Stockfish Testing Queue

Finished - 30415 tests

16-06-15 pec mult_stop diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 9733 W: 1591 L: 1675 D: 6467
sprt @ 10+0.1 th 3 Take 2. Allow multiple threads to stop search. Do not care about easy move stop for now
16-06-15 sni keep_pieces diff
LLR: -3.20 (-2.94,2.94) [0.00,5.00]
Total: 2430 W: 568 L: 705 D: 1157
sprt @ 10+0.1 th 1 Take 2, midgameLove = 100. Submitted with priority = -2.
16-06-15 sni keep_pieces diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 10012 W: 2116 L: 2198 D: 5698
sprt @ 10+0.1 th 1 Try to avoid early exchanges, prefering midgame positions. Take 1, midgameLove = 40.
16-06-15 Voy checkExtC diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 9169 W: 1687 L: 1773 D: 5709
sprt @ 10+0.1 th 1 Version C: Extend check if capture or promotion with positive SEE.
16-06-15 Voy checkExtB diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 4926 W: 871 L: 976 D: 3079
sprt @ 10+0.1 th 1 Version B. Don't extend check if bad history.
16-06-15 Voy checkExt diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 14191 W: 2632 L: 2696 D: 8863
sprt @ 10+0.1 th 1 Don't extend checks at low iterations.
16-06-15 Elb init_blockp diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 8534 W: 1577 L: 1666 D: 5291
sprt @ 10+0.1 th 1 Take 2: Exclude blocked pawns from initiative calculation
16-06-15 pb0 ext_checks_one_evasion diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 7051 W: 1315 L: 1411 D: 4325
sprt @ 10+0.1 th 1 Extend checks with only one evasion move as this can't produce search explosions.
16-06-15 sni coordination5 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 7347 W: 1289 L: 1383 D: 4675
sprt @ 10+0.1 th 1 Trying the semi-hanging idea
16-06-15 sni coordination5 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 3390 W: 613 L: 725 D: 2052
sprt @ 10+0.1 th 1 Introduce coordination bonus, calculated from squares that we double-control.
16-06-15 mco doubled_and_passed3 diff
LLR: -0.18 (-2.94,2.94) [1.00,-3.00]
Total: 782 W: 143 L: 135 D: 504
sprt @ 10+0.1 th 1 Reschedule as a fix: Testing for mcostalba, as suggested in https://github.com/official-stockfish/Stockfish/pull/701
16-06-15 Elb init_blockp diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 6270 W: 1071 L: 1170 D: 4029
sprt @ 10+0.1 th 1 Exclude blocked pawns from initiative calculation
16-06-15 sni doubled_and_passed3 diff
LLR: -2.98 (-2.94,2.94) [0.00,4.00]
Total: 31040 W: 5828 L: 5878 D: 19334
sprt @ 10+0.1 th 1 Testing for mcostalba, as suggested in https://github.com/official-stockfish/Stockfish/pull/701
16-06-15 sg hanging diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 9822 W: 1803 L: 1886 D: 6133
sprt @ 10+0.1 th 1 Consider double attacked pieces with only one defender as semi-hanging. Use for them the half bonus as for hanging pieces. Inspired by Stephane Nicolet.
16-06-15 sni pawn_mobility7 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 9263 W: 1672 L: 1758 D: 5833
sprt @ 10+0.1 th 1 pawn mobility, bonus=S(0,7)
16-06-15 pb0 pin_aware_SEE4checkext diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 15249 W: 2772 L: 2832 D: 9645
sprt @ 10+0.1 th 1 Use a pin-aware SEE when deciding to extend a checkmove or not.
16-06-15 eel TuneForMichaelB diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 19714 W: 3636 L: 3725 D: 12353
sprt @ 10+0.1 th 1 Test for Michael as documented http://www.talkchess.com/forum/viewtopic.php?t=60462
16-06-14 Fis strongKnight diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 24076 W: 4527 L: 4548 D: 15001
sprt @ 10+0.1 th 1 Discourage knights for strong side.
16-06-14 Roc KingForward diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 31039 W: 5867 L: 5857 D: 19315
sprt @ 10+0.1 th 1 Bonus for king with forward mobility in the middlegame/endgame.
16-06-14 sni pawn_mobility6 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 14816 W: 1981 L: 2050 D: 10785
sprt @ 60+0.6 th 1 LTC: Try endgame bonus, S(0,5)
16-06-14 Voy HSpt diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 10436 W: 1915 L: 1996 D: 6525
sprt @ 10+0.1 th 1 History Split off Capture PieceType
16-06-14 IIv tmm_new diff
LLR: -2.96 (-2.94,2.94) [-3.00,1.00]
Total: 18868 W: 3378 L: 3568 D: 11922
sprt @ 10+0.1 th 1 A variant inspired by tuning.
16-06-14 lbr kingRing diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 3745 W: 585 L: 694 D: 2466
sprt @ 20+0.2 th 1 reshaped kingRing, with tuned weights
16-06-11 mco master diff
ELO: 36.29 +-1.6 (95%) LOS: 100.0%
Total: 40000 W: 6769 L: 2606 D: 30625
40000 @ 60+0.6 th 1 Regression test
16-06-12 lbr kingRingTune diff
37878/40000 iterations
76771/80000 games played
80000 @ 20+0.2 th 1 Tune KS weights with a more logical form for kingRing
16-06-14 cib pawn_capture diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 12617 W: 2331 L: 2402 D: 7884
sprt @ 10+0.1 th 1 Different counter move stats for pawn capture.
16-06-14 sg knights diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 9633 W: 1749 L: 1833 D: 6051
sprt @ 10+0.1 th 1 Bonus if opponent knight can't move because its the only defender of a pawn
16-06-12 IIv tmm_tune diff
134288/150000 iterations
300000/300000 games played
300000 @ 2+0.02 th 1 Continue tuning with somewhat higher ck values.
16-06-14 Fis strongQueen diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 13116 W: 2404 L: 2473 D: 8239
sprt @ 10+0.1 th 1 Simpler version Take 2.
16-06-14 sni pawn_mobility6 diff
LLR: 2.96 (-2.94,2.94) [0.00,5.00]
Total: 54265 W: 10279 L: 9916 D: 34070
sprt @ 10+0.1 th 1 Try endgame bonus: S(0,5)
16-06-14 Mys rep diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 5852 W: 1093 L: 1194 D: 3565
sprt @ 10+0.1 th 1 First cut at repetition contempt
16-06-14 sni pawn_mobility6 diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 7063 W: 1294 L: 1390 D: 4379
sprt @ 10+0.1 th 1 Try endgame bonus: S(0,15)
16-06-14 Fis strongQueen diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 8892 W: 1587 L: 1674 D: 5631
sprt @ 10+0.1 th 1 The attacking side should try to keep its queen.
16-06-13 Elb lmr_cut4B diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 31750 W: 5926 L: 5973 D: 19851
sprt @ 10+0.1 th 1 In IID we should not expect a PvNode to be a cut node. Resubmitted with [0,4] bounds.
16-06-13 Voy impHS diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 21128 W: 3933 L: 3967 D: 13228
sprt @ 10+0.1 th 1 Split history stats based on improving.
16-06-13 sni exposed_king2 diff
LLR: -2.94 (-2.94,2.94) [0.00,5.00]
Total: 22910 W: 4235 L: 4261 D: 14414
sprt @ 10+0.1 th 1 Take 2 (factor=10)
16-06-13 cib stats_update diff
LLR: -2.95 (-2.94,2.94) [-3.00,1.00]
Total: 35650 W: 4708 L: 4904 D: 26038
sprt @ 60+0.6 th 1 LTC: Do not multiply bonus by 32, when updating stats. (With correct bounds)
16-06-13 sni exposed_king2 diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 24228 W: 4583 L: 4603 D: 15042
sprt @ 10+0.1 th 1 Discovered checks by pieces (not pawns) in attackUnits
16-06-13 luc scaleEqual diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 4871 W: 839 L: 944 D: 3088
sprt @ 10+0.1 th 1 Scale endgames with same pieces and few pawn asymmetries
16-06-13 sni pawn_mobility6 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 4802 W: 886 L: 992 D: 2924
sprt @ 10+0.1 th 1 Pawn mobility
16-06-12 Fis doKingSafety diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 27340 W: 5039 L: 5101 D: 17200
sprt @ 10+0.1 th 1 Patch of @Mindbreaker1 https://github.com/official-stockfish/Stockfish/pull/489
16-06-13 sni imbalance2 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 14420 W: 2714 L: 2777 D: 8929
sprt @ 10+0.1 th 1 Take 3
16-06-13 sni knight_mobility diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 2315 W: 417 L: 535 D: 1363
sprt @ 10+0.1 th 1 Only count knight mobility for moves to ranks 4..8
16-06-12 cib stats_update diff
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 24459 W: 4632 L: 4517 D: 15310
sprt @ 10+0.1 th 1 Do not multiply bonus by 32, when updating stats.
16-06-12 sni imbalance2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 20225 W: 3800 L: 3837 D: 12588
sprt @ 10+0.1 th 1 Take 2, using XOR to compute imbalance.
16-06-12 SC escape diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 8120 W: 1513 L: 1604 D: 5003
sprt @ 10+0.1 th 1 Scale reductions for moves that escape a capture
16-06-12 sni imbalance2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 15407 W: 2954 L: 3012 D: 9441
sprt @ 10+0.1 th 1 Try a variant of mstembera and mindbreaker idea for piece imbalance in initiative()
16-06-12 Fis imbalance diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 48276 W: 9168 L: 9082 D: 30026
sprt @ 10+0.1 th 1 On behalf of @Mindbreaker1 w/ his tested values. Final try.
16-06-12 jos king_safety_three diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 26074 W: 4924 L: 4990 D: 16160
sprt @ 10+0.1 th 1 Re-tuned KingDanger array.
16-06-11 IIv tmm_tune diff
61464/200000 iterations
129016/400000 games played
400000 @ 2+0.02 th 1 I promise this is my last tuning attempt with time management simplification ever, but I got +4 elo with 200K@40/2 in movestogo case, and I want to put this method (and SPSA) on ultimate temptations.