Stockfish Testing Queue

Finished - 20665 tests

28-04-14 sn pinned_pawns diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 12751 W: 2245 L: 2317 D: 8189
sprt @ 15+0.05 th 1 Faster way to calculate attacks from pinned pawns, using a mask
28-04-14 sg no_king_mobility diff
LLR: -2.97 (-2.94,2.94) [0.00,6.00]
Total: 6428 W: 915 L: 988 D: 4525
sprt @ 60+0.05 th 1 LTC: skip razoring if king cannot move
28-04-14 mc master diff
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 24139 W: 4336 L: 4221 D: 15582
sprt @ 15+0.05 th 1 A quick test just to verify we don't have a regression with a supposed harmless patch.
28-04-14 in closed_position_storm diff
LLR: -2.96 (-2.94,2.94) [0.00,6.00]
Total: 14313 W: 2124 L: 2162 D: 10027
sprt @ 60+0.05 th 1 LTC: Take 3: Using Marco's definition of closed position, also a slightly better approach.
28-04-14 sg no_king_mobility diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 19503 W: 3390 L: 3445 D: 12668
sprt @ 15+0.05 th 1 skip null move if king cannot move
28-04-14 sg no_king_mobility diff
LLR: 2.95 (-2.94,2.94) [-1.50,4.50]
Total: 5951 W: 1117 L: 998 D: 3836
sprt @ 15+0.05 th 1 skip razoring if king cannot move
28-04-14 Ro tt_any_size diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 14629 W: 2218 L: 2286 D: 10125
sprt @ 15+0.05 th 3 Faster probe/store. Fair test, with both engines using the same hash size.
28-04-14 in spacetrade diff
LLR: -2.95 (-2.94,2.94) [0.00,6.00]
Total: 15604 W: 2315 L: 2347 D: 10942
sprt @ 60+0.05 th 1 LTC: Take 5
28-04-14 in closed_position_storm diff
LLR: 2.96 (-2.94,2.94) [-1.50,4.50]
Total: 23926 W: 4356 L: 4191 D: 15379
sprt @ 15+0.05 th 1 Take 3: Using Marco's definition of closed position, also a slightly better approach.
28-04-14 in spacetrade diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 23875 W: 4260 L: 4304 D: 15311
sprt @ 15+0.05 th 1 Final take // CLOP
27-04-14 pe sf_dd_eval_based_contem diff
ELO: 0.54 +-2.1 (95%) LOS: 69.4%
Total: 40000 W: 7538 L: 7476 D: 24986
40000 @ 15+0.05 th 1 Test how eval based contempt performs against stronger opponent. 1. Confirm that eval based contempt improvement for sf dd.
28-04-14 in spacetrade diff
LLR: 2.96 (-2.94,2.94) [-1.50,4.50]
Total: 10071 W: 1909 L: 1778 D: 6384
sprt @ 15+0.05 th 1 Take 5
28-04-14 jh endgame_speedup diff
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 20365 W: 3735 L: 3613 D: 13017
sprt @ 15+0.05 th 1 Remove superfluous stalemate detection.
28-04-14 sn pinned_pawns diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 6196 W: 1080 L: 1169 D: 3947
sprt @ 15+0.05 th 1 Use pinned info for pawn attacks
28-04-14 Ro tt_any_size diff
LLR: 2.97 (-2.94,2.94) [-1.50,4.50]
Total: 4981 W: 831 L: 718 D: 3432
sprt @ 15+0.05 th 3 Arbitrary sized TT (with more threads and smaller TT)
27-04-14 in closed_position_minors2 diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 8617 W: 1486 L: 1568 D: 5563
sprt @ 15+0.05 th 1 Take 3: Values tuned with CLOP.
23-04-14 Me no50move diff
LLR: 1.42 (-2.94,2.94) [-1.50,4.50]
Total: 135627 W: 24004 L: 23615 D: 88008
sprt @ 15+0.05 th 1 is it possible that 50moverole and stalemates should have same need to know status ?
27-04-14 lb simplify_king_safety diff
ELO: -3.01 +-2.2 (95%) LOS: 0.3%
Total: 40000 W: 8001 L: 8347 D: 23652
40000 @ 6+0.03 th 1 measure value of contact checks at fast tc
26-04-14 hx mob_area diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 15007 W: 2647 L: 2713 D: 9647
sprt @ 15+0.05 th 1 bishop-rook xray
26-04-14 hx mob_area diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 6823 W: 1239 L: 1327 D: 4257
sprt @ 15+0.05 th 1 no rook-rook xray
26-04-14 gl master diff
ELO: 57.08 +-2.2 (95%) LOS: 100.0%
Total: 40000 W: 11371 L: 4858 D: 23771
40000 @ 60+0.05 th 1 Regression test after "speed picking of killers"
23-04-14 in rook_passers3 diff
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 106327 W: 16258 L: 16265 D: 73804
sprt @ 60+0.05 th 1 LTC: Remove rook passers (test as simplification)
27-04-14 lb simplify_king_safety diff
ELO: -13.10 +-2.2 (95%) LOS: 0.0%
Total: 40000 W: 7704 L: 9212 D: 23084
40000 @ 6+0.03 th 1 measure the value of safe distant checks at fast tc.
27-04-14 Ro tt_any_size diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 2522 W: 399 L: 497 D: 1626
sprt @ 15+0.05 th 1 Support arbitrary (<any> MB) sized TT.
26-04-14 sn little_combinaisons3 diff
LLR: 2.95 (-2.94,2.94) [-1.50,4.50]
Total: 8519 W: 1565 L: 1440 D: 5514
sprt @ 15+0.05 th 1 Remove the attack vs defense test
26-04-14 in closed_position_minors2 diff
LLR: -2.94 (-2.94,2.94) [-1.50,4.50]
Total: 6150 W: 1032 L: 1120 D: 3998
sprt @ 15+0.05 th 1 Based on Marco's definition of closed position (initial idea by Lyudmil Tsvetkov)
26-04-14 mc closed_position diff
LLR: -2.95 (-2.94,2.94) [0.00,6.00]
Total: 11064 W: 1619 L: 1671 D: 7774
sprt @ 60+0.05 th 1 LTC: Generalized code position detector: take2
26-04-14 ra outpost_pawns_A diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 16444 W: 2923 L: 2986 D: 10535
sprt @ 15+0.05 th 1 Bonus for outpost knight attacking weak pawn
26-04-14 sn little_combinaisons2 diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 13677 W: 2329 L: 2399 D: 8949
sprt @ 15+0.05 th 1 Little combinaisons with values tuned by CLOP
26-04-14 mc closed_position diff
LLR: 2.97 (-2.94,2.94) [-1.50,4.50]
Total: 12068 W: 2215 L: 2080 D: 7773
sprt @ 15+0.05 th 1 Generalized code position detector: take2
26-04-14 mc closed_position diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 24476 W: 4329 L: 4371 D: 15776
sprt @ 15+0.05 th 1 Generalized code position detector
26-04-14 mc closed_position diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 28524 W: 5029 L: 5061 D: 18434
sprt @ 15+0.05 th 1 Test closed position detector as described by Lyudmil.
26-04-14 in minor_behind_pawn diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 18549 W: 3244 L: 3301 D: 12004
sprt @ 15+0.05 th 1 Take 4: Based on corrections proposed by Lyudmil Tsvetkov.
26-04-14 jh contempt_2 diff
ELO: -0.81 +-2.3 (95%) LOS: 24.2%
Total: 32000 W: 5693 L: 5768 D: 20539
32000 @ 15+0.05 th 1 Attempt to verify source of ELO gain.
25-04-14 jh contempt diff
ELO: -1.29 +-2.3 (95%) LOS: 13.2%
Total: 32000 W: 5622 L: 5741 D: 20637
32000 @ 15+0.05 th 1 Test whether draw avoidance or search preferencing is more important.
23-04-14 pe contempt diff
LLR: -2.97 (-2.94,2.94) [0.00,6.00]
Total: 23451 W: 3461 L: 3460 D: 16530
sprt @ 60+0.05 th 1 LTC. Decrease eval-based contempt for negative eval. I scheduled it for now with low priority in case queue ever gets empty.
25-04-14 hx mob_area diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 917 W: 136 L: 240 D: 541
sprt @ 15+0.05 th 1 queen xray attacks
25-04-14 in knight_pawn_span diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 13013 W: 2319 L: 2390 D: 8304
sprt @ 15+0.05 th 1 Take 2: Higher penalty.
25-04-14 in minor_behind_pawn diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 16005 W: 2787 L: 2851 D: 10367
sprt @ 15+0.05 th 1 Final take: Lower bonus.
25-04-14 in closed_position_minors diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 7275 W: 1300 L: 1386 D: 4589
sprt @ 15+0.05 th 1 Take 2 (some tuning)
25-04-14 in stormdanger diff
LLR: -2.97 (-2.94,2.94) [0.00,6.00]
Total: 15491 W: 2360 L: 2392 D: 10739
sprt @ 60+0.05 th 1 One last try: LTC: Final Take: 75K iterations.
25-04-14 ra shelter diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 3896 W: 696 L: 791 D: 2409
sprt @ 15+0.05 th 1 Penalize undefended shelter pawns.
25-04-14 in minor_behind_pawn diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 23396 W: 4148 L: 4193 D: 15055
sprt @ 15+0.05 th 1 Take 2: Smaller bonus
25-04-14 in minor_behind_pawn diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 15929 W: 2749 L: 2813 D: 10367
sprt @ 15+0.05 th 1 Bonus for minor behind pawn that is part of a chain (Idea by Lyudmil from talkchess)
25-04-14 in knight_pawn_span diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 23586 W: 4223 L: 4267 D: 15096
sprt @ 15+0.05 th 1 Penalty for knights when pawns are far apart in endgame.
25-04-14 jr 3fold_root diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 8866 W: 1546 L: 1628 D: 5692
sprt @ 15+0.05 th 1 Avoid 3 fold repetition only at root node. Bench signature is the same, but it shoud win more games by avoiding repetitions.
25-04-14 My 3_pawns_vs_minor diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 18339 W: 3184 L: 3242 D: 11913
sprt @ 15+0.05 th 1 Favour 3 pawns over a minor (Take 2).
25-04-14 lb contempt diff
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 25548 W: 4046 L: 3933 D: 17569
sprt @ 60+0.05 th 1 LTC: use fixd contempt. Now that many attempts have been made to tune the weird contempt patch, it's time to LTC this one.
25-04-14 in stormdanger diff
LLR: 2.96 (-2.94,2.94) [-1.50,4.50]
Total: 8384 W: 1581 L: 1455 D: 5348
sprt @ 15+0.05 th 1 One last try: STC: Final Take: 75K iterations.
25-04-14 My 3pawns_vs_minor diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 21002 W: 3678 L: 3729 D: 13595
sprt @ 15+0.05 th 1 Favour 3 pawns over a minor