Stockfish Testing Queue

Finished - 40562 tests

17-02-01 Gua combo+tmr diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 2584 W: 410 L: 524 D: 1650
sprt @ 10+0.1 th 1 combo+TrappedMajorRook
17-02-01 Gua combo diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 47365 W: 8432 L: 8429 D: 30504
sprt @ 10+0.1 th 1 2 not bad tunes together
17-02-01 sni king_danger_tuned diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 76521 W: 13773 L: 13670 D: 49078
sprt @ 10+0.1 th 1 Try tuned values for king danger
17-02-01 sg null_history diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 11276 W: 1960 L: 2038 D: 7278
sprt @ 10+0.1 th 1 Do no nullmove pruning for about the 1% lowest values for depth >= 13
17-02-01 pb0 lmr_check diff
LLR: -2.96 (-2.94,2.94) [-3.00,1.00]
Total: 7049 W: 1198 L: 1367 D: 4484
sprt @ 10+0.1 th 1 I hope this qualifies for simplification with [-3,1] bounds. Local stc-result: W: 57 L: 47 D:196 Tot: 300
17-02-01 sni other3 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 6024 W: 1046 L: 1146 D: 3832
sprt @ 10+0.1 th 1 Tweak other checks
17-02-01 sg null_history diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 8862 W: 1500 L: 1588 D: 5774
sprt @ 10+0.1 th 1 Adjusting the move reduction depth shows no success. So try another usage of the nullmove history stats. Do no nullmove pruning for about the 1% lowest values.
17-02-01 Gua NoThreats diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 3821 W: 674 L: 784 D: 2363
sprt @ 10+0.1 th 1 Bonus if there are no opponent pieces besides king in king camp
17-02-01 pb0 lmr_king_move diff
LLR: -2.94 (-2.94,2.94) [0.00,4.00]
Total: 5878 W: 993 L: 1129 D: 3756
sprt @ 10+0.1 th 1 lmr: don't reduce king moves (submit with [0,4] bounds as complexity doesn't change IMO)
17-02-01 Mys pp diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 21642 W: 3852 L: 3886 D: 13904
sprt @ 10+0.1 th 1 favour blocking passers with minors
17-02-01 pb0 capture_escape diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 13505 W: 2449 L: 2517 D: 8539
sprt @ 10+0.1 th 1 Cache see calls detecting capture escapes (idea inspired by VoyagerOne). On a normal bench run cache stats are as follows: Total 994038 Hits 508906 hit rate (%) 51. Also this is more accurate in case of absolute pins & check evasions. (bugfixed version)
17-02-01 Elb kingpawn diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 17405 W: 3130 L: 3181 D: 11094
sprt @ 10+0.1 th 1 Take 2: a more correct version
17-02-01 pb0 capture_escape diff
Pending...
sprt @ 10+0.1 th 1 Cache see calls detecting capture escapes (idea inspired by VoyagerOne). On a normal bench run cache stats are as follows: Total 1051169 Hits 498380 hit rate (%) 47. Furthermore this is more accurate in case of absolute pins.
17-01-31 sni tune_king_danger2 diff
19362/20000 iterations
39559/40000 games played
40000 @ 20+0.2 th 1 Tune king danger
17-02-01 Elb kingpawn diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 17996 W: 3233 L: 3282 D: 11481
sprt @ 10+0.1 th 1 Bonus if opposing king can castle but has to defend a pawn.
17-01-30 fau pfntune diff
19678/20000 iterations
40000/40000 games played
40000 @ 50+0.5 th 1 As suggested by Rocky, increased ck to 10 and continuing.
17-01-31 jos matimb_fix diff
ELO: -3.82 +-2.5 (95%) LOS: 0.1%
Total: 30000 W: 5735 L: 6065 D: 18200
30000 @ 7+0.07 th 1 Values have noticeably changed. To the better or worse?
17-02-01 Voy hs diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 36253 W: 6434 L: 6407 D: 23412
sprt @ 10+0.1 th 1 history split idea revisit...
17-02-01 fau PfnTuned diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 10522 W: 1864 L: 1945 D: 6713
sprt @ 10+0.1 th 1 Testing the tuned values.
17-02-01 Roc PawnLessFlankSimplify diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 13744 W: 2469 L: 2536 D: 8739
sprt @ 10+0.1 th 1 Take 2
17-01-31 sni windmill3' diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 17244 W: 3093 L: 3145 D: 11006
sprt @ 10+0.1 th 1 Try with windmill king danger penalty = 400
17-01-31 Voy cf-cn diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 12033 W: 2120 L: 2194 D: 7719
sprt @ 10+0.1 th 1 Apply correction factor only for cutnodes...
17-01-31 sni fork3 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 15726 W: 2759 L: 2818 D: 10149
sprt @ 10+0.1 th 1 Take 2
17-01-31 Gua sf diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 5542 W: 936 L: 1038 D: 3568
sprt @ 10+0.1 th 1 try to increase sf for pawns imbalance
17-01-31 sni fork3 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 7914 W: 1385 L: 1477 D: 5052
sprt @ 10+0.1 th 1 Bonus for safe queen checks threatening to capture loose material
17-01-31 Voy hpt diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 8149 W: 1425 L: 1516 D: 5208
sprt @ 10+0.1 th 1 stc
17-01-31 sni windmill3 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 23468 W: 4190 L: 4216 D: 15062
sprt @ 10+0.1 th 1 Include smaller penalty for normal discovered checks too
17-01-31 Gua sf diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 9978 W: 1733 L: 1816 D: 6429
sprt @ 10+0.1 th 1 Scale factor for rook endings only 3-2 pawns and less
17-01-31 Roc QueenAlign diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 24300 W: 4396 L: 4418 D: 15486
sprt @ 10+0.1 th 1 Take 2, normal bonus
17-01-31 sg rookMob diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 7719 W: 1347 L: 1439 D: 4933
sprt @ 10+0.1 th 1 includes squares with opponent non-pawn pieces
17-01-31 Voy queenFreedom'' diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 9553 W: 1659 L: 1744 D: 6150
sprt @ 10+0.1 th 1 Bonus for number of paths queen can take...
17-01-31 Roc PawnLessFlankSimplify diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 13936 W: 2431 L: 2497 D: 9008
sprt @ 10+0.1 th 1 Running 0, 5 just to see how far it goes before trying some -3, 1
17-01-31 Roc QueenAlign diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 7435 W: 1306 L: 1400 D: 4729
sprt @ 10+0.1 th 1 Compute missing queen "other checks" (but only for the obvious aligned case)
17-01-31 Gua sf diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 8036 W: 1368 L: 1459 D: 5209
sprt @ 10+0.1 th 1 Another scale factor try for 4-3 rook endings and less
17-01-31 sg rookMob diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 5389 W: 952 L: 1055 D: 3382
sprt @ 10+0.1 th 1 My try on rook mobility. Count by knight or bishop attacked squares if defended.
17-01-31 Elb ttStat diff
LLR: -1.16 (-2.94,2.94) [-3.00,1.00]
Total: 62714 W: 11216 L: 11384 D: 40114
sprt @ 10+0.1 th 1 Can we get rid of the captureOrPromotion check?
17-01-31 sg null_history diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 20463 W: 3707 L: 3745 D: 13011
sprt @ 10+0.1 th 1 Try negative correction
17-01-31 Mys ks diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 26680 W: 4722 L: 4788 D: 17170
sprt @ 10+0.1 th 1 seeing queen contacts are removed, do we need to worry about king safety with a queen or less anymore?
17-01-31 sni windmill2' diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 117327 W: 21270 L: 20897 D: 75160
sprt @ 10+0.1 th 1 Try after passed "safe2" patch
17-01-30 SC tweak diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 18052 W: 2850 L: 2947 D: 12255
sprt @ 5+0.05 th 7 Use tt depth instead of depth. Rebased. Try SMP as suggested by pb.
17-01-31 sni windmill2' diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 41396 W: 7449 L: 7399 D: 26548
sprt @ 10+0.1 th 1 Try Stefan's suggestion to exclude the squares where the king can recapture
17-01-31 sg null_history diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 16668 W: 3011 L: 3065 D: 10592
sprt @ 10+0.1 th 1 no correction
17-01-31 sg null_history diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 7118 W: 1217 L: 1312 D: 4589
sprt @ 10+0.1 th 1 Collect nullmove history stats indexed by last move to adjust reduction like in LMR
17-01-30 Voy ttStatBonus diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 26702 W: 4835 L: 4846 D: 17021
sprt @ 10+0.1 th 1 stc
17-01-31 Roc PopQueen diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 7441 W: 1319 L: 1413 D: 4709
sprt @ 10+0.1 th 1 Try popcount adjustment for QueenCheck on top of new safe check patch
17-01-31 Gua OnPawn diff
LLR: -2.94 (-2.94,2.94) [0.00,5.00]
Total: 16662 W: 2971 L: 3025 D: 10666
sprt @ 10+0.1 th 1 try to use OnPawn bonus for queen too
17-01-29 Roc MaxThreat2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 61142 W: 11141 L: 11005 D: 38996
sprt @ 10+0.1 th 1 with SPSA tuned values "as is "
17-01-30 Voy rookMob diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 17430 W: 3197 L: 3247 D: 10986
sprt @ 10+0.1 th 1 Don't include knight/bishop attack squares for rook mobility.
17-01-30 sni other2 diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 17441 W: 3133 L: 3230 D: 11078
sprt @ 10+0.1 th 1 Try using kingDanger in a more regular/simpler way (for midgame score, for other checks). Run as a parameter tweak against passed "safe2" patch.
17-01-30 fau pfntune diff
5068/20000 iterations
10381/40000 games played
40000 @ 50+0.5 th 1 Tuning pawnless flank us since its showing promising, at LTC since there is where it has some problems. (Please accept only after/if pfn2 fails).