Summarize the given text. Stockfish Testing Framework Profile Logout Tests…
- User
User: Summarize the given text.
Stockfish Testing Framework
Profile Logout Tests Overview LTC Greens Yellows Fishtest Contributors Top Month Events Users Blocked Workers Stockfish Official Releases Prereleases Contribute Progress NN Repo Resources Discord Fishtest Wiki Stockfish Wiki NN Trainer Wiki SPRT Calc Development Stockfish Fishtest NN Trainer BooksStockfish Testing Queue Cores 3313 Nodes / sec 1824M Games / min 2307 Time remaining 112.7h Workers - 233 machines Pending approval - 6 tests 24-05-21 fau ves diff
LLR: 0.00 (-2.94,2.94) <-1.75,0.25> Total: 0 W: 0 L: 0 D: 0 Ptnml(0-2): 0, 0, 0, 0, 0
sprt @ 10+0.1 th 1cores: 0 (0) ves 24-05-21 fau ves1 diff
LLR: 0.00 (-2.94,2.94) <-1.75,0.25> Total: 0 W: 0 L: 0 D: 0 Ptnml(0-2): 0, 0, 0, 0, 0
sprt @ 10+0.1 th 1cores: 0 (0) ves1 24-05-21 fau ves2 diff
LLR: 0.00 (-2.94,2.94) <-1.75,0.25> Total: 0 W: 0 L: 0 D: 0 Ptnml(0-2): 0, 0, 0, 0, 0
sprt @ 10+0.1 th 1cores: 0 (0) ves2 24-05-21 fau ves3 diff
LLR: 0.00 (-2.94,2.94) <-1.75,0.25> Total: 0 W: 0 L: 0 D: 0 Ptnml(0-2): 0, 0, 0, 0, 0
sprt @ 10+0.1 th 1cores: 0 (0) ves3 24-05-21 fau ves4 diff
LLR: 0.00 (-2.94,2.94) <-1.75,0.25> Total: 0 W: 0 L: 0 D: 0 Ptnml(0-2): 0, 0, 0, 0, 0
sprt @ 10+0.1 th 1cores: 0 (0) ves4 24-05-21 xot close3 diff
LLR: 0.00 (-2.94,2.94) <0.50,2.50> Total: 0 W: 0 L: 0 D: 0 Ptnml(0-2): 0, 0, 0, 0, 0
sprt @ 60+0.6 th 1cores: 0 (0) LTC: Move quicker if good alternative moves - rebased. [Note: Comment is wrong, but retained to correspond to STC test. Moves quicker if larger gaps between best moves.] Paused - 9 tests Failed - 0 tests Active - 41 tests 24-05-21 Dub ttUncutoff6a diff
LLR: 2.71 (-2.94,2.94) <0.00,2.00> Total: 30528 W: 7971 L: 7681 D: 14876 Ptnml(0-2): 109, 3524, 7715, 3800, 116
sprt @ 10+0.1 th 1cores: 43 (6) if(cut) d--; if(d<1)retqs (fixed, forgot
git addthe first time) 24-05-21 Dub ttUncutoff6c diffLLR: 2.56 (-2.94,2.94) <0.00,2.00> Total: 22496 W: 5875 L: 5609 D: 11012 Ptnml(0-2): 71, 2607, 5663, 2799, 108
sprt @ 10+0.1 th 1cores: 54 (6) if(cut) d -= (d>6) 24-05-21 lin eval-5p-re4 diff
LLR: 2.23 (-2.94,2.94) <0.00,2.00> Total: 42976 W: 11247 L: 10979 D: 20750 Ptnml(0-2): 156, 5085, 10822, 5185, 240
sprt @ 10+0.1 th 1cores: 40 (5) spsa 30k 24-05-21 cj5 pairwise-max-less diff
LLR: 2.09 (-2.94,2.94) <0.00,2.00> Total: 61856 W: 16127 L: 15854 D: 29875 Ptnml(0-2): 210, 6851, 16569, 7052, 246
sprt @ 10+0.1 th 1cores: 81 (6) Optimise pairwise multiplication, no functional change (new net was required to make these optimisations, the only difference is all the FT weights and biases being multiplied by 2). Retry this test as all the ARM workers are gone 24-05-21 lin eval-5p-re2 diff
LLR: 1.99 (-2.94,2.94) <0.50,2.50> Total: 39342 W: 10036 L: 9788 D: 19518 Ptnml(0-2): 31, 4302, 10763, 4538, 37
sprt @ 60+0.6 th 1cores: 158 (10) spsa 10k 24-05-21 lin eval-5p-re3 diff
LLR: 1.93 (-2.94,2.94) <0.00,2.00> Total: 64288 W: 16686 L: 16414 D: 31188 Ptnml(0-2): 245, 7607, 16203, 7809, 280
sprt @ 10+0.1 th 1cores: 44 (4) spsa 21k 24-05-20 Vir new358 diff
LLR: 1.78 (-2.94,2.94) <0.00,2.00> Total: 39012 W: 10089 L: 9907 D: 19016 Ptnml(0-2): 0, 3539, 12247, 3719, 1
sprt @ 70+0.7 th 7cores: 274 (9) Following https://tests.stockfishchess.org/tests/view/664b292e830eb9f886614a89 : Try decrease depth by 2 instead of 3 and only when depth >= 8 24-05-20 Vir new357 diff
LLR: 1.75 (-2.94,2.94) <-1.75,0.25> Total: 38720 W: 9971 L: 9876 D: 18873 Ptnml(0-2): 2, 3541, 12179, 3636, 2
sprt @ 70+0.7 th 7cores: 285 (10) LTC: Take 1 24-05-21 lin eval-5p-re diff
LLR: 1.22 (-2.94,2.94) <0.50,2.50> Total: 45252 W: 11439 L: 11244 D: 22569 Ptnml(0-2): 21, 4995, 12409, 5170, 31
sprt @ 60+0.6 th 1cores: 147 (7) LTC: 5p eval rebased 24-05-21 Vir new361 diff
LLR: 1.00 (-2.94,2.94) <0.00,2.00> Total: 154560 W: 39584 L: 39265 D: 75711 Ptnml(0-2): 454, 18505, 39169, 18572, 580
sprt @ 10+0.1 th 1cores: 40 (4) Take 2 24-05-21 Dub ttUncutoff6b diff
LLR: 0.60 (-2.94,2.94) <0.00,2.00> Total: 28800 W: 7444 L: 7346 D: 14010 Ptnml(0-2): 94, 3469, 7213, 3493, 131
sprt @ 10+0.1 th 1cores: 32 (3) if(cut) d -= (d>3) 24-05-20 xot ocbbad1 diff
LLR: 0.24 (-2.94,2.94) <0.00,2.00> Total: 84960 W: 21957 L: 21810 D: 41193 Ptnml(0-2): 284, 9980, 21790, 10157, 269
sprt @ 10+0.1 th 1cores: 52 (7) Reduce optimism if ocb, -40. 24-05-21 Dub ttUncutoff6d diff
LLR: 0.22 (-2.94,2.94) <0.00,2.00> Total: 14400 W: 3760 L: 3718 D: 6922 Ptnml(0-2): 53, 1741, 3586, 1751, 69
sprt @ 10+0.1 th 1cores: 31 (4) if(cut) d -= (d>9) 24-05-21 fau max2 diff
LLR: 0.19 (-2.94,2.94) <0.00,2.00> Total: 10752 W: 2760 L: 2727 D: 5265 Ptnml(0-2): 31, 1283, 2728, 1290, 44
sprt @ 10+0.1 th 1cores: 27 (4) max2 24-05-20 Vir new360 diff
LLR: 0.04 (-2.94,2.94) <0.00,2.00> Total: 55522 W: 14014 L: 13938 D: 27570 Ptnml(0-2): 8, 5722, 16222, 5804, 5
sprt @ 180+1.8 th 1cores: 164 (11) Try tweaking slightly 24-05-20 Tie tt_change_1 diff
LLR: -0.04 (-2.94,2.94) <0.00,2.00> Total: 484704 W: 125328 L: 124658 D: 234718 Ptnml(0-2): 1331, 50801, 137453, 51401, 1366
sprt @ 10+0.1 th 1cores: 107 (10) Merge branch 'official-stockfish:master' into tt_change_1 24-05-20 Min 2d325816-3 diff
LLR: -0.07 (-2.94,2.94) <0.00,2.00> Total: 316960 W: 82014 L: 81550 D: 153396 Ptnml(0-2): 1081, 37485, 80907, 37903, 1104
sprt @ 10+0.1 th 1cores: 58 (6) Allow PV reduction when eval is improving 24-05-21 Viz corrHistForC7 diff
LLR: -0.23 (-2.94,2.94) <0.50,2.50> Total: 58662 W: 14842 L: 14740 D: 29080 Ptnml(0-2): 36, 6564, 16019, 6686, 26
sprt @ 60+0.6 th 1cores: 132 (5) Take 7 24-05-21 fau max3 diff
LLR: -0.25 (-2.94,2.94) <0.00,2.00> Total: 10304 W: 2636 L: 2643 D: 5025 Ptnml(0-2): 42, 1248, 2582, 1235, 45
sprt @ 10+0.1 th 1cores: 27 (2) max3 24-05-21 fau max1 diff
LLR: -0.38 (-2.94,2.94) <0.00,2.00> Total: 12320 W: 3162 L: 3178 D: 5980 Ptnml(0-2): 50, 1499, 3085, 1469, 57
sprt @ 10+0.1 th 1cores: 25 (4) max1 24-05-21 Viz corrHistForC9 diff
LLR: -0.38 (-2.94,2.94) <0.00,2.00> Total: 19808 W: 5113 L: 5118 D: 9577 Ptnml(0-2): 81, 2394, 4960, 2387, 82
sprt @ 10+0.1 th 1cores: 37 (6) Take 9 24-05-21 fau max diff
LLR: -0.39 (-2.94,2.94) <0.00,2.00> Total: 11552 W: 2954 L: 2972 D: 5626 Ptnml(0-2): 45, 1400, 2906, 1378, 47
sprt @ 10+0.1 th 1cores: 21 (3) max 24-05-21 sha test221 diff
LLR: -0.60 (-2.94,2.94) <0.00,2.00> Total: 22048 W: 5746 L: 5768 D: 10534 Ptnml(0-2): 112, 2675, 5449, 2699, 89
sprt @ 10+0.1 th 1cores: 32 (3) +50% update when multicut 24-05-19 lin nCmplx3 diff
LLR: -0.65 (-2.94,2.94) <0.00,2.00> Total: 235456 W: 60720 L: 60429 D: 114307 Ptnml(0-2): 694, 28235, 59734, 28216, 849
sprt @ 10+0.1 th 1cores: 28 (3) weigh nc more in opt inc 24-05-21 Viz corrHistForCReb1 diff
LLR: -0.91 (-2.94,2.94) <0.50,2.50> Total: 14910 W: 3763 L: 3808 D: 7339 Ptnml(0-2): 11, 1718, 4045, 1667, 14
sprt @ 60+0.6 th 1cores: 129 (7) Rebased LTC 24-05-21 Viz corrHistForC10 diff
LLR: -0.93 (-2.94,2.94) <0.00,2.00> Total: 20384 W: 5261 L: 5315 D: 9808 Ptnml(0-2): 84, 2482, 5108, 2440, 78
sprt @ 10+0.1 th 1cores: 31 (4) Take 10 24-05-21 Min c86ec8ec-1 diff
LLR: -1.06 (-2.94,2.94) <-1.75,0.25> Total: 20032 W: 5142 L: 5262 D: 9628 Ptnml(0-2): 85, 2475, 5031, 2325, 100
sprt @ 10+0.1 th 1cores: 36 (5) Remove razoring 24-05-20 sg tuned_lmr_cond_rand5 diff
LLR: -1.09 (-2.94,2.94) <0.50,2.50> Total: 33078 W: 8349 L: 8369 D: 16360 Ptnml(0-2): 2, 3014, 10527, 2994, 2
sprt @ 70+0.7 th 7cores: 221 (11) LTC: Try first at VVLTC (with LTC bounds) because of vondeles fleet, later i try smaller TC: Test VVLTC tuned combinations of up to 3 conditions for more reduction (C=30, 28 conditions, random proportional selection). Use largest term P[11]. 24-05-21 rn5 precalcFmc diff
LLR: -1.30 (-2.94,2.94) <0.00,2.00> Total: 28928 W: 7404 L: 7477 D: 14047 Ptnml(0-2): 145, 3226, 7774, 3195, 124
sprt @ 10+0.1 th 1cores: 48 (4) Calculate fmc once and only update on depth changes 24-05-21 Min 2d325816-1 diff
LLR: -1.56 (-2.94,2.94) <0.50,2.50> Total: 85056 W: 21632 L: 21586 D: 41838 Ptnml(0-2): 52, 9492, 23394, 9538, 52
sprt @ 60+0.6 th 1cores: 172 (9) LTC default 24-05-21 rn5 weirdReductionsCN diff
LLR: -1.77 (-2.94,2.94) <0.00,2.00> Total: 29792 W: 7634 L: 7751 D: 14407 Ptnml(0-2): 140, 3632, 7462, 3529, 133
sprt @ 10+0.1 th 1cores: 43 (6) Only do "weird reductions" on cutnodes 24-05-21 sha test220 diff
LLR: -1.78 (-2.94,2.94) <0.00,2.00> Total: 24512 W: 6251 L: 6376 D: 11885 Ptnml(0-2): 108, 3021, 6117, 2908, 102
sprt @ 10+0.1 th 1cores: 32 (4) ttPv scaling? 24-05-21 Dub test205 diff
LLR: -1.93 (-2.94,2.94) <0.00,2.00> Total: 18000 W: 4482 L: 4615 D: 8903 Ptnml(0-2): 13, 1993, 5114, 1874, 6
sprt @ 60+0.6 th 1cores: 48 (6) speculative LTC of sx's ttmv reduc idea 24-05-21 sha test218 diff
LLR: -2.46 (-2.94,2.94) <-1.75,0.25> Total: 43872 W: 11127 L: 11399 D: 21346 Ptnml(0-2): 194, 5362, 11045, 5192, 143
sprt @ 10+0.1 th 1cores: 32 (2) simplify opponentworsening 24-05-20 xot tmcombine1 diff
112292/200000 iterations 224584/400000 games played
400000 @ 20+0.2 th 1cores: 69 (8) Retry tuning further from ceiling tune values. 24-05-21 fau wild73 diff
68135/200000 iterations 136270/400000 games played
400000 @ 10+0.1 th 1cores: 24 (4) wild73 24-05-21 fau wild73 diff
59058/200000 iterations 118116/400000 games played
400000 @ 20+0.2 th 1cores: 54 (4) wild73 24-05-20 sha nmp_formula_tune diff
52282/60000 iterations 104564/120000 games played
120000 @ 60+0.6 th 1cores: 69 (5) LTC: new tune without eval-beta limit 24-05-20 lin spsa-eval-5p diff
19972/60000 iterations 39944/120000 games played
120000 @ 60+0.6 th 1cores: 44 (3) rebased with latest eval patches 24-05-21 sg tune_lmr_cond_rand2 diff
14833/50000 iterations 29666/100000 games played
100000 @ 10+0.1 th 1cores: 48 (4) Tune combinations of up to 3 conditions for more reduction (C=50, 26 conditions, random proportional selection). 24-05-21 jmu vvltc_search_tune_21052 diff
6830/80000 iterations 13660/160000 games played
160000 @ 70+0.7 th 7cores: 274 (9) VVLTC search tune, rebased, continue from tuned 127k games. (new) Finished - 156875 tests
Prev 1 2 3 4 5 ... Next24-05-21 xot tmstableiter1 diff
LLR: -2.94 (-2.94,2.94) <0.50,2.50> Total: 47100 W: 11759 L: 11903 D: 23438 Ptnml(0-2): 15, 5185, 13307, 5015, 28
sprt @ 60+0.6 th 1 Gradual change for stable iterations.24-05-21 Dub ttUncutoff4c diff
LLR: -2.97 (-2.94,2.94) <0.00,2.00> Total: 35456 W: 8985 L: 9200 D: 17271 Ptnml(0-2): 170, 4256, 9050, 4123, 129
sprt @ 10+0.1 th 1 d += !cut && (d > 9)24-05-21 xot close3 diff
LLR: 2.93 (-2.94,2.94) <0.00,2.00> Total: 84416 W: 21964 L: 21581 D: 40871 Ptnml(0-2): 280, 9584, 22133, 9895, 316
sprt @ 10+0.1 th 1 Move quicker if good alternative moves.24-05-21 sha test219 diff
LLR: -2.95 (-2.94,2.94) <0.00,2.00> Total: 16448 W: 4185 L: 4432 D: 7831 Ptnml(0-2): 91, 2089, 4095, 1874, 75
sprt @ 10+0.1 th 1 rebased and fixed, test against minetas24-05-21 Dub ttUncutoff3b diff
LLR: -2.93 (-2.94,2.94) <0.00,2.00> Total: 21760 W: 5560 L: 5795 D: 10405 Ptnml(0-2): 99, 2749, 5399, 2554, 79
sprt @ 10+0.1 th 1 d += !cut && (d < 7)24-05-21 Dub ttUncutoff5c diff
LLR: -2.96 (-2.94,2.94) <0.00,2.00> Total: 7104 W: 1723 L: 1985 D: 3396 Ptnml(0-2): 38, 982, 1765, 738, 29
sprt @ 10+0.1 th 1 if(!cut) d += (ttd-d)/324-05-20 Viz corrHistForC2 diff
LLR: 2.93 (-2.94,2.94) <0.00,2.00> Total: 818848 W: 211452 L: 209975 D: 397421 Ptnml(0-2): 2693, 97183, 208262, 98526, 2760
sprt @ 10+0.1 th 1 Take 224-05-21 Tie limit_mtg diff
LLR: -1.72 (-2.94,2.94) <0.00,2.00> Total: 8192 W: 2043 L: 2184 D: 3965 Ptnml(0-2): 34, 1012, 2138, 885, 27
sprt @ 10+0.1 th 1 use tuned values24-05-21 Dub ttUncutoff5b diff
LLR: -2.93 (-2.94,2.94) <0.00,2.00> Total: 2944 W: 623 L: 893 D: 1428 Ptnml(0-2): 28, 466, 734, 236, 8
sprt @ 10+0.1 th 1 if(!cut) d += (ttd-d)/224-05-21 Dub ttUncutoff5a diff
LLR: -2.95 (-2.94,2.94) <0.00,2.00> Total: 1024 W: 53 L: 561 D: 410 Ptnml(0-2): 93, 330, 81, 8, 0
sprt @ 10+0.1 th 1 if(!cut) d += ttd-d (bench -> garbage)24-05-21 lin reSe4 diff
LLR: -1.87 (-2.94,2.94) <0.00,2.00> Total: 60384 W: 15598 L: 15677 D: 29109 Ptnml(0-2): 230, 7311, 15199, 7212, 240
sprt @ 10+0.1 th 1 re-eval only if over 2k24-05-21 Dub ttUncutoff4b diff
LLR: -2.94 (-2.94,2.94) <0.00,2.00> Total: 10656 W: 2623 L: 2870 D: 5163 Ptnml(0-2): 45, 1382, 2700, 1177, 24
sprt @ 10+0.1 th 1 d += !cut && (d > 6)24-05-21 Dub ttUncutoff3c diff
LLR: -2.93 (-2.94,2.94) <0.00,2.00> Total: 9888 W: 2434 L: 2681 D: 4773 Ptnml(0-2): 40, 1283, 2532, 1062, 27
sprt @ 10+0.1 th 1 d += !cut && (d < 10)24-05-21 Dub ttUncutoff4a diff
LLR: -2.95 (-2.94,2.94) <0.00,2.00> Total: 8736 W: 2097 L: 2349 D: 4290 Ptnml(0-2): 44, 1144, 2218, 944, 18
sprt @ 10+0.1 th 1 d += !cut && (d > 3)24-05-21 Vir new361 diff
LLR: -2.95 (-2.94,2.94) <0.00,2.00> Total: 65984 W: 16930 L: 17096 D: 31958 Ptnml(0-2): 206, 8006, 16756, 7796, 228
sprt @ 10+0.1 th 1 Take 424-05-21 lin reSe3 diff
LLR: -1.04 (-2.94,2.94) <0.00,2.00> Total: 42176 W: 10814 L: 10845 D: 20517 Ptnml(0-2): 188, 5009, 10671, 5086, 134
sprt @ 10+0.1 th 1 over 250024-05-21 lin reSe diff
LLR: -1.86 (-2.94,2.94) <0.00,2.00> Total: 59136 W: 15160 L: 15239 D: 28737 Ptnml(0-2): 197, 7159, 14955, 7040, 217
sprt @ 10+0.1 th 1 reeval if nnue se diff over 2k24-05-20 Viz corrHistForC3 diff
LLR: -2.94 (-2.94,2.94) <0.50,2.50> Total: 224418 W: 56896 L: 56678 D: 110844 Ptnml(0-2): 110, 24894, 61979, 25120, 106
sprt @ 60+0.6 th 1 Take 324-05-21 lin eval-5p-re2 diff
LLR: 2.99 (-2.94,2.94) <0.00,2.00> Total: 48864 W: 12742 L: 12403 D: 23719 Ptnml(0-2): 146, 5659, 12516, 5932, 179
sprt @ 10+0.1 th 1 spsa 10k Prev 1 2 3 4 5 ... NextAssistant: The text outlines various tests being conducted on the Stockfish testing framework, including specific test details, results, and future plans for testing. There is a list of different types of tests being carried out along with their corresponding outcomes. Additionally, the text highlights the number of cores being used, the speed of processing, and the progress made in the testing queue. User: tell me about fishtest Assistant: Fishtest is a part of the Stockfish testing framework where various contributors run tests on different versions of the Stockfish chess engine to evaluate performance improvements and changes. The tests involve playing a large number of chess games to assess the strength and effectiveness of different versions of Stockfish. The outcomes of these tests help in making decisions about which changes should be implemented in future official releases of Stockfish. FishTest involves contributors, top contributors of the month, events, users, and blocked workers. User: what are the tests right now? Assistant: The tests currently being conducted involve various versions of Stockfish chess engine being tested for their performance and effectiveness. These tests include evaluating specific changes such as evaluating alternative moves, evaluating different evaluation parameters, tuning reductions, and assessing move generation algorithms. The outcomes of these tests help in determining the best configurations for Stockfish. User: what are some of the test names? Assistant:
- Assistant
Some of the test names include "ttUncutoff6a," "eval-5p-re4," "pairwise-max-less," "eval-5p-re3," "new358," "new361," "tmcombine1," "ttUncutoff6d," "max2," "max1," "minetas," "nmp_formula_tune," "vvltc_search_tune_21052," "precalcFmc," and "weirdReductionsCN." These tests focus on different aspects of the Stockfish chess engine, such as evaluation tweaks, move generation optimizations, and reduction tuning.