscott/nnue-pytorch/pull/251 3. greedy 2- and 3- cycle permuting with https://github.com/official-stockfish/Stockfish/pull/4640 Local elo at 25k nodes per move (vs. L1-1536 nn-fdc1d0fe6455.nnue): nn-epoch739.nnue : 20.2 +/- 1.7 Passed STC: https://tests.stockfishchess.org/tests/view/64a050b33ee09aa549c4e4c8 LLR: 2.95 (-2.94,2.94) <0.00,2.00> Total: 195552 W: 49977 L: 49430 D: 96145 Elo +0.97 Ptnml(0-2): 556, 22775, 50607, 23242, 596 Passed LTC: https://tests.stockfishchess.org/tests/view/64a127bd3ee09aa549c4f60c LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 235452 W: 60327 L: 59609 D: 115516 Elo +1.06 Ptnml(0-2): 119, 25173, 66426, 25887, 121 closes https://github.com/official-stockfish/Stockfish/pull/4666 bench 2427629 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Joost VandeVondele
Date: Thu Jul 6 23:03:16 2023 +0200 Timestamp: 1688677396 Add armv8-dotprod to CI binaries also generate binaries for more recent Android hardware. closes https://github.com/official-stockfish/Stockfish/pull/4663 No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: mstembera
Date: Thu Jul 6 23:02:11 2023 +0200 Timestamp: 1688677331 Simplify away lookup_count. https://tests.stockfishchess.org/tests/view/64a3c1a93ee09aa549c53167 LLR: 2.94 (-2.94,2.94) <-1.75,0.25> Total: 32832 W: 8497 L: 8280 D: 16055 Elo +2.30 Ptnml(0-2): 80, 3544, 8967, 3729, 96 closes https://github.com/official-stockfish/Stockfish/pull/4662 No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Joost VandeVondele
Date: Thu Jul 6 23:01:27 2023 +0200 Timestamp: 1688677287 Revise extract bench from git log in CI order commits differently closes https://github.com/official-stockfish/Stockfish/pull/4668 No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: ppigazzini
Date: Mon Jul 3 20:17:14 2023 +0200 Timestamp: 1688408234 Remove leftover braces for if conditional in CI closes https://github.com/official-stockfish/Stockfish/pull/4660 No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: ppigazzini
Date: Mon Jul 3 19:07:06 2023 +0200 Timestamp: 1688404026 Extract bench according to wiki instructions - loop through the commits starting from the latest one - read the bench value from the last match, if any, of the template in the commit body text closes https://github.com/official-stockfish/Stockfish/pull/4627 No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: rn5f107s2
Date: Mon Jul 3 18:54:22 2023 +0200 Timestamp: 1688403262 simplified gives check castling tested verifying perft and bench is unchanged on a larger set of epds for both standard and FRC chess. Passed non-regression STC: https://tests.stockfishchess.org/tests/live_elo/648587be65ffe077ca123d78 LLR: 2.95 (-2.94,2.94) <-1.75,0.25> Total: 153632 W: 41015 L: 40928 D: 71689 Elo +0.20 Ptnml(0-2): 377, 16077, 43816, 16174, 372 closes https://github.com/official-stockfish/Stockfish/pull/4628 No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: FauziAkram
Date: Mon Jul 3 18:38:41 2023 +0200 Timestamp: 1688402321 Improving grammar and readability of comments closes https://github.com/official-stockfish/Stockfish/pull/4643 No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Muzhen Gaming
Date: Mon Jul 3 18:33:27 2023 +0200 Timestamp: 1688402007 Simplify score improvement reduction Reduce depth by 2 based on score improvement, only for depths 3 to 11. Simplification STC: https://tests.stockfishchess.org/tests/view/64929a53dc7002ce609c7807 LLR: 2.95 (-2.94,2.94) <-1.75,0.25> Total: 238912 W: 63466 L: 63468 D: 111978 Elo -0.00 Ptnml(0-2): 564, 26262, 65805, 26262, 563 Simplification LTC: https://tests.stockfishchess.org/tests/view/64942e47dc7002ce609c9e07 LLR: 2.94 (-2.94,2.94) <-1.75,0.25> Total: 64452 W: 17485 L: 17320 D: 29647 Elo +0.89 Ptnml(0-2): 19, 6161, 19706, 6316, 24 closes https://github.com/official-stockfish/Stockfish/pull/4637 Bench: 2740142 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Muzhen Gaming
Date: Mon Jul 3 18:27:33 2023 +0200 Timestamp: 1688401653 Simplify away improvement term in null move search passed STC: https://tests.stockfishchess.org/tests/view/649c0d2edc7002ce609d33b5 LLR: 2.95 (-2.94,2.94) <-1.75,0.25> Total: 271104 W: 72181 L: 72217 D: 126706 Elo -0.05 Ptnml(0-2): 691, 30042, 74129, 29992, 698 passed LTC: https://tests.stockfishchess.org/tests/view/649d0dd7dc7002ce609d4efa LLR: 2.94 (-2.94,2.94) <-1.75,0.25> Total: 183120 W: 49469 L: 49418 D: 84233 Elo +0.10 Ptnml(0-2): 84, 17636, 56063, 17699, 78 closes https://github.com/official-stockfish/Stockfish/pull/4650 Bench: 2642851 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: peregrineshahin
Date: Mon Jul 3 18:24:41 2023 +0200 Timestamp: 1688401481 Fix pruning to (in TB loss) in Null move pruning. Current logic can apply Null move pruning on a dead-lost position returning an unproven loss (i.e. in TB loss score or mated in losing score) on nonPv nodes. on a default bench, this can be observed by adding this debugging line: ``` if (nullValue >= beta) { // Do not return unproven mate or TB scores nullValue = std::min(nullValue, VALUE_TB_WIN_IN_MAX_PLY-1); dbg_hit_on(nullValue <= VALUE_TB_LOSS_IN_MAX_PLY); // Hit #0: Total 73983 Hits 1 Hit Rate (%) 0.00135166 if (thisThread->nmpMinPly || depth < 14) return nullValue; ``` This fixes this very rare issue (happens at ~0.00135166% of the time) by eliminating the need to try Null Move Pruning with dead-lost positions and leaving it to be determined by a normal searching flow. The previous try to fix was not as safe enough because it was capping the returned value to (out of TB range) thus reviving the dead-lost position based on an artificial clamp (i.e. the in TB score/mate score can be lost on that nonPv node): https://tests.stockfishchess.org/tests/view/649756d5dc7002ce609cd794 Final fix: Passed STC: https://tests.stockfishchess.org/tests/view/649a5446dc7002ce609d1049 LLR: 2.93 (-2.94,2.94) <-1.75,0.25> Total: 577280 W: 153613 L: 153965 D: 269702 Elo -0.21 Ptnml(0-2): 1320, 60594, 165190, 60190, 1346 Passed LTC: https://tests.stockfishchess.org/tests/view/649cd048dc7002ce609d4801 LLR: 2.94 (-2.94,2.94) <-1.75,0.25> Total: 246432 W: 66769 L: 66778 D: 112885 Elo -0.01 Ptnml(0-2): 83, 22105, 78847, 22100, 81 closes https://github.com/official-stockfish/Stockfish/pull/4649 Bench: 2425978 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: mstembera
Date: Mon Jul 3 18:20:10 2023 +0200 Timestamp: 1688401210 Simplify lookup_count and clean up pieces(). https://github.com/official-stockfish/Stockfish/pull/4656 No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: disservin
Date: Mon Jul 3 18:17:20 2023 +0200 Timestamp: 1688401040 Add bmi2 to CI generated binaries verify bench for avx2 and bmi2 as well closes https://github.com/official-stockfish/Stockfish/pull/4658 No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: ppigazzini
Date: Sun Jul 2 10:32:36 2023 +0200 Timestamp: 1688286756 Make posix and msys2 shells consistent in CI In CI, it is typical for the process to halt immediately when an error is encountered. However, with our `shell: bash {0}` configuration, the process continues despite errors for posix shells. This commit updates the behavior of posix and msys2 shells to ensure consistency in terms of pipeline exit codes and stop conditions. We adopt the most appropriate default behavior as recommended by the GitHub documentation. Update the code that searches for the bench value in the git log: - to be compatible with the new shell settings - to retry the value from the first line that contains only the template and spaces/tabs/newlines see also https://docs.github.com/en/actions/using-workflows/workflow-syntax-for-github-actions#jobsjob_idstepsshell https://docs.github.com/en/actions/using-workflows/workflow-syntax-for-github-actions#exit-codes-and-error-action-preference https://github.com/msys2/setup-msys2/blob/main/main.js closes https://github.com/official-stockfish/Stockfish/pull/4653 No functional change see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: Linmiao Xu
Date: Sat Jul 1 13:34:30 2023 +0200 Timestamp: 1688211270 Update NNUE architecture to SFNNv7 with larger L1 size of 2048 Creating this net involved: - a 5-step training process from scratch - greedy permuting L1 weights with https://github.com/official-stockfish/Stockfish/pull/4620 - leb128 compression with https://github.com/glinscott/nnue-pytorch/pull/251 - greedy 2- and 3- cycle permuting with https://github.com/official-stockfish/Stockfish/pull/4640 The 5 training steps were: 1. 400 epochs, lambda 1.0, lr 9.75e-4 UHOx2-wIsRight-multinet-dfrc-n5000-largeGensfen-d9.binpack (178G) nodes5000pv2_UHO.binpack data_pv-2_diff-100_nodes-5000.binpack wrongIsRight_nodes5000pv2.binpack multinet_pv-2_diff-100_nodes-5000.binpack dfrc_n5000.binpack large_gensfen_multipvdiff_100_d9.binpack ep399 chosen as start model for step2 2. 800 epochs, end-lambda 0.75, skip 16 LeelaFarseer-T78juntoaugT79marT80dec.binpack (141G) T60T70wIsRightFarseerT60T74T75T76.binpack test78-junjulaug2022-16tb7p.no-db.min.binpack test79-mar2022-16tb7p.no-db.min.binpack test80-dec2022-16tb7p.no-db.min.binpack ep559 chosen as start model for step3 3. 800 epochs, end-lambda 0.725, skip 20 leela96-dfrc99-v2-T80dectofeb-sk20-mar-v6-T77decT78janfebT79apr.binpack (223G) leela96-filt-v2.min.binpack dfrc99-16tb7p-eval-filt-v2.min.binpack test80-dec2022-16tb7p-filter-v6-sk20.min-mar2023.binpack test80-jan2023-16tb7p-filter-v6-sk20.min-mar2023.binpack test80-feb2023-16tb7p-filter-v6-sk20.min-mar2023.binpack test80-mar2023-2tb7p-filter-v6.min.binpack test77-dec2021-16tb7p.no-db.min.binpack test78-janfeb2022-16tb7p.no-db.min.binpack test79-apr2022-16tb7p.no-db.min.binpack ep499 chosen as start model for step4 4. 800 epochs, end-lambda 0.7, skip 24 0dd1cebea57 dataset https://github.com/official-stockfish/Stockfish/pull/4606 ep599 chosen as start model for step5 5. 800 epochs, end-lambda 0.7, skip 28 same dataset as step4 ep619 became nn-1b951f8b449d.nnue For the final step5 training: python3 easy_train.py \ --experiment-name L1-2048-S5-sameData-sk28-S4-0dd1cebea57-shuffled-S3-leela96-dfrc99-v2-T80dectofeb-sk20-mar-v6-T77decT78janfebT79apr-sk20-S2-LeelaFarseerT78T79T80-ep399-S1-UHOx2-wIsRight-multinet-dfrc-n5000-largeGensfen-d9 \ --training-dataset /data/leela96-dfrc99-T60novdec-v2-T80juntonovjanfebT79aprmayT78jantosepT77dec-v6dd-T80apr.binpack \ --early-fen-skipping 28 \ --nnue-pytorch-branch linrock/nnue-pytorch/misc-fixes-L1-2048 \ --engine-test-branch linrock/Stockfish/L1-2048 \ --start-from-engine-test-net False \ --start-from-model /data/experiments/experiment_L1-2048-S4-0dd1cebea57-shuffled-S3-leela96-dfrc99-v2-T80dectofeb-sk20-mar-v6-T77decT78janfebT79apr-sk20-S2-LeelaFarseerT78T79T80-ep399-S1-UHOx2-wIsRight-multinet-dfrc-n5000-largeGensfen-d9/training/run_0/nn-epoch599.nnue --max_epoch 800 \ --lr 4.375e-4 \ --gamma 0.995 \ --start-lambda 1.0 \ --end-lambda 0.7 \ --tui False \ --seed $RANDOM \ --gpus 0 SF training data components for the step1 dataset: https://drive.google.com/drive/folders/1yLCEmioC3Xx9KQr4T7uB6GnLm5icAYGU Leela training data for steps 2-5 can be found at: https://robotmoon.com/nnue-training-data/ Due to larger L1 size and slower inference, the speed penalty loses elo at STC. Measurements from 100 bench runs at depth 13 with x86-64-modern on Intel Core i5-1038NG7 2.00GHz: sf_base = 1240730 +/- 3443 (95%) sf_test = 1153341 +/- 2832 (95%) diff = -87388 +/- 1616 (95%) speedup = -7.04330% +/- 0.130% (95%) Local elo at 25k nodes per move (vs. L1-1536 nn-fdc1d0fe6455.nnue): nn-epoch619.nnue : 21.1 +/- 3.2 Failed STC: https://tests.stockfishchess.org/tests/view/6498ee93dc7002ce609cf979 LLR: -2.95 (-2.94,2.94) <0.00,2.00> Total: 11680 W: 3058 L: 3299 D: 5323 Elo -7.17 Ptnml(0-2): 44, 1422, 3149, 1181, 44 LTC: https://tests.stockfishchess.org/tests/view/649b32f5dc7002ce609d20cf Elo: 0.68 ± 1.5 (95%) LOS: 80.5% Total: 40000 W: 10887 L: 10809 D: 18304 Elo +0.68 Ptnml(0-2): 36, 3938, 11958, 4048, 20 nElo: 1.50 ± 3.4 (95%) PairsRatio: 1.02 Passed VLTC 180+1.8: https://tests.stockfishchess.org/tests/view/64992b43dc7002ce609cfd20 LLR: 3.06 (-2.94,2.94) <0.00,2.00> Total: 38086 W: 10612 L: 10338 D: 17136 Elo +2.50 Ptnml(0-2): 9, 3316, 12115, 3598, 5 Passed VLTC SMP 60+0.6 th 8: https://tests.stockfishchess.org/tests/view/649a21fedc7002ce609d0c7d LLR: 2.95 (-2.94,2.94) <0.50,2.50> Total: 38936 W: 11091 L: 10820 D: 17025 Elo +2.42 Ptnml(0-2): 1, 2948, 13305, 3207, 7 closes https://github.com/official-stockfish/Stockfish/pull/4646 Bench: 2505168 see source |
| Windows x64 for Haswell CPUs Windows x64 for modern computers + AVX2 Windows x64 for modern computers Windows x64 + SSSE3 Windows x64 Linux x64 for Haswell CPUs Linux x64 for modern computers + AVX2 Linux x64 for modern computers Linux x64 + SSSE3 Linux x64 | Author: cj5716
Date: Sat Jul 1 13:06:49 2023 +0200 Timestamp: 1688209609 Negative extension on cutNodes based on depth This patch was inspired by candirufish original attempt at negative extensions here that failed yellow: https://tests.stockfishchess.org/tests/view/6486529065ffe077ca124f32 I tested some variations of the idea and tuned a depth condition for a modified version of it here https://tests.stockfishchess.org/tests/view/648db80a91c58631ce31fe00 after noticing abnormal scaling (ie many passed STC but not LTC) After some small tweaks I got the final version here Passed STC: LLR: 2.98 (-2.94,2.94) <0.00,2.00> Total: 122208 W: 32776 L: 32350 D: 57082 Elo +1.21 Ptnml(0-2): 310, 13250, 33553, 13686, 305 https://tests.stockfishchess.org/tests/view/64997934dc7002ce609d01e3 Passed LTC: LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 145092 W: 39617 L: 39115 D: 66360 Elo +1.20 Ptnml(0-2): 54, 13691, 44552, 14197, 52 https://tests.stockfishchess.org/tests/view/649a1c5ddc7002ce609d0bff closes https://github.com/official-stockfish/Stockfish/pull/4644 Bench: 2637784 see source |