Stockfish Testing Queue

Finished - 38881 tests

17-01-08 Roc SafeSafeCheck diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 6677 W: 1131 L: 1228 D: 4318
sprt @ 10+0.1 th 1 Fix of the fix. (shame on me !)
17-01-07 Roc PawnCaptureFollowUp2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 55599 W: 9984 L: 9874 D: 35741
sprt @ 10+0.1 th 1 Take 2: only look at forcing attacks
17-01-07 Roc PawnCaptureFollowUp3 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 15349 W: 2755 L: 2815 D: 9779
sprt @ 10+0.1 th 1 Take 3: Just consider the "consolidation" factor
17-01-07 Voy pruningT diff
LLR: -3.05 (-2.94,2.94) [0.00,5.00]
Total: 49178 W: 8786 L: 8708 D: 31684
sprt @ 10+0.1 th 1 Take 2
17-01-08 Roc SafeSafeCheck diff
LLR: -0.18 (-2.94,2.94) [0.00,5.00]
Total: 11725 W: 2124 L: 2082 D: 7519
sprt @ 10+0.1 th 1 A check experiment.
17-01-08 sni queen_threats3 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 6464 W: 1116 L: 1214 D: 4134
sprt @ 10+0.1 th 1 Queen domination. Inspired by Alain's green patch "QueenThreats3".
17-01-07 moj minorcenter2 diff
LLR: -1.60 (-2.94,2.94) [0.00,5.00]
Total: 12289 W: 2165 L: 2181 D: 7943
sprt @ 10+0.1 th 1 middle center squares is better for minors
17-01-07 Roc PawnCaptureFollowUp diff
LLR: 2.96 (-2.94,2.94) [0.00,5.00]
Total: 9083 W: 1691 L: 1528 D: 5864
sprt @ 10+0.1 th 1 Bonus for pawn captures which consolidate or create new threat
17-01-07 fau QueenMob3 diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 81257 W: 14609 L: 14490 D: 52158
sprt @ 10+0.1 th 1 Since now Queen mobility is more accurate we can decrease Queen EG psqt +few extra change.
17-01-07 sni IID diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 23035 W: 4025 L: 4104 D: 14906
sprt @ 10+0.1 th 1 IID margin = 200
17-01-07 Mic tte_skipEarlyPruning_Co diff
LLR: -2.21 (-2.94,2.94) [0.00,5.00]
Total: 9092 W: 1632 L: 1687 D: 5773
sprt @ 10+0.1 th 1 2nd try with PawnValueEg
17-01-07 sni risk_asymmetry diff
LLR: -3.58 (-2.94,2.94) [0.00,5.00]
Total: 5516 W: 845 L: 972 D: 3699
sprt @ 10+0.1 th 1 Asymmetric risk management for SF and her opponent
17-01-07 moj PSQT2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 3297 W: 545 L: 656 D: 2096
sprt @ 10+0.1 th 1 re-score pawn psq table
17-01-07 moj minorcenter diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 13735 W: 2387 L: 2454 D: 8894
sprt @ 10+0.1 th 1
17-01-07 moj PSQT diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 39707 W: 7101 L: 7059 D: 25547
sprt @ 10+0.1 th 1 Re-score pawn psq table
17-01-07 sg lmr2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 5591 W: 928 L: 1029 D: 3634
sprt @ 10+0.1 th 1 Less reduction for counter move at cut nodes.
17-01-07 lbr tune diff
LLR: -2.94 (-2.94,2.94) [0.00,5.00]
Total: 36974 W: 4822 L: 4819 D: 27333
sprt @ 60+0.6 th 1 LTC: test tuned values. SPRT(0,5), because (0,4) is really too costly.
17-01-07 Mic tte_skipEarlyPruning_Co diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 12138 W: 2094 L: 2168 D: 7876
sprt @ 10+0.1 th 1 test tt value vs static eval
17-01-06 lbr tune diff
LLR: 2.95 (-2.94,2.94) [0.00,4.00]
Total: 224957 W: 40981 L: 40060 D: 143916
sprt @ 10+0.1 th 1 test tuned values: rescheduling, as it was stopped by biffhero (i don't know if this is the known fishtest bug causing wrong bench from time to time...)
17-01-07 Mys pp diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 18086 W: 3265 L: 3313 D: 11508
sprt @ 10+0.1 th 1 kingdanger tweak
17-01-07 Mys pp diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 10978 W: 1904 L: 1983 D: 7091
sprt @ 10+0.1 th 1 and take 2 flip sign for comparison
17-01-07 Mys spc diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 6424 W: 1144 L: 1242 D: 4038
sprt @ 10+0.1 th 1 take 2
17-01-07 sg lmr2 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 17368 W: 3062 L: 3114 D: 11192
sprt @ 10+0.1 th 1 Less reduction for killer moves at cut nodes.
17-01-07 Mys spc diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 15272 W: 2787 L: 2847 D: 9638
sprt @ 10+0.1 th 1 tweak space scoring
17-01-06 moj minorSP2 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 19740 W: 3509 L: 3551 D: 12680
sprt @ 10+0.1 th 1 Control enemy pawn stop point by a minor
17-01-06 fau QueenMob2 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 70849 W: 12753 L: 12579 D: 45517
sprt @ 10+0.1 th 1 Queen Mob Take 2 (More effective on EG) Last try if red.
17-01-06 SC GLMpickBest diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 17053 W: 3000 L: 3053 D: 11000
sprt @ 5+0.05 th 3 Use threads with higher depths if score is not much worse.
17-01-06 sni kingtropism3 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 41893 W: 7358 L: 7308 D: 27227
sprt @ 10+0.1 th 1 king tropism: protected pieces
17-01-06 Mys ps diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 30731 W: 5462 L: 5458 D: 19811
sprt @ 10+0.1 th 1 no bonus for pieces in passed pawn promotion square (except queen) as can invite drawish tendancies i.e. when enemy rook attacking from behind
17-01-06 sg passed_pawns diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 16700 W: 2981 L: 3035 D: 10684
sprt @ 10+0.1 th 1 If opposite bishops exist give a penalty if promotion path is defended by enemy bishop
17-01-06 Voy fdsBonus diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 21503 W: 3886 L: 3920 D: 13697
sprt @ 10+0.1 th 1 Extra stat bonus at full depth search.
17-01-06 Mys space diff
LLR: -2.95 (-2.94,2.94) [-3.00,1.00]
Total: 18129 W: 3127 L: 3313 D: 11689
sprt @ 10+0.1 th 1 i think we can always eval space,may even be better
17-01-06 moj minorSP diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 13770 W: 2423 L: 2490 D: 8857
sprt @ 10+0.1 th 1 Add a score for minor stop point
17-01-06 cib mcp diff
LLR: -3.56 (-2.94,2.94) [0.00,5.00]
Total: 6421 W: 1124 L: 1248 D: 4049
sprt @ 10+0.1 th 1 MoveCountPruning for cutNodes
17-01-06 Voy pruningT diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 11449 W: 2001 L: 2078 D: 7370
sprt @ 10+0.1 th 1 stc
17-01-06 jos MCP1 diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 8186 W: 1468 L: 1559 D: 5159
sprt @ 10+0.1 th 1 Try the same with movecount based pruning.
17-01-06 fau QueenMob diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 7422 W: 1267 L: 1361 D: 4794
sprt @ 10+0.1 th 1 My Take on Queen Mob (Less effective on EG)
17-01-06 Mys RE3 diff
LLR: -0.42 (-2.94,2.94) [0.00,5.00]
Total: 78 W: 9 L: 27 D: 42
sprt @ 10+0.1 th 1 take 3 - simpler less destructive go
17-01-06 jos FUT1 diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 11234 W: 1989 L: 2067 D: 7178
sprt @ 10+0.1 th 1 Slowly increase amount of futility pruning. Is this too oldschool?
17-01-06 Mys RE2 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 616 W: 82 L: 211 D: 323
sprt @ 10+0.1 th 1 OK try only adding the pawns to potential mobility, keeping attacks
17-01-06 SC fracDepth02 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 1588 W: 242 L: 361 D: 985
sprt @ 10+0.1 th 1 Try something completely different with fractional depths: search twice for every depth with different pruning margins.
17-01-06 Mys RE diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 860 W: 131 L: 295 D: 434
sprt @ 10+0.1 th 1 Treat rook mobility differently, important in endgames with pawns and rook. test with 0,4 as not actually adding code and slightly removing some.
17-01-06 SC fracDepth02 diff
LLR: -2.95 (-2.94,2.94) [-3.00,1.00]
Total: 846 W: 83 L: 239 D: 524
sprt @ 10+0.1 th 1 Can we replace IID by fractional search? (99% no, but let me see how wrong this is).
17-01-06 SC fracDepth02 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 731 W: 75 L: 196 D: 460
sprt @ 10+0.1 th 1 Invert order
17-01-06 Mys mo diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 20898 W: 3689 L: 3726 D: 13483
sprt @ 10+0.1 th 1 An initial attempt at opening capture array and putting preference on central captures.
17-01-06 lbr tune diff
LLR: -0.09 (-2.94,2.94) [0.00,4.00]
Total: 2065 W: 365 L: 363 D: 1337
sprt @ 10+0.1 th 1 test tuned values
17-01-06 Mys mo diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 6718 W: 1126 L: 1223 D: 4369
sprt @ 10+0.1 th 1 take 2 opposite
17-01-06 sni attackedBy2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 7737 W: 1340 L: 1432 D: 4965
sprt @ 10+0.1 th 1 Include some protected pieces in attackedBy2
17-01-05 lbr tune diff
42643/40000 iterations
87272/90000 games played
90000 @ 10+0.1 th 1 tune checks
17-01-05 Mys HL diff
LLR: -0.90 (-2.94,2.94) [0.00,5.00]
Total: 2644 W: 389 L: 416 D: 1839
sprt @ 10+0.1 th 3 wonder if skip sizes should be missed when fail high/low as we need to increase delta to do more careful search