Stockfish Testing Queue

Finished - 19388 tests

26-08-13 jh move_importance diff
LLR: -2.95 (-2.94,2.94)
Total: 12589 W: 2556 L: 2628 D: 7405
sprt @ 15+0.05 th 1 Shorten move importance array.
26-08-13 pe opp_bishops diff
LLR: -2.02 (-2.94,2.94)
Total: 42777 W: 9074 L: 9025 D: 24678
sprt @ 15+0.05 th 1 Discourage trading down when ahead with opposite bishops. Take 6: Limit effect only to 2 last exchanges before transition to opposite bishop endgame centering around original scaling factor
26-08-13 mc slow_mover diff
LLR: -2.96 (-2.94,2.94)
Total: 7085 W: 1423 L: 1511 D: 4151
sprt @ 15+0.05 th 1 Set slow mover to 75 instead of 50
26-08-13 pe opp_bishops diff
LLR: -2.95 (-2.94,2.94)
Total: 14976 W: 2820 L: 2846 D: 9310
sprt @ 60+0.05 th 1 LTC. Discourage trading down when ahead with opposite bishops. Take 7. Decrease penalty for traded piece
25-08-13 ur time5 diff
ELO: 5.20 +-2.8 (95%) LOS: 100.0%
Total: 19311 W: 3380 L: 3091 D: 12840
20000 @ 120+0.05 th 1 testing as marco suggested 20,000 games at longer time control.
26-08-13 pe opp_bishops diff
LLR: 2.95 (-2.94,2.94)
Total: 7814 W: 1694 L: 1564 D: 4556
sprt @ 15+0.05 th 1 Discourage trading down when ahead with opposite bishops. Take 7. Decrease penalty for traded piece
26-08-13 ra Passed_Pawns_C2 diff
LLR: -2.96 (-2.94,2.94)
Total: 34772 W: 6586 L: 6508 D: 21678
sprt @ 60+0.05 th 1 Ltc:If our rook is behind passed pawn, all squares are defended. take 2.
26-08-13 sg bm_extension_reduced_an diff
LLR: -2.96 (-2.94,2.94)
Total: 10662 W: 2261 L: 2339 D: 6062
sprt @ 15+0.05 th 1 Botvinnik-Markov extension: if last move reduced and move is non quiet or refutes threat move
25-08-13 pe opp_bishops diff
LLR: -2.95 (-2.94,2.94)
Total: 8215 W: 1755 L: 1840 D: 4620
sprt @ 15+0.05 th 1 Discourage trading down when ahead with opposite bishops. Take 5: Shift scale a little more up
25-08-13 pe opp_bishops diff
LLR: -2.96 (-2.94,2.94)
Total: 9611 W: 2019 L: 2100 D: 5492
sprt @ 15+0.05 th 1 Discourage trading down when ahead with opposite bishops. Take 4: Surprisingly take 2 failed strongly. Shift little in opposite direction. Countdown now starts from 4 extra pieces
25-08-13 jk capPrune diff
LLR: -2.96 (-2.94,2.94)
Total: 6503 W: 1331 L: 1421 D: 3751
sprt @ 15+0.05 th 1 Bad captures pruning, 256
25-08-13 jk capPrune diff
LLR: -2.97 (-2.94,2.94)
Total: 6706 W: 1393 L: 1483 D: 3830
sprt @ 15+0.05 th 1 bad captures futility pruning, 128
25-08-13 tv 18796f04 diff
LLR: -2.94 (-2.94,2.94)
Total: 14203 W: 2584 L: 2615 D: 9004
sprt @ 60+0.05 th 1 LTC of opposition
25-08-13 jk time5 diff
LLR: 2.97 (-2.94,2.94)
Total: 7892 W: 1544 L: 1438 D: 4910
sprt @ 40/30 th 1 Last test for this patch: SPRT: -3 +3. 40moves/30 seconds
25-08-13 sg bm_extension_reduced_an diff
LLR: -2.95 (-2.94,2.94)
Total: 11231 W: 2378 L: 2454 D: 6399
sprt @ 15+0.05 th 1 Botvinnik-markov extension: take 6 (last move have to reduced and move refutes threat move)
25-08-13 mc search diff
LLR: -2.95 (-2.94,2.94)
Total: 17955 W: 3678 L: 3735 D: 10542
sprt @ 15+0.05 th 1 Inspired by Passed_Pawns_F results
25-08-13 pe opp_bishops diff
LLR: -2.95 (-2.94,2.94)
Total: 4705 W: 958 L: 1053 D: 2694
sprt @ 15+0.05 th 1 Discourage trading down when ahead with opposite bishops. Take 2: make effect larger by shifting scale down
25-08-13 mc master diff
ELO: -0.07 +-3.1 (95%) LOS: 48.2%
Total: 20000 W: 4087 L: 4091 D: 11822
20000 @ 15+0.05 th 1 Verify Chris fix of KBPK bug
25-08-13 sg bm_extension_refute_onl diff
LLR: -2.96 (-2.94,2.94)
Total: 9911 W: 2018 L: 2098 D: 5795
sprt @ 15+0.05 th 1 Botvinnik-markov extension: take 5 (extend only if move refutes threat move)
25-08-13 tv f339ac8 diff
LLR: -2.95 (-2.94,2.94)
Total: 11236 W: 2348 L: 2424 D: 6464
sprt @ 15+0.05 th 1 12 bonus 6 malus see https://groups.google.com/forum/?fromgroups=#!topic/fishcooking/t96AM8Wiz7o
25-08-13 tv eea64a7 diff
LLR: -2.97 (-2.94,2.94)
Total: 9708 W: 1983 L: 2064 D: 5661
sprt @ 15+0.05 th 1 8 bonus 4 malus see https://groups.google.com/forum/?fromgroups=#!topic/fishcooking/t96AM8Wiz7o
25-08-13 jk prob2 diff
LLR: -2.96 (-2.94,2.94)
Total: 4814 W: 855 L: 935 D: 3024
sprt @ 60+0.05 th 1 LTC: Legacy probCut when depth >= 2 plies
25-08-13 ee AlwaysDancingII diff
LLR: -2.95 (-2.94,2.94)
Total: 3610 W: 719 L: 817 D: 2074
sprt @ 15+0.05 th 1 Take 2. Many changes because the old version 20130731 is hard to beat under my testconditions. It still is. I made some comments here in last post of https://groups.google.com/forum/?fromgroups=#!topic/fishcooking/FJ606_kDFRo
25-08-13 ra Passed_Pawns_C2 diff
LLR: 2.95 (-2.94,2.94)
Total: 11335 W: 2469 L: 2328 D: 6538
sprt @ 15+0.05 th 1 If our rook is behind passed pawn, all squares are defended. take 2.
25-08-13 ur time5 diff
LLR: 2.96 (-2.94,2.94)
Total: 6703 W: 1238 L: 1134 D: 4331
sprt @ 30+0.5 th 1 decided to test it with -3 3 as other suggested and I plan to stop the test with fixed number of games.
25-08-13 pe opp_bishops diff
LLR: -2.96 (-2.94,2.94)
Total: 23197 W: 4839 L: 4881 D: 13477
sprt @ 15+0.05 th 1 Discourage trading down when ahead with opposite bishops. Take 3: Make effect larger by penalizing more for every extra piece traded
24-08-13 tk Multiple-Not-Improving diff
LLR: -2.96 (-2.94,2.94)
Total: 17325 W: 3115 L: 3131 D: 11079
sprt @ 60+0.05 th 1 Long TC: Try 12
24-08-13 sg bm_extension_pv_only diff
LLR: -0.28 (-2.94,2.94)
Total: 14760 W: 3126 L: 3094 D: 8540
sprt @ 15+0.05 th 1 Botvinnik-Markox extension: only at PV nodes
24-08-13 sg bm_extension_reduced_on diff
LLR: -2.95 (-2.94,2.94)
Total: 40617 W: 8355 L: 8348 D: 23914
sprt @ 15+0.05 th 1 Botvinnik-Markov extension: take 5, only if last move reduced
24-08-13 ra Passed_Pawns_F diff
LLR: -2.96 (-2.94,2.94)
Total: 8536 W: 1744 L: 1828 D: 4964
sprt @ 15+0.05 th 1 bonus for positive see passed pawn push. take 2.
24-08-13 jk prob2 diff
LLR: 2.96 (-2.94,2.94)
Total: 19103 W: 4146 L: 3982 D: 10975
sprt @ 15+0.05 th 1 Legacy probCut when depth >= 2 plies
24-08-13 ur time5 diff
LLR: 2.96 (-2.94,2.94)
Total: 9266 W: 1777 L: 1624 D: 5865
sprt @ 60+0.05 th 1 same as last time4 that passed short time control
24-08-13 mc search diff
LLR: -2.97 (-2.94,2.94)
Total: 5124 W: 1029 L: 1123 D: 2972
sprt @ 15+0.05 th 1 Trade when down, keep when up
24-08-13 tv Opposition3 diff
LLR: -3.03 (-2.94,2.94)
Total: 21636 W: 4544 L: 4593 D: 12499
sprt @ 15+0.05 th 1 Diagonal bonus 16/8
23-08-13 pe mc_pruning diff
LLR: -2.95 (-2.94,2.94)
Total: 39186 W: 8347 L: 8342 D: 22497
sprt @ 15+0.05 th 1 Take 2: Increasing pruning of non-improving nodes at higher depths did not help. Check to see if opposite helps.
23-08-13 pe mc_pruning diff
LLR: -2.95 (-2.94,2.94)
Total: 11899 W: 2484 L: 2558 D: 6857
sprt @ 15+0.05 th 1 Increasing pruning of non-improving nodes at higher depths did not help. Check to see if opposite helps
23-08-13 tk Multiple-Not-Improving diff
LLR: -2.94 (-2.94,2.94)
Total: 9376 W: 1914 L: 1995 D: 5467
sprt @ 15+0.05 th 1 Try 14
23-08-13 tk Multiple-Not-Improving diff
LLR: 2.96 (-2.94,2.94)
Total: 12322 W: 2711 L: 2566 D: 7045
sprt @ 15+0.05 th 1 Try 12
23-08-13 sg bm_extension_depth_4 diff
LLR: -2.96 (-2.94,2.94)
Total: 6916 W: 1445 L: 1534 D: 3937
sprt @ 15+0.05 th 1 Botvinnik-Markov extension: take 4 (depth <= 4)
23-08-13 sg bm_extension_depth_6 diff
LLR: -2.97 (-2.94,2.94)
Total: 3781 W: 778 L: 877 D: 2126
sprt @ 15+0.05 th 1 Botvinnik-Markov extension: take 3 (depth <= 6)
23-08-13 tv Opposition2 diff
LLR: 2.95 (-2.94,2.94)
Total: 40830 W: 8761 L: 8535 D: 23534
sprt @ 15+0.05 th 1 Test code which resembles the best passed version http://tests.stockfishchess.org/tests/view/52133d9f0ebc593f4bb9a3c4
23-08-13 tv Opposition2 diff
LLR: -1.18 (-2.94,2.94)
Total: 45289 W: 9631 L: 9544 D: 26114
sprt @ 15+0.05 th 1 Bonus when not in zugzwang
23-08-13 sg bm_extension_depth_5 diff
LLR: -2.97 (-2.94,2.94)
Total: 7102 W: 1507 L: 1596 D: 3999
sprt @ 15+0.05 th 1 Botvinnik-Markov-Extension: take2 (depth<=5)
23-08-13 ur time4 diff
LLR: 2.93 (-2.94,2.94)
Total: 13895 W: 3030 L: 2882 D: 7983
sprt @ 15+0.05 th 1 double the change in time control in the patch that passed stage 1. Reducing slowmover to 50 does not work but maybe it works with using more time for pv instability and I simply deleted the division by 2 in the pv_instability function to double the change of using more time)
22-08-13 tv Opposition2 diff
LLR: -1.45 (-2.94,2.94)
Total: 47734 W: 10208 L: 10123 D: 27403
sprt @ 15+0.05 th 1 Add file/rank restriction
23-08-13 tv Opposition2 diff
LLR: -2.97 (-2.94,2.94)
Total: 15599 W: 3229 L: 3293 D: 9077
sprt @ 15+0.05 th 1 Bonus 8/8
23-08-13 sg bm_extension diff
LLR: -2.97 (-2.94,2.94)
Total: 4689 W: 965 L: 1061 D: 2663
sprt @ 15+0.05 th 1 Add Botvinnik-Markov search extension
23-08-13 ra Passed_Pawns_F diff
LLR: -2.95 (-2.94,2.94)
Total: 10284 W: 2178 L: 2257 D: 5849
sprt @ 15+0.05 th 1 Bonus for positive SEE passed pawn push
23-08-13 ur slow75 diff
LLR: -2.96 (-2.94,2.94)
Total: 17195 W: 3647 L: 3706 D: 9842
sprt @ 15+0.05 th 1 changing only the slow mover parameter to 75 as marco suggested
22-08-13 ra Passed_Pawns_E diff
LLR: -2.95 (-2.94,2.94)
Total: 9817 W: 2061 L: 2141 D: 5615
sprt @ 15+0.05 th 1 Allow double bonus for passed pawn that supported twice.