Stockfish Testing Queue

Finished - 3733 tests

19-05-16 sg lmr_depth2 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 11251 W: 2424 L: 2533 D: 6294
sprt @ 10+0.1 th 1 Increase LMR depth if the whole path would be less than a third of root depth.
19-05-16 sg lmr_depth2 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 28547 W: 6248 L: 6272 D: 16027
sprt @ 10+0.1 th 1 More reduction for higher depth at non-PV nodes. reduction = depth/12.
19-05-16 sg lmr_depth2 diff
LLR: -2.94 (-2.94,2.94) [0.50,4.50]
Total: 25057 W: 5518 L: 5558 D: 13981
sprt @ 10+0.1 th 1 Increase LMR depth if the whole path would be less than the half of root depth.
19-05-16 sg lmr_singular_mc diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 12037 W: 2562 L: 2667 D: 6808
sprt @ 10+0.1 th 1 Use singularExtensionLMRmultiplier only if move count > 15.
19-05-15 sg outposts diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 66699 W: 14751 L: 14586 D: 37362
sprt @ 10+0.1 th 1 If outpost attacks opponent king keep bonus the same else decrease bonus by 1/8.
19-05-15 sg ks_outpost diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 33359 W: 7415 L: 7414 D: 18530
sprt @ 10+0.1 th 1 Try the opposite and give bishop outposts double weight. Bonus factor = 50.
19-05-15 sg pawn_phase^ diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 36673 W: 8198 L: 8180 D: 20295
sprt @ 10+0.1 th 1 Use quarter pawn value for phase calculation. Also adjust endgame limit.
19-05-15 sg pawn_phase diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 33740 W: 7477 L: 7474 D: 18789
sprt @ 10+0.1 th 1 Use quarter pawn value for phase calculation.
19-05-15 sg pawn_phase2 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 23327 W: 5167 L: 5216 D: 12944
sprt @ 10+0.1 th 1 Scale directly phase by pawns. Max allowed decrease of 1/8
19-05-15 sg pawn_phase2 diff
LLR: -2.94 (-2.94,2.94) [0.50,4.50]
Total: 972 W: 176 L: 343 D: 453
sprt @ 10+0.1 th 1 Scale directly phase by pawns.
19-05-15 sg pawn_phase diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 7665 W: 1659 L: 1786 D: 4220
sprt @ 10+0.1 th 1 Use quarter pawn value for phase calculation. Doesn't adjust midgame limit.
19-05-15 sg ks_outpost diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 37620 W: 8411 L: 8388 D: 20821
sprt @ 10+0.1 th 1 Use outposts which attack the enemy king area in king danger.
19-05-15 sg pawn_phase diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 22291 W: 4933 L: 4987 D: 12371
sprt @ 10+0.1 th 1 Use half pawn value for phase calculation.
19-05-15 sg ks_outpost diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 21596 W: 4805 L: 4862 D: 11929
sprt @ 10+0.1 th 1 Give knight outposts double weight. Bonus factor = 50.
19-05-15 sg root_order_new diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 20506 W: 4498 L: 4561 D: 11447
sprt @ 10+0.1 th 1 Collect the two last best quiet moves at root (like killers) and use them at root as killers. Retest after rebase. There where many search-based commits and the STC and LTC struggles. The old tests has been stopped.
19-05-15 sg pawn_phase diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 15609 W: 3496 L: 3583 D: 8530
sprt @ 10+0.1 th 1 Add pawns to phase calculation (Inspired by miguel's test).
19-05-15 sg ks_outpost diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 14486 W: 3203 L: 3296 D: 7987
sprt @ 10+0.1 th 1 Lower bonus 40.
19-05-15 sg ks_outpost diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 7379 W: 1604 L: 1733 D: 4042
sprt @ 10+0.1 th 1 Higher bonus 100.
19-05-13 sg root_order diff
LLR: 0.56 (-2.94,2.94) [0.00,3.50]
Total: 33590 W: 5795 L: 5666 D: 22129
sprt @ 60+0.6 th 1 Speculative LTC: Collect the two last best quiet moves at root (like killers) and use them at root as killers. The STC has already over 100k games but seems to fail yellow. I want to test if this scales well. Low throughput. STC has been set to prio -1.
19-05-15 sg outposts diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 28030 W: 6226 L: 6251 D: 15553
sprt @ 10+0.1 th 1 If outpost attacks opponent king keep bonus the same else decrease by 1/4.
19-05-15 sg outposts diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 10709 W: 2286 L: 2398 D: 6025
sprt @ 10+0.1 th 1 If outpost attacks opponent king increase bonus by 1/4 else kepp it the same.
19-05-15 sg outposts2 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 8271 W: 1783 L: 1907 D: 4581
sprt @ 10+0.1 th 1 If outpost attacks opponent king keep bonus the same else half bonus.
19-05-15 sg mobility2 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 1244 W: 226 L: 388 D: 630
sprt @ 10+0.1 th 1 Extended mobility to squares which are reachable in two moves.
19-05-15 sg mobility diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 1121 W: 186 L: 347 D: 588
sprt @ 10+0.1 th 1 Use additional scaling for new mobility.
19-05-15 sg outposts diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 19753 W: 4393 L: 4459 D: 10901
sprt @ 10+0.1 th 1 Depending if the king area is attacked change the outpost score by +-1/8.
19-05-15 sg mobility diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 1944 W: 358 L: 513 D: 1073
sprt @ 10+0.1 th 1 Calculate new extended mobility by looking one move ahead. Use then the average of old and new mobility.
19-05-14 sg root_order4c diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 101555 W: 22943 L: 22600 D: 56012
sprt @ 10+0.1 th 1 Use only pseudo legal moves.
19-05-13 sg singular_margin diff
LLR: -1.18 (-2.94,2.94) [0.00,4.00]
Total: 166797 W: 37765 L: 37162 D: 91870
sprt @ 10+0.1 th 1 1/8 nore singular depth margin.
19-05-14 sg root_order4 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 93631 W: 21106 L: 20803 D: 51722
sprt @ 10+0.1 th 1 Set at root a quiet move from later in pv as killer if missing.
19-05-14 sg root_order4a diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 48892 W: 10981 L: 10902 D: 27009
sprt @ 10+0.1 th 1 Fixed: used the wrong move as killer.
19-05-14 sg root_order5 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 34660 W: 7822 L: 7813 D: 19025
sprt @ 10+0.1 th 1 Use first quiet root move (beginning from second move) as killer if killer missing.
19-05-14 sg root_order4b diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 20278 W: 4525 L: 4589 D: 11164
sprt @ 10+0.1 th 1 Try now also the first move in the pv.
19-05-14 sg root_order4 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 6988 W: 1505 L: 1635 D: 3848
sprt @ 10+0.1 th 1 Replace killer also if not missing.
19-05-14 sg root_order4a diff
LLR: -0.89 (-2.94,2.94) [0.50,4.50]
Total: 10301 W: 2350 L: 2348 D: 5603
sprt @ 10+0.1 th 1 Set at root a quiet move from later in pv as killer if missing. Try whole pv.
19-05-14 sg root_ply1_order diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 24879 W: 5604 L: 5644 D: 13631
sprt @ 10+0.1 th 1 Collect now also for ply=1 the best two responses and use them as killers if found.
19-05-14 sg singular_margin3 diff
LLR: -1.01 (-2.94,2.94) [0.00,4.00]
Total: 78500 W: 17722 L: 17464 D: 43314
sprt @ 10+0.1 th 1 1/4 more singular depth margin.
19-05-13 sg singular_margin diff
LLR: -0.57 (-2.94,2.94) [0.00,4.00]
Total: 89500 W: 20031 L: 19707 D: 49762
sprt @ 10+0.1 th 1 Fix: Retest 1/16 more singular depth margin.
19-05-13 sg singular_margin^ diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 48990 W: 11065 L: 11034 D: 26891
sprt @ 10+0.1 th 1 Fix: i overlocked the factor 2 (Thanks to @Viz). So retest 1/16 less singular depth margin.
19-05-13 sg weak_pawns2 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 30573 W: 6828 L: 6840 D: 16905
sprt @ 10+0.1 th 1 Try quadratic distance. Half effect.
19-05-13 sg weak_pawns2 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 6350 W: 1382 L: 1516 D: 3452
sprt @ 10+0.1 th 1 Try quadratic distance. 1/3 effect.
19-05-13 sg weak_pawns2 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 7652 W: 1623 L: 1750 D: 4279
sprt @ 10+0.1 th 1 Try quadratic distance.
19-05-13 sg weak_pawns diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 37992 W: 8478 L: 8491 D: 21023
sprt @ 10+0.1 th 1 Increase endgame value of islated and backward pawns by 1/3.
19-05-13 sg singular_margin2a diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 36154 W: 8160 L: 8144 D: 19850
sprt @ 10+0.1 th 1 Use weighted average of ttValue and beta (ratio 3:1) as base margin.
19-05-13 sg weak_pawns2 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 19118 W: 4275 L: 4345 D: 10498
sprt @ 10+0.1 th 1 Further half bonus (0,3).
19-05-13 sg weak_pawns2 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 8542 W: 1894 L: 2017 D: 4631
sprt @ 10+0.1 th 1 Half bonus (0,6).
19-05-13 sg weak_pawns2 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 4056 W: 849 L: 994 D: 2213
sprt @ 10+0.1 th 1 use only weak pawns which are not defended.
19-05-13 sg weak_pawns2 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 2055 W: 424 L: 583 D: 1048
sprt @ 10+0.1 th 1 Endgame bonus(=12) for enemy weak pawns dependant on distance to our king.
19-05-13 sg singular_margin2 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 28786 W: 6543 L: 6563 D: 15680
sprt @ 10+0.1 th 1 Use average of ttValue and beta as base margin.
19-05-13 sg singular_margin^ diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 15580 W: 3498 L: 3585 D: 8497
sprt @ 10+0.1 th 1 1/16 less singular margin.
19-05-13 sg singular_margin2a diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 9223 W: 2035 L: 2164 D: 5024
sprt @ 10+0.1 th 1 Use beta as base singular margin.