Stockfish Testing Queue

Finished - 18536 tests

20-10-13 mc safe_checks2 diff
ELO: 48.93 +-3.0 (95%) LOS: 100.0%
Total: 20000 W: 5458 L: 2660 D: 11882
20000 @ 60+0.05 th 1 Tentative regression test of Ralph 3x safe checks increase
19-10-13 mc master diff
ELO: 50.11 +-2.1 (95%) LOS: 100.0%
Total: 40000 W: 10547 L: 4817 D: 24636
40000 @ 60+0.05 th 1 Regression test after Ralph second increase of safe checks bonus
21-10-13 mc passed_fix diff
LLR: 2.96 (-2.94,2.94) [-1.50,4.50]
Total: 83831 W: 16466 L: 16133 D: 51232
sprt @ 15+0.05 th 1 Retest this passed pawns tweak after fixing pos.count<ALL_PIECES>
20-10-13 pe chains diff
LLR: -2.94 (-2.94,2.94) [-3.00,3.00]
Total: 26301 W: 4477 L: 4580 D: 17244
sprt @ 60+0.05 th 1 Two Ralph's successful versions versus each other.
20-10-13 mc master diff
ELO: 49.72 +-3.1 (95%) LOS: 100.0%
Total: 20000 W: 5520 L: 2677 D: 11803
20000 @ 60+0.05 th 1 Regression test after Ralph chain pawn patch
21-10-13 jo knight2 diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 8655 W: 1675 L: 1758 D: 5222
sprt @ 15+0.05 th 1 Give penalty to knight for each pawn less than 8.
21-10-13 lb test diff
LLR: -2.96 (-2.94,2.94) [-3.00,3.00]
Total: 747 W: 139 L: 254 D: 354
sprt @ 15+0.05 th 1 eval simplification, inspired by new chained pawn scoring.
21-10-13 gl queenrook diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 5320 W: 979 L: 1071 D: 3270
sprt @ 15+0.05 th 1 Decrease bonus for queen vs rooks
20-10-13 mc master^ diff
LLR: -2.97 (-2.94,2.94) [-3.00,3.00]
Total: 5564 W: 1082 L: 1190 D: 3292
sprt @ 15+0.05 th 1 Validate the pos.count<ALL_PIECES> fix. In particular passed pawns rule to increase the bonus if we have more non-pawn pieces
20-10-13 sg lmr_conditions diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 3620 W: 643 L: 739 D: 2238
sprt @ 15+0.05 th 1 LMR: reduce more if SEE<0 and static eval + futility margin <= alpha
20-10-13 sg lmr_conditions diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 34542 W: 6676 L: 6689 D: 21177
sprt @ 15+0.05 th 1 LMR: reduce more if SEE<0 and depth > 2*ply
20-10-13 sg lmr_conditions diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 13718 W: 2606 L: 2675 D: 8437
sprt @ 15+0.05 th 1 LMR: reduce more if SEE<0 and 2*depth > ply
20-10-13 sg lmr_conditions diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 21264 W: 4061 L: 4110 D: 13093
sprt @ 15+0.05 th 1 LMR: reduce more if SEE<0 and depth > ply (ignore !PvNode && cutNode). Schedule some tests because framework almost empty.
20-10-13 jo knight1 diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 5199 W: 985 L: 1078 D: 3136
sprt @ 15+0.05 th 1 Give small bonus to knight for each additional pawn more than 8.
20-10-13 gl lmr_conditions diff
LLR: -2.95 (-2.94,2.94) [0.00,6.00]
Total: 6760 W: 1096 L: 1166 D: 4498
sprt @ 60+0.05 th 1 (speculative long tc) LMR: reduce more if SEE<0 and depth > ply
20-10-13 jo bishop2 diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 3190 W: 623 L: 722 D: 1845
sprt @ 15+0.05 th 1 Try opposite of take 1. Give penalty for opponent's pawns on that colour as well.
20-10-13 sg lmr_conditions diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 9395 W: 1779 L: 1860 D: 5756
sprt @ 15+0.05 th 1 LMR: reduce more if SEE < 0 and static eval <= alpha
20-10-13 jo bishop1 diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 3181 W: 583 L: 681 D: 1917
sprt @ 15+0.05 th 1 Bishop with same coloured pawns gets a penalty for each own pawn of that colour. Why not a bonus for each opponent's pawn of that colour?
20-10-13 tv BishopPawn diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 4992 W: 932 L: 1025 D: 3035
sprt @ 15+0.05 th 1 Take 2, no bonus when we have the bishop pair
20-10-13 do imbalance diff
ELO: -54.66 +-3.4 (95%) LOS: 0.0%
Total: 20000 W: 3446 L: 6567 D: 9987
20000 @ 15+0.05 th 1 Measure ELO of imbalance evaluation.
20-10-13 tv BishopPawn diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 3974 W: 757 L: 853 D: 2364
sprt @ 15+0.05 th 1 Bonus for passed pawn with matching bishop
20-10-13 pe chains_opposed diff
LLR: 2.96 (-2.94,2.94) [0.00,6.00]
Total: 9738 W: 1797 L: 1644 D: 6297
sprt @ 60+0.05 th 1 LTC attempt to improve pawn chains take 2
19-10-13 do rooks diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 8223 W: 1575 L: 1659 D: 4989
sprt @ 15+0.05 th 1 Remove redundant rook penalty.
19-10-13 rs chains_opposed diff
LLR: 2.96 (-2.94,2.94) [-1.50,4.50]
Total: 28299 W: 5854 L: 5667 D: 16778
sprt @ 15+0.05 th 1 attempt to improve pawn chains take 2
19-10-13 mc test diff
LLR: -2.97 (-2.94,2.94) [0.00,6.00]
Total: 6239 W: 1041 L: 1114 D: 4084
sprt @ 60+0.05 th 1 LTC for Lucas: try to fix various orthogonality issues introduced by chained bonus patch: take 1.
19-10-13 jo congregation1 diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 8747 W: 1658 L: 1741 D: 5348
sprt @ 15+0.05 th 1 Penalty for more than 4 pieces on same file or rank. Kicks in very rare, bench unchanged, but I want to give it a try.
19-10-13 lb test diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 18066 W: 3557 L: 3602 D: 10907
sprt @ 15+0.05 th 1 try to restore orthogonality after chained patch: take 2.
19-10-13 lb test diff
LLR: 2.95 (-2.94,2.94) [0.00,5.00]
Total: 28272 W: 5731 L: 5474 D: 17067
sprt @ 15+0.05 th 1 try to fix various orthogonality issues introduced by chained bonus patch: take 1.
19-10-13 sg lmr_conditions diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 9496 W: 1793 L: 1874 D: 5829
sprt @ 15+0.05 th 1 LMR: reduce more if SEE<0 and not dangerous
19-10-13 sg lmr_conditions diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 17466 W: 3421 L: 3480 D: 10565
sprt @ 15+0.05 th 1 LMR: reduce more if SEE<0 and no PV node and no cut node
19-10-13 sg lmr_conditions diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 28990 W: 5702 L: 5729 D: 17559
sprt @ 15+0.05 th 1 LMR: reduce more if SEE<0 and depth > ply
19-10-13 sg lmr_conditions diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 8441 W: 1614 L: 1698 D: 5129
sprt @ 15+0.05 th 1 LMR: reduce more if SEE<0 and singularExtensionNode
18-10-13 rs chains_from_joerg diff
LLR: 2.95 (-2.94,2.94) [0.00,6.00]
Total: 31858 W: 5497 L: 5240 D: 21121
sprt @ 60+0.05 th 1 LTC: attempt to improve on pawn chains
18-10-13 mc safe_checks diff
LLR: -2.96 (-2.94,2.94) [-3.00,3.00]
Total: 44484 W: 7360 L: 7463 D: 29661
sprt @ 60+0.05 th 1 Pit the 2 safe_checks successful versions from Ralph
18-10-13 mc chain_tune diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 4100 W: 734 L: 829 D: 2537
sprt @ 15+0.05 th 1 Fine tune chain bonus: take 2
18-10-13 mc chain_tune^ diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 14092 W: 2683 L: 2751 D: 8658
sprt @ 15+0.05 th 1 Fine tune chain bonus: take 1
18-10-13 pe storm diff
LLR: -2.97 (-2.94,2.94) [0.00,6.00]
Total: 9430 W: 1537 L: 1595 D: 6298
sprt @ 60+0.05 th 1 LTC for neutral at short TC. tune rank 3 storm
18-10-13 rs chains_from_joerg diff
LLR: 2.96 (-2.94,2.94) [-1.50,4.50]
Total: 23460 W: 4727 L: 4556 D: 14177
sprt @ 15+0.05 th 1 attempt to improve on pawn chains
18-10-13 pe storm diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 15491 W: 3013 L: 3077 D: 9401
sprt @ 15+0.05 th 1 tune storm rank 3
18-10-13 gl queen_mate_threat diff
LLR: -2.96 (-2.94,2.94) [0.00,6.00]
Total: 20649 W: 3423 L: 3427 D: 13799
sprt @ 60+0.05 th 1 (Long TC since neutral at short TC) Queen contact mates - fix pin detection, reduce bonus, final try
17-10-13 rs queen_safe_check diff
LLR: -2.95 (-2.94,2.94) [0.00,6.00]
Total: 18595 W: 3109 L: 3122 D: 12364
sprt @ 60+0.05 th 1 LTC: more bonus for non-diagonal check, less bonus for diagonal check
17-10-13 rs safe_checks diff
LLR: 2.94 (-2.94,2.94) [0.00,6.00]
Total: 36065 W: 6176 L: 5900 D: 23989
sprt @ 60+0.05 th 1 LTC: more bonus for safe checks take 2
16-10-13 tv SE2 diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 53141 W: 10539 L: 10500 D: 32102
sprt @ 15+0.05 th 1 Adapt SE margin to move gain
15-10-13 rs safe_checks diff
LLR: 2.96 (-2.94,2.94) [0.00,6.00]
Total: 26334 W: 4540 L: 4310 D: 17484
sprt @ 60+0.05 th 1 LTC: more bonus for safe checks
11-10-13 pe storm diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 91962 W: 17973 L: 17830 D: 56159
sprt @ 15+0.05 th 1 tune rank 3 storm
15-10-13 gl queen_mate_threat diff
LLR: 0.63 (-2.94,2.94) [-1.50,4.50]
Total: 64997 W: 12604 L: 12407 D: 39986
sprt @ 15+0.05 th 1 Queen contact mates - fix pin detection, reduce bonus, final try
17-10-13 jo chained5 diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 6636 W: 1404 L: 1494 D: 3738
sprt @ 15+0.05 th 1 Chained pawn, Take 4. Further increase rank bonus. It would be areal surprise, if this would pass, as now the full bonus for candidate passers would be given to chain pawns, too.
17-10-13 rs chained3 diff
LLR: 2.97 (-2.94,2.94) [0.00,6.00]
Total: 4116 W: 769 L: 645 D: 2702
sprt @ 60+0.05 th 1 (LTC for Joerg) Chained pawn, Take 3. Increase rank bonus.
17-10-13 tv SE2 diff
LLR: -3.62 (-2.94,2.94) [-1.50,4.50]
Total: 25354 W: 4821 L: 4883 D: 15650
sprt @ 15+0.05 th 1 Take 2, halve the gain as margin
15-10-13 gl chained1 diff
LLR: -0.78 (-2.94,2.94) [0.00,6.00]
Total: 41717 W: 7170 L: 6994 D: 27553
sprt @ 60+0.05 th 1 (Long TC for joster) Bonus for more advanced chained pawns. I use 1/3 value of CandidatePassed[] array as a starting point.