Stockfish Testing Queue

Finished - 37472 tests

16-12-15 Mys CR2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 10359 W: 1781 L: 1862 D: 6716
sprt @ 10+0.1 th 1 Some improvements - take 2 on safe queen checks
16-12-14 Gua MinorBehindPawnQueen diff
24479/25000 iterations
49460/50000 games played
50000 @ 20+0.2 th 1 MinorBehindPawn for Queen
16-12-15 Gua MinorQueen diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 8739 W: 1525 L: 1613 D: 5601
sprt @ 10+0.1 th 1 Minor behind pawn bonus for queen
16-12-14 vdv reductTune diff
27078/30000 iterations
54830/60000 games played
60000 @ 20+0.2 th 1 Continue tuning with recentered params.
16-12-14 Gua DefenceZone-reversed diff
14848/15000 iterations
30000/30000 games played
30000 @ 20+0.2 th 1 How it will work, if positions of our pieces in opponent king defence zone
16-12-14 Ser threefold_syzygy diff
ELO: -1.36 +-2.9 (95%) LOS: 17.8%
Total: 20000 W: 3528 L: 3606 D: 12866
20000 @ 10+0.1 th 1 Test syzygy's idea against my original idea. See pull request #925. (Note: root key can not be zeroed, its real value is used.)
16-12-14 sni semi_hanging2 diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 32245 W: 5716 L: 5706 D: 20823
sprt @ 10+0.1 th 1 General semi-hanging bonus, bonus = S(100,-50)
16-12-14 mbo matethreatpiece2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 18242 W: 3205 L: 3253 D: 11784
sprt @ 10+0.1 th 1 When we have a single piece defending against a mate threat that piece is sometimes out of play. Less penalty.
16-12-15 sni semi_hanging2 diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 4704 W: 742 L: 847 D: 3115
sprt @ 10+0.1 th 1 Try general weak pawns bonus
16-12-14 Voy pruningS diff
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 29310 W: 5225 L: 5118 D: 18967
sprt @ 10+0.1 th 1 Fixed Bench
16-12-14 sni quasi_passed_pawn3 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 18043 W: 3179 L: 3228 D: 11636
sprt @ 10+0.1 th 1 Bonus = S(100,-50), including hanging critical stoppers
16-12-14 SC criticalPosition diff
LLR: -2.95 (-2.94,2.94) [-3.00,1.00]
Total: 5234 W: 833 L: 997 D: 3404
sprt @ 10+0.1 th 1 Advanced pawn push is also critical.
16-12-14 Gua DefenceZone-reversed diff
240/15000 iterations
510/30000 games played
30000 @ 20+0.2 th 1 How it will work, if positions of our pieces in opponent king defence zone
16-12-14 vdv reductTune diff
25928/30000 iterations
53366/60000 games played
60000 @ 20+0.2 th 1 Tune three param version
16-12-14 mbo matethreatpiece diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 29418 W: 5210 L: 5211 D: 18997
sprt @ 10+0.1 th 1 When we have a single piece defending against a mate threat that piece is sometimes out of play.
16-12-14 hex search_delta_tuned diff
LLR: -3.30 (-2.94,2.94) [0.00,4.00]
Total: 24624 W: 4204 L: 4296 D: 16124
sprt @ 10+0.1 th 1 Aspiration window tuning, only non failed value changed
16-12-14 pb0 verification_search7 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 37536 W: 6671 L: 6638 D: 24227
sprt @ 10+0.1 th 1 Take 7
16-12-14 Gua PawnsPressureIndian3 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 4271 W: 723 L: 830 D: 2718
sprt @ 10+0.1 th 1 take 3
16-12-14 sg nullmove2 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 12136 W: 2113 L: 2187 D: 7836
sprt @ 10+0.1 th 1 less null move reduction at all nodes
16-12-14 Elb attackedBy2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 29136 W: 5115 L: 5118 D: 18903
sprt @ 10+0.1 th 1 Extend attackedBy2 with attacks from 2 pawns
16-12-14 hex search_delta_tuned diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 7651 W: 1272 L: 1402 D: 4977
sprt @ 10+0.1 th 1 Aspiration window parameters tuned in SPSA
16-12-14 Mys cr diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 9811 W: 1695 L: 1779 D: 6337
sprt @ 10+0.1 th 1 Noticed some games where SF is allowing queen checks at later stages in our half so try ranked queen safe checks.
16-12-14 Gua PawnsPressureIndian2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 371 W: 45 L: 183 D: 143
sprt @ 10+0.1 th 1 take 2
16-12-14 Mys uns diff
LLR: -1.36 (-2.94,2.94) [0.00,4.00]
Total: 20514 W: 2540 L: 2556 D: 15418
sprt @ 60+0.6 th 1 Try this at LTC since empty framework & is really an endgame patch anyway. unstoppable tweak take 3
16-12-14 Gua Lever2 diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 21985 W: 3823 L: 3905 D: 14257
sprt @ 10+0.1 th 1 Lever settings after 2 SPSA tests
16-12-14 Elb outpost diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 16420 W: 2851 L: 2907 D: 10662
sprt @ 10+0.1 th 1 Extra bonus if outpost supported by 2 pawns
16-12-14 Mys ks diff
LLR: -2.94 (-2.94,2.94) [0.00,5.00]
Total: 3472 W: 571 L: 681 D: 2220
sprt @ 10+0.1 th 1 Penalty when we can't castle during opening phase.
16-12-14 Mys pt diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 4452 W: 725 L: 831 D: 2896
sprt @ 10+0.1 th 1 If we do hit the pawn then focus more on center pawns
16-12-12 Gua ThreatByRank-16-0 diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 30650 W: 5432 L: 5485 D: 19733
sprt @ 10+0.1 th 1 ThreatByRank-16-0
16-12-13 Gua PawnsPressureIndian diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 8222 W: 1440 L: 1530 D: 5252
sprt @ 10+0.1 th 1 Trying to take into account the old indian pawns pressure
16-12-14 Roc PawnEnding1 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 18213 W: 3225 L: 3273 D: 11715
sprt @ 10+0.1 th 1 Bonus for having a pawn-supported passed pawn in pawn endings.
16-12-13 Gua PassivePawnsGame diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 14815 W: 2525 L: 2588 D: 9702
sprt @ 10+0.1 th 1 Attempt to use penalties for passive pawns structures
16-12-13 Gua tune-lever2 diff
29433/30000 iterations
60000/60000 games played
60000 @ 20+0.2 th 1 Trying to improve lever tunes (take 2)
16-12-12 Mys uns diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 237338 W: 41862 L: 41228 D: 154248
sprt @ 10+0.1 th 1 unstoppable tweak take 3
16-12-13 fau mobility2 diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 25228 W: 4418 L: 4489 D: 16321
sprt @ 10+0.1 th 1 Mobility Take 3 (This time big changes, if it finish red then its the last try)
16-12-12 hex search_delta diff
29667/30000 iterations
60000/60000 games played
60000 @ 20+0.2 th 1 Aspiration window tuning with better granularity
16-12-12 atu tt diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 9605 W: 1199 L: 1286 D: 7120
sprt @ 10+0.1 th 10 Thread-safe transposition table; ClusterSize = 2
16-12-13 fau mixtune diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 55992 W: 9863 L: 9832 D: 36297
sprt @ 10+0.1 th 1 Mix tuning.
16-12-13 Voy lmrt2 diff
LLR: -2.94 (-2.94,2.94) [0.00,4.00]
Total: 22205 W: 3878 L: 3959 D: 14368
sprt @ 10+0.1 th 1 stc
16-12-13 fau mobility diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 31916 W: 5638 L: 5687 D: 20591
sprt @ 10+0.1 th 1 mobility tuning.
16-12-13 vdv reduct-v3 diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 24290 W: 4246 L: 4321 D: 15723
sprt @ 10+0.1 th 1 Take 2, tuned params.
16-12-13 fau mobility1 diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 16912 W: 2991 L: 3090 D: 10831
sprt @ 10+0.1 th 1 mobility tune Take 2.
16-12-13 Voy lmrt diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 22402 W: 3839 L: 3871 D: 14692
sprt @ 10+0.1 th 1 stc
16-12-13 Voy pruning diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 11914 W: 2051 L: 2126 D: 7737
sprt @ 10+0.1 th 1 stc
16-12-13 Gua NoPawnsCenter diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 11420 W: 1988 L: 2065 D: 7367
sprt @ 10+0.1 th 1 Penalties, if there are no pawns on the central lines
16-12-13 Elb pawn_threat4 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 5859 W: 961 L: 1061 D: 3837
sprt @ 10+0.1 th 1 Bonus for pawn that can get close to enemy king
16-12-13 vdv reductTune diff
27050/30000 iterations
55231/60000 games played
60000 @ 20+0.2 th 1 Continue tuning after centering and reducing c.
16-12-12 Ser threefold_opt diff
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 36775 W: 4739 L: 4639 D: 27397
sprt @ 60+0.6 th 1 LTC. Optimized version of my previous solution. Testing as a bugfix: [-3.00,1.00].
16-12-13 Voy cmEvasion diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 24132 W: 4152 L: 4177 D: 15803
sprt @ 10+0.1 th 1 Separate cm table for evasions (fixed patch)
16-12-12 Roc PasserCreation2 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 49048 W: 8667 L: 8587 D: 31794
sprt @ 10+0.1 th 1 take 3: this time, consider instead only the pawn takes piece cases. (Fixed bench, sorry !)