Stockfish Testing Queue

Finished - 2285 tests

19-09-06 MJZ PassedP_blockSq diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 12148 W: 2635 L: 2739 D: 6774
sprt @ 10+0.1 th 1 Increase bonus if blocking Sq is protected by minor piece (k+7)
19-09-05 MJZ Check_pruningB diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 22098 W: 4851 L: 4906 D: 12341
sprt @ 10+0.1 th 1 Decrease reduction for safe checks at shallow depths (d < 8) and losing side (staticEval < alpha - 800).
19-09-05 MJZ Check_pruningB2 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 19892 W: 4341 L: 4407 D: 11144
sprt @ 10+0.1 th 1 Decrease reduction for safe checks at shallow depths (d < 8) and losing side (staticEval < alpha - 400).
19-09-04 MJZ Ply1-LMR diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 15178 W: 3285 L: 3375 D: 8518
sprt @ 10+0.1 th 1 Decrease reduction for first opponent moves @PV line.
19-09-03 MJZ Lazy-KSafety diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 9294 W: 1957 L: 2075 D: 5262
sprt @ 10+0.1 th 1 Adapt LazyThreshold to first evaluation of kingsafety. 1370 + KS / 2
19-09-03 MJZ Lazy-KSafety diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 4427 W: 929 L: 1072 D: 2426
sprt @ 10+0.1 th 1 Adapt LazyThreshold to first evaluation of kingsafety. 1430 + KS / 2
19-09-03 MJZ Lazy-KSafety diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 9166 W: 1974 L: 2093 D: 5099
sprt @ 10+0.1 th 1 Adapt LazyThreshold to first evaluation of kingsafety. 1400 + KS / 2 (mainly to evaluate cost of NPS loss)
19-09-03 MJZ Lazy-KSafety2 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 5746 W: 1166 L: 1301 D: 3279
sprt @ 10+0.1 th 1 Adapt LazyThreshold to first evaluation of kingsafety. 1400 +KS x 2 (new run after correction)
19-09-03 MJZ Lazy-KSafety diff
LLR: -1.83 (-2.94,2.94) [0.50,4.50]
Total: 1765 W: 343 L: 436 D: 986
sprt @ 10+0.1 th 1 Adapt LazyThreshold to first evaluation of kingsafety. +KS x 2
19-09-03 MJZ Check_pruningB diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 13848 W: 3020 L: 3116 D: 7712
sprt @ 10+0.1 th 1 Decrease reduction for safe checks at shallow depths (d < 8).
19-09-03 MJZ Check_pruningB diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 18575 W: 4088 L: 4161 D: 10326
sprt @ 10+0.1 th 1 Verify futility before pruning.
19-09-03 MJZ Futility_verif diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 9757 W: 2087 L: 2203 D: 5467
sprt @ 10+0.1 th 1 Verify futility before pruning + decrease margin
19-09-02 MJZ Check_pruningB diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 31079 W: 6863 L: 6874 D: 17342
sprt @ 10+0.1 th 1 Decrease reduction for safe checks in bad positions - staticEval < alpha - 800
19-09-02 MJZ CheckIndex-futility diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 13128 W: 2922 L: 3022 D: 7184
sprt @ 10+0.1 th 1 Increase futility margin in tactical positions. margin = (240 - CI) x d
19-09-02 MJZ Check_pruningB diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 16996 W: 3690 L: 3771 D: 9535
sprt @ 10+0.1 th 1 It seems that pruning some checks don't let SF to find some obvious mates. Decrease reduction for safe checks (1 time in 2)
19-09-02 MJZ CheckIndex-futility diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 11860 W: 2547 L: 2653 D: 6660
sprt @ 10+0.1 th 1 Increase futility margin in tactical positions. margin = (260 - CI) x d
19-09-02 MJZ Check_pruningB diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 9416 W: 2049 L: 2167 D: 5200
sprt @ 10+0.1 th 1 It seems that pruning some checks don't let SF to find some obvious mates. Try to find the source - decrease reduction for safe checks
19-09-02 MJZ Check_pruningB diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 2213 W: 428 L: 583 D: 1202
sprt @ 10+0.1 th 1 It seems that pruning some checks don't let SF to find some obvious mates. Try to find the source - no LMR for safe checks
19-09-02 MJZ Check_pruningB diff
LLR: -2.96 (-2.94,2.94) [-3.00,1.00]
Total: 11833 W: 2539 L: 2728 D: 6566
sprt @ 10+0.1 th 1 It seems that pruning some checks don't let SF to find some obvious mates. Try to find the source - remove "!extension" condition in shallow LMR pruning
19-09-01 MJZ Alpha-ExtensionF diff
LLR: -2.95 (-2.94,2.94) [0.00,3.50]
Total: 8746 W: 1334 L: 1485 D: 5927
sprt @ 60+0.6 th 1 Extend ttmove if ttValue > 260, PvNode and d > 12. (limited LTC test to check the idea, needs high depth search)
19-09-01 MJZ Alpha-ReductionB diff
ELO: -2.02 +-4.6 (95%) LOS: 19.3%
Total: 9791 W: 2142 L: 2199 D: 5450
10000 @ 10+0.1 th 1 Increase reduction at PvNode if we already had a good move. bestValue = alpha > min(260, beta - 10), d > 7
19-08-31 MJZ Initiative_threshold diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 11137 W: 2368 L: 2477 D: 6292
sprt @ 10+0.1 th 1 Add a threshold to initiative (10)
19-08-31 MJZ Check_Index-TC4 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 12304 W: 2702 L: 2806 D: 6796
sprt @ 10+0.1 th 1 Check_Index : more thinking time if several queen checks in search tree. More increase of reduction 2.1 >> 1.9
19-08-30 MJZ NMP-ply1 diff
ELO: -5.18 +-4.8 (95%) LOS: 1.6%
Total: 10000 W: 2367 L: 2516 D: 5117
10000 @ 5+0.05 th 1 NMP formula : 60 -22d +10p. Test at VSTC
19-08-30 MJZ NMP-ply4 diff
ELO: -0.87 +-4.8 (95%) LOS: 36.0%
Total: 10000 W: 2428 L: 2453 D: 5119
10000 @ 5+0.05 th 1 NMP formula : 80 -22d +10p. Test at VSTC
19-08-30 MJZ NMP-ply1 diff
ELO: -2.36 +-4.8 (95%) LOS: 16.5%
Total: 10000 W: 2410 L: 2478 D: 5112
10000 @ 5+0.05 th 1 NMP formula : 100 -22d +10p. Test at VSTC
19-08-30 MJZ NMP-ply1 diff
ELO: -9.56 +-4.8 (95%) LOS: 0.0%
Total: 10000 W: 2319 L: 2594 D: 5087
10000 @ 5+0.05 th 1 NMP formula : 0 + 5x ply. Test at VSTC for information.
19-08-30 MJZ NMP-ply1 diff
ELO: -11.16 +-4.3 (95%) LOS: 0.0%
Total: 10000 W: 1821 L: 2142 D: 6037
10000 @ 20+0.2 th 1 NPM formula : limited test @MTC to see how it scales.
19-08-30 MJZ NMP-ply1 diff
ELO: -12.23 +-4.8 (95%) LOS: 0.0%
Total: 10000 W: 2317 L: 2669 D: 5014
10000 @ 5+0.05 th 1 NPM formula : limited test @VSTC to see how it scales.
19-08-30 MJZ Check_Index-TC4 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 27253 W: 5938 L: 5968 D: 15347
sprt @ 10+0.1 th 1 Check_Index : more thinking time if several queen checks in search tree. One last try with an overall increase of reduction
19-08-30 MJZ TTsave-bestMove2 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 3837 W: 723 L: 867 D: 2247
sprt @ 5+0.05 th 4 Save the best Move if no "bestMove" - investigating the source of the problem - best Move must pass LMR at least
19-08-30 MJZ TTsave-bestMove2 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 2470 W: 451 L: 602 D: 1417
sprt @ 5+0.05 th 4 Save the best Move if no "bestMove" - investigating the source of the problem - only depth > 4
19-08-30 MJZ TTsave-bestMove2 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 1879 W: 296 L: 447 D: 1136
sprt @ 5+0.05 th 8 Save the best Move if no "bestMove" (fail low in general) - multi thread
19-08-30 MJZ TTsave-bestMove2 diff
LLR: -1.12 (-2.94,2.94) [0.50,4.50]
Total: 633 W: 101 L: 159 D: 373
sprt @ 5+0.05 th 4 Save the best Move if no "bestMove" - investigating the source of the problem - no ttMove penalty in case of fail low
19-08-22 MJZ psq_npm diff
LLR: -2.95 (-2.94,2.94) [0.00,3.50]
Total: 114976 W: 19064 L: 18920 D: 76992
sprt @ 60+0.6 th 1 Increase score in late EG. +1/2, NPM==0 and abs(v) > 120 - LTC
19-08-24 MJZ psq_npm4 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 72456 W: 16424 L: 16226 D: 39806
sprt @ 10+0.1 th 1 Increase score in late EG. +1/2, NPM==0 and abs(v) > 100, extend PV until abs(alpha-staticEval) > 400
19-08-29 MJZ Ab2Minors_2 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 25810 W: 5718 L: 5755 D: 14337
sprt @ 10+0.1 th 1 Bonus if we attack an apponent piece by 2 minors pieces (pawn, knight or bishop). (8, 4) if not strongly protected
19-08-29 MJZ Ab2Minors_3 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 9817 W: 2094 L: 2210 D: 5513
sprt @ 10+0.1 th 1 Bonus if we attack an apponent piece by 2 minors pieces (pawn, knight or bishop). (6, 3) if not strongly protected
19-08-29 MJZ Ab2Minors_3 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 11899 W: 2599 L: 2705 D: 6595
sprt @ 10+0.1 th 1 Bonus if we attack an apponent piece by 2 minors pieces (pawn, knight or bishop). (12, 6) if not strongly protected
19-08-29 MJZ Ab2Minors_3 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 7550 W: 1636 L: 1764 D: 4150
sprt @ 10+0.1 th 1 Bonus if we attack an apponent piece or pawn by 2 minors pieces (pawn, knight or bishop). (8, 4) if not strongly protected
19-08-29 MJZ Check_Index-TC4 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 17852 W: 3848 L: 3924 D: 10080
sprt @ 10+0.1 th 1 Check_Index : more thinking time if several queen checks in search tree. New test after tuning
19-08-28 MJZ Check_Index-TC_tune diff
19184/20000 iterations
40003/40000 games played
40000 @ 20+0.2 th 1 Check_Index : more thinking time if several queen checks in search tree. Tuning
19-08-23 MJZ psq_npm3 diff
LLR: -2.95 (-2.94,2.94) [0.00,3.50]
Total: 159110 W: 26702 L: 26432 D: 105976
sprt @ 60+0.6 th 1 Increase score in late EG. +1/2, NPM==0 and abs(v) > 100 - LTC
19-08-28 MJZ Check_Index-TC4 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 44720 W: 9724 L: 9669 D: 25327
sprt @ 10+0.1 th 1 Check_Index : more thinking time if several queen checks in search tree. Progressive formula 0.88 + 20x (check + 5)
19-08-28 MJZ QCheckCount-MCP diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 5229 W: 1081 L: 1220 D: 2928
sprt @ 10+0.1 th 1 More MCP for less tactical positions : MCP limit - 1/8 if QCI > 500
19-08-28 MJZ Check_Index-TC4 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 35923 W: 7966 L: 7952 D: 20005
sprt @ 10+0.1 th 1 Check_Index : more thinking time if several queen checks in search tree. Progressive formula 0.94 + 10x (check + 5)
19-08-28 MJZ QCheckCount-MCP diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 5842 W: 1253 L: 1389 D: 3200
sprt @ 10+0.1 th 1 More MCP for tactical positions : MCP limit - 1/8 if QCI < 30
19-08-28 MJZ QCheckCount-MCP diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 13182 W: 2823 L: 2922 D: 7437
sprt @ 10+0.1 th 1 Less MCP for tactical positions : MCP limit + 1/8 if QCI < 30
19-08-28 MJZ Check_Index-TC5 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 27856 W: 6125 L: 6152 D: 15579
sprt @ 10+0.1 th 1 Check_Index : more thinking time if several queen checks in search tree. Q_check_index < 20, +40%
19-08-28 MJZ Check_Index-TC4 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 24077 W: 5297 L: 5342 D: 13438
sprt @ 10+0.1 th 1 Check_Index : more thinking time if several queen checks in search tree. Q_check_index < 20, +20%