Stockfish Testing Queue

Finished - 24198 tests

17-04-15 sn backward3 diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 24125 W: 4606 L: 4647 D: 14872
sprt @ 15+0.05 th 1 Backward pawn penalty by opposed flag and rank
17-04-15 Ro QueenContactChecks diff
LLR: -2.95 (-2.94,2.94) [-3.00,1.00]
Total: 18827 W: 3561 L: 3754 D: 11512
sprt @ 15+0.05 th 1 Retire Queen contact checks AND Rook contact checks in evaluation
16-04-15 Ro RookSupportChecks diff
ELO: 0.21 +-2.8 (95%) LOS: 55.9%
Total: 25000 W: 5181 L: 5166 D: 14653
25000 @ 9+0.05 th 1 What is the value of Rook Contact Checks ?
17-04-15 SC pieceValuesMP_tuned diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 15738 W: 2970 L: 3034 D: 9734
sprt @ 15+0.05 th 1 Use values from SPSA tuning for ordering of captures in MovePicker.
17-04-15 sn backward3 diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 16107 W: 3166 L: 3229 D: 9712
sprt @ 15+0.05 th 1 Only consider backward pawns on semi-open files
17-04-15 Ro RookSupportChecks diff
LLR: 2.96 (-2.94,2.94) [-1.50,4.50]
Total: 8095 W: 1597 L: 1470 D: 5028
sprt @ 15+0.05 th 1 Do we need Rook Contact Checks Supported by Queen ? Simplification test.
17-04-15 Ro SafeChecks diff
ELO: -13.88 +-2.8 (95%) LOS: 0.0%
Total: 25000 W: 4882 L: 5880 D: 14238
25000 @ 9+0.05 th 1 What is the value of Queen and Rook safe checks ?
16-04-15 Ro RookSupportChecks diff
ELO: 1.11 +-2.8 (95%) LOS: 78.4%
Total: 25000 W: 5210 L: 5130 D: 14660
25000 @ 9+0.05 th 1 Do we need Rook Contact Checks Supported by Queen ? See git notes.
17-04-15 Ro QueenContactChecks diff
ELO: -0.51 +-2.8 (95%) LOS: 35.8%
Total: 25000 W: 5182 L: 5219 D: 14599
25000 @ 9+0.05 th 1 What is the value of Queen Contact Checks ?
17-04-15 lb test diff
LLR: -0.55 (-2.94,2.94) [0.00,4.50]
Total: 1273 W: 234 L: 255 D: 784
sprt @ 15+0.05 th 1 tuned ThreatenedByPawn array locally
17-04-15 lb test diff
LLR: 2.96 (-2.94,2.94) [-0.50,4.50]
Total: 6399 W: 1280 L: 1131 D: 3988
sprt @ 15+0.05 th 1 tuned ThreatenedByPawn array locally
17-04-15 Vo Talipia diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 5471 W: 979 L: 1070 D: 3422
sprt @ 15+0.05 th 1 Make sure move is OK before using CMH stat and sync up with LMR.
15-04-15 la rook_psqt_doe diff
ELO: -5.39 +-9.5 (95%) LOS: 13.4%
Total: 2000 W: 376 L: 407 D: 1217
2000 @ 15+0.05 th 1 DOE rook psqt, a-file, run 8
17-04-15 Fi easyCandidate diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 8302 W: 1591 L: 1675 D: 5036
sprt @ 15+0.05 th 1 Tweaked easy move parameters. Take 2
17-04-15 Ro SafeChecks diff
ELO: -13.08 +-4.7 (95%) LOS: 0.0%
Total: 8979 W: 1745 L: 2083 D: 5151
25000 @ 9+0.05 th 1 What is the value of safe checks ?
16-04-15 Vo Goldfish diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 8774 W: 1596 L: 1678 D: 5500
sprt @ 15+0.05 th 1 Quite Move Order and LMR Criteria: h+cmh*(depth+1)
15-04-15 la rook_psqt_doe diff
ELO: -5.39 +-9.6 (95%) LOS: 13.6%
Total: 2000 W: 382 L: 413 D: 1205
2000 @ 15+0.05 th 1 DOE rook psqt a-file, run 10
15-04-15 la rook_psqt_doe diff
ELO: -8.69 +-9.7 (95%) LOS: 3.9%
Total: 2000 W: 379 L: 429 D: 1192
2000 @ 15+0.05 th 1 DOE rook psqt a-file, run 7
15-04-15 la rook_psqt_doe diff
ELO: -4.52 +-9.4 (95%) LOS: 17.5%
Total: 2000 W: 373 L: 399 D: 1228
2000 @ 15+0.05 th 1 DOE rook psqt a-file, run 5
15-04-15 la rook_psqt_doe diff
ELO: -3.13 +-9.3 (95%) LOS: 25.5%
Total: 2000 W: 364 L: 382 D: 1254
2000 @ 15+0.05 th 1 DOE rook psqt a-file, run 16 last
15-04-15 la rook_psqt_doe diff
ELO: -11.47 +-9.6 (95%) LOS: 1.0%
Total: 2000 W: 366 L: 432 D: 1202
2000 @ 15+0.05 th 1 DOE rook psqt a-file, run 9
15-04-15 la rook_psqt_doe diff
ELO: 3.65 +-9.4 (95%) LOS: 77.7%
Total: 2000 W: 391 L: 370 D: 1239
2000 @ 15+0.05 th 1 DOE rook psqt a-file, run 12
16-04-15 jo doubled_opposed diff
LLR: -2.96 (-2.94,2.94) [0.00,6.00]
Total: 14623 W: 2323 L: 2357 D: 9943
sprt @ 60+0.05 th 1 LTC: Reintroduce opposed flag for doubled pawns.
15-04-15 la rook_psqt_doe diff
ELO: 4.17 +-9.4 (95%) LOS: 80.6%
Total: 2000 W: 398 L: 374 D: 1228
2000 @ 15+0.05 th 1 DOE rook psqt a-file, run 13
15-04-15 la rook_psqt_doe diff
ELO: 2.08 +-9.6 (95%) LOS: 66.4%
Total: 2000 W: 407 L: 395 D: 1198
2000 @ 15+0.05 th 1 DOE rook psqt a-file, run 11
15-04-15 la rook_psqt_doe diff
ELO: -3.65 +-9.4 (95%) LOS: 22.5%
Total: 2000 W: 375 L: 396 D: 1229
2000 @ 15+0.05 th 1 DOE rook psqt a-file, run 15
15-04-15 la rook_psqt_doe diff
ELO: -11.30 +-9.6 (95%) LOS: 1.1%
Total: 2000 W: 369 L: 434 D: 1197
2000 @ 15+0.05 th 1 DOE rook psqt a-file, run 6
15-04-15 la rook_psqt_doe diff
ELO: -4.17 +-9.5 (95%) LOS: 19.6%
Total: 2000 W: 381 L: 405 D: 1214
2000 @ 15+0.05 th 1 DOE rook psqt a-file, run 14
16-04-15 vi doubled_shelter2 diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 126250 W: 24397 L: 24164 D: 77689
sprt @ 15+0.05 th 1 SPSA values did worse, so try applying only half the change.
15-04-15 jo doubled_opposed diff
LLR: 2.96 (-2.94,2.94) [-1.50,4.50]
Total: 102659 W: 20059 L: 19676 D: 62924
sprt @ 15+0.05 th 1 Reintroduce opposed flag for doubled pawns.
16-04-15 lb safe_checks diff
LLR: -2.96 (-2.94,2.94) [0.00,6.00]
Total: 10857 W: 1691 L: 1743 D: 7423
sprt @ 60+0.05 th 1 reschedule with correct SPRT bounds at LTC for ajithc. Failed STC(-1.5,4.5)
15-04-15 vi doubled_shelter2 diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 45134 W: 8685 L: 8670 D: 27779
sprt @ 15+0.05 th 1 Doubled pawn effect on king shelter. Try SPSA tuned values. Not convinced about one of them, but only fair not to second-guess to begin with..
16-04-15 Fi scoreQuietsSquared diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 8564 W: 1575 L: 1658 D: 5331
sprt @ 15+0.05 th 1 Final long shot at a different way of scoring quiets.
16-04-15 Fi scoreQuietsMax diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 2472 W: 436 L: 536 D: 1500
sprt @ 15+0.05 th 1 Just a long shot at a different way of scoring quiets.
16-04-15 Fi scoreQuietsMin diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 1177 W: 168 L: 270 D: 739
sprt @ 15+0.05 th 1 Just a long shot at a different way of scoring quiets.
15-04-15 aj hanging_tuned diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 5536 W: 1016 L: 1107 D: 3413
sprt @ 15+0.05 th 1 spsa tuned values for hanging pieces: STC
16-04-15 SC pieceValuesMP_tune diff
27147/30000 iterations
58000/60000 games played
60000 @ 30+0.05 th 1 Tune capture score table. Rescheduling after fixing the nodestime in base. Again a bugfix, hope I got everything right this time.
15-04-15 Vo Minnow-3 diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 17588 W: 3394 L: 3453 D: 10741
sprt @ 15+0.05 th 1 Don't update CMH if depth<=1
15-04-15 aj safe_checks diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 11659 W: 2147 L: 2222 D: 7290
sprt @ 15+0.05 th 1 evaluate safe checks even when ei.kingAttackersCount[Them] = 0. This must have been an unintentional bug in existing code : STC
15-04-15 lb inline diff
ELO: 0.26 +-1.8 (95%) LOS: 61.2%
Total: 60000 W: 12437 L: 12392 D: 35171
60000 @ 9+0.03 th 1 verify that inline is inline is useless in fishtest also (ie. with the blend of compilers, mostly driven by the mingw cross compiler for windows machines)
14-04-15 vi doubled_shelter_tune diff
46106/25000 iterations
69000/50000 games played
50000 @ 30+0.05 th 1 Tune doubled pawn strength after yellow LTC result. (Resubmit after fixing errors in test spec)
14-04-15 la rook_psqt_doe diff
ELO: -2.78 +-9.7 (95%) LOS: 28.7%
Total: 2000 W: 397 L: 413 D: 1190
2000 @ 15+0.05 th 1 DOE for rook psqt, run 1
15-04-15 la rook_psqt_doe diff
ELO: -3.82 +-9.7 (95%) LOS: 21.9%
Total: 2000 W: 392 L: 414 D: 1194
2000 @ 15+0.05 th 1 DOE for rook psqt a-file, run 2
15-04-15 la rook_psqt_doe diff
ELO: -3.30 +-9.6 (95%) LOS: 25.1%
Total: 2000 W: 392 L: 411 D: 1197
2000 @ 15+0.05 th 1 DOE for rook psqt a-file, run 4
15-04-15 la rook_psqt_doe diff
ELO: -1.91 +-9.6 (95%) LOS: 34.9%
Total: 2000 W: 395 L: 406 D: 1199
2000 @ 15+0.05 th 1 DOE for rook psqt a-file, run 3
14-04-15 sn supported_lever4 diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 31743 W: 6011 L: 6032 D: 19700
sprt @ 15+0.05 th 1 Supported levers: take 3. 
Add some bonus to the endgame values of levers.
15-04-15 Vo Minnow-2 diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 17864 W: 3391 L: 3449 D: 11024
sprt @ 15+0.05 th 1 Don't update stats if depth<=1, except for killers and counter moves.
15-04-15 Vo Minnow diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 24287 W: 4666 L: 4707 D: 14914
sprt @ 15+0.05 th 1 Don't update history stats if depth<=1.
14-04-15 sg quiet_move_order diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 34340 W: 6698 L: 6733 D: 20909
sprt @ 15+0.05 th 1 Allow bad quiet moves sorting at depth = 2
14-04-15 sg quiet_move_order diff
LLR: -3.28 (-2.94,2.94) [0.00,4.00]
Total: 28873 W: 5495 L: 5568 D: 17810
sprt @ 15+0.05 th 1 Allow bad quiet moves sorting only at depth >= 4