Stockfish Testing Queue

Finished - 19388 tests

18-08-13 mc master diff
ELO: 56.66 +-3.3 (95%) LOS: 100.0%
Total: 20000 W: 6221 L: 2988 D: 10791
20000 @ 60+0.05 th 1 Last regression before TCEC with Leonid's patch in
17-08-13 tv QSStale3 diff
LLR: -2.97 (-2.94,2.94)
Total: 30628 W: 5462 L: 5412 D: 19754
sprt @ 60+0.05 th 1 Ltc: Stalemate detection for lone king in eval
18-08-13 tv QSStale3 diff
LLR: -2.97 (-2.94,2.94)
Total: 31655 W: 6471 L: 6490 D: 18694
sprt @ 15+0.05 th 1 Fix stalemate for king + pawns
11-08-13 mc master^ diff
ELO: -1.67 +-2.2 (95%) LOS: 6.8%
Total: 40000 W: 8183 L: 8375 D: 23442
40000 @ 15+0.05 th 1 Verify the fix on the improving condition does not regress
18-08-13 tv QSStale3 diff
LLR: -2.95 (-2.94,2.94)
Total: 27328 W: 5510 L: 5541 D: 16277
sprt @ 15+0.05 th 1 Stalemate for king + pawns
18-08-13 ur timetry diff
LLR: -2.95 (-2.94,2.94)
Total: 9773 W: 1953 L: 2033 D: 5787
sprt @ 15+0.05 th 1 simplification instead of having 2 variables of number of times that you changed your mind and number of times that you changed your mind in previous ply having only one variable.
17-08-13 do 2.3.1_patches diff
LLR: -2.97 (-2.94,2.94)
Total: 19045 W: 3409 L: 3417 D: 12219
sprt @ 60+0.05 th 1 Old patch commited after SF 2.3.1 and reverted few months later. Check with SPRT.
16-08-13 pe decrease_improving diff
LLR: -2.97 (-2.94,2.94)
Total: 23544 W: 4242 L: 4227 D: 15075
sprt @ 60+0.05 th 1 LTC: Decrease pruning if eval is improving. Take 7 : 50% of take 6. Pri is set to -1 for now. Maybe there is something better
17-08-13 tv QSStale3 diff
LLR: 2.96 (-2.94,2.94)
Total: 12017 W: 2558 L: 2416 D: 7043
sprt @ 15+0.05 th 1 Stalemate detection for lone king in eval
17-08-13 pe lmr diff
LLR: -2.95 (-2.94,2.94)
Total: 5111 W: 1004 L: 1097 D: 3010
sprt @ 15+0.05 th 1 Limit reductions for improved by smaller depths d < 10
17-08-13 pe decrease_improving diff
LLR: -2.95 (-2.94,2.94)
Total: 9067 W: 1593 L: 1651 D: 5823
sprt @ 60+0.05 th 1 LTC Shift by 2
16-08-13 jh easymove3 diff
ELO: -0.74 +-3.2 (95%) LOS: 32.7%
Total: 16000 W: 2877 L: 2911 D: 10212
16000 @ 60+0.05 th 1 Ensure new easymove not a significant regression at testable TC.
16-08-13 pe reshape_improving diff
LLR: -2.96 (-2.94,2.94)
Total: 29907 W: 6049 L: 6073 D: 17785
sprt @ 15+0.05 th 1 Check idea to pump more improving nodes in front, but then prune them more aggressively than master at higher depths. Here breakeven is at depth 7
16-08-13 tv QSStale2 diff
LLR: -2.97 (-2.94,2.94)
Total: 5807 W: 1154 L: 1246 D: 3407
sprt @ 15+0.05 th 1 Test for stalemate in QS (less generic but faster fix)
16-08-13 pe decrease_improving diff
LLR: 2.95 (-2.94,2.94)
Total: 17118 W: 3165 L: 2974 D: 10979
sprt @ 60+0.05 th 1 Long TC. Take 8. Try original functional from shifted by 1. This is mostly in between take 6 and take 7, but prunes more at higher depth so it may be better at LTC
16-08-13 tv QSStale2 diff
LLR: -2.97 (-2.94,2.94)
Total: 10753 W: 2194 L: 2272 D: 6287
sprt @ 15+0.05 th 1 Test for stalemate in QS (fix)
16-08-13 ra control diff
LLR: -2.97 (-2.94,2.94)
Total: 9323 W: 1879 L: 1961 D: 5483
sprt @ 15+0.05 th 1 Tweak jkiiski's control patch.
16-08-13 jk null_depth_1 diff
LLR: -2.96 (-2.94,2.94)
Total: 8470 W: 1684 L: 1768 D: 5018
sprt @ 15+0.05 th 1 Null move at depth 1
16-08-13 jk control diff
LLR: -2.95 (-2.94,2.94)
Total: 11186 W: 2358 L: 2434 D: 6394
sprt @ 15+0.05 th 1 Bonus for overall board control
16-08-13 ra Pawns_A8 diff
LLR: -2.95 (-2.94,2.94)
Total: 16916 W: 3411 L: 3471 D: 10034
sprt @ 15+0.05 th 1 Blocked isolated pawns. take 7
16-08-13 pe decrease_improving diff
LLR: 0.41 (-2.94,2.94)
Total: 19000 W: 3879 L: 3811 D: 11310
sprt @ 15+0.05 th 1 Shift by 3. It has front end like take 2, but greatly reduced numbers for higher depth to get more pruning for LTC
16-08-13 ra Pawns_A6 diff
LLR: -2.96 (-2.94,2.94)
Total: 4128 W: 789 L: 885 D: 2454
sprt @ 15+0.05 th 1 Blocked isolated pawns are worse than just opposed, but better than not opposed pawns.
16-08-13 pe decrease_improving diff
LLR: 2.96 (-2.94,2.94)
Total: 12787 W: 2652 L: 2509 D: 7626
sprt @ 15+0.05 th 1 Shift by 2
16-08-13 pe decrease_improving diff
LLR: 2.95 (-2.94,2.94)
Total: 15140 W: 3125 L: 2976 D: 9039
sprt @ 15+0.05 th 1 Take 8. Try original functional from shifted by 1. This is mostly in between take 6 and take 7, but prunes more at higher depth so it may be better at LTC
15-08-13 ra Pawns_A7 diff
LLR: -2.95 (-2.94,2.94)
Total: 42327 W: 8721 L: 8709 D: 24897
sprt @ 15+0.05 th 1 Try another definition for blocked
15-08-13 ra Pawns_A5 diff
LLR: -2.96 (-2.94,2.94)
Total: 2804 W: 527 L: 627 D: 1650
sprt @ 15+0.05 th 1 Penalty for blocked isolated pawns, little bug fix
15-08-13 pe decrease_improving diff
LLR: -2.96 (-2.94,2.94)
Total: 12758 W: 2315 L: 2354 D: 8089
sprt @ 60+0.05 th 1 LTC: Decrease pruning if eval is improving. Take 6 : 50% of take 5
15-08-13 pe decrease_improving diff
LLR: 2.97 (-2.94,2.94)
Total: 24908 W: 5245 L: 5066 D: 14597
sprt @ 15+0.05 th 1 Decrease pruning if eval is improving. Take 7 : 50% of take 6
15-08-13 ee fianchettoWeak diff
LLR: -2.95 (-2.94,2.94)
Total: 12985 W: 2719 L: 2790 D: 7476
sprt @ 15+0.05 th 1 Another take on Joona's fianchetto bonus code, plus a bonus for bishop controlling squares in front of pawns. It is, so far, reasonably stable (≈50%) in my local tests but not with high percentages above that
15-08-13 mc out_of_book diff
LLR: -2.96 (-2.94,2.94)
Total: 15256 W: 3035 L: 3100 D: 9121
sprt @ 15+0.05 th 1 Double available time on first move out of book
15-08-13 ra Pawns_A4 diff
LLR: -2.97 (-2.94,2.94)
Total: 3622 W: 702 L: 800 D: 2120
sprt @ 15+0.05 th 1 Give more penalty for blocked isolated pawn
15-08-13 ra Pawns_A3 diff
LLR: -2.97 (-2.94,2.94)
Total: 3556 W: 744 L: 844 D: 1968
sprt @ 15+0.05 th 1 Give bonus for advanced blocked pawns
15-08-13 ra Pawns_A1 diff
LLR: -2.97 (-2.94,2.94)
Total: 13500 W: 2741 L: 2811 D: 7948
sprt @ 15+0.05 th 1 Detect more backward pawns
15-08-13 pe decrease_improving diff
LLR: 2.95 (-2.94,2.94)
Total: 11198 W: 2326 L: 2188 D: 6684
sprt @ 15+0.05 th 1 Decrease pruning if eval is improving. Take 6 : 50% of take 5
15-08-13 jh easymove2 diff
LLR: -2.97 (-2.94,2.94)
Total: 22729 W: 4669 L: 4713 D: 13347
sprt @ 15+0.05 th 1 Increase base depth for easy move.
15-08-13 pe decrease_improving diff
LLR: -2.97 (-2.94,2.94)
Total: 17586 W: 3111 L: 3127 D: 11348
sprt @ 60+0.05 th 1 Long TC: Decrease pruning if eval is improving. Take 5. Lower curve of previous test by 1.
07-08-13 tv CounterMoves_A diff
ELO: -1.56 +-2.9 (95%) LOS: 14.6%
Total: 20000 W: 3607 L: 3697 D: 12696
20000 @ 60+0.05 th 1 Test at ltc Reuven's patch because I think this is a genuine fix with a potential small elo gain. Try to fix bug in null move counter moves
12-08-13 jk fianchetto_bishop diff
LLR: -2.96 (-2.94,2.94)
Total: 53492 W: 9909 L: 9737 D: 33846
sprt @ 60+0.05 th 1 LTC: Give a bonus for fianchettoed bishop protecting the king
14-08-13 pe decrease_improving diff
LLR: 2.96 (-2.94,2.94)
Total: 17344 W: 3600 L: 3444 D: 10300
sprt @ 15+0.05 th 1 Decrease pruning if eval is improving. Take 5. Lower curve of previous test by 1.
14-08-13 pe decrease_improving diff
LLR: -2.95 (-2.94,2.94)
Total: 11566 W: 2332 L: 2407 D: 6827
sprt @ 15+0.05 th 1 Decrease pruning if eval is improving. Try about half of take 2
14-08-13 ra Bonus_A diff
LLR: 0.02 (-2.94,2.94)
Total: 48 W: 20 L: 19 D: 9
sprt @ 15+0.05 th 1 Change depth based history update. Take 1
14-08-13 pe decrease_improving diff
LLR: -2.95 (-2.94,2.94)
Total: 17185 W: 3482 L: 3541 D: 10162
sprt @ 15+0.05 th 1 Decrease pruning if eval is improving. Take 3
14-08-13 ra master diff
ELO: 6.81 +-75.4 (95%) LOS: 57.1%
Total: 51 W: 16 L: 15 D: 20
1000 @ 15+0.05 th 1 Quick test for crashes.
13-08-13 pe decrease_improving diff
LLR: -2.95 (-2.94,2.94)
Total: 16045 W: 2906 L: 2928 D: 10211
sprt @ 60+0.05 th 1 Long TC: Decrease pruning if eval is improving. Take 2
13-08-13 jh easymove diff
ELO: -2.10 +-2.9 (95%) LOS: 7.6%
Total: 20000 W: 3499 L: 3620 D: 12881
20000 @ 60+0.05 th 1 Determine value of current easy move implementation.
13-08-13 pe decrease_improving diff
LLR: 2.95 (-2.94,2.94)
Total: 18613 W: 3907 L: 3747 D: 10959
sprt @ 15+0.05 th 1 Decrease pruning if eval is improving. Take 2
13-08-13 ra TT_Store_A diff
ELO: 0.19 +-3.1 (95%) LOS: 54.9%
Total: 20000 W: 4067 L: 4056 D: 11877
20000 @ 15+0.05 th 1 Tweak TT replacement. A little simplification.
13-08-13 mc lmr diff
LLR: -2.95 (-2.94,2.94)
Total: 22140 W: 4562 L: 4607 D: 12971
sprt @ 15+0.05 th 1 Reduce less when improving: take 3
13-08-13 mc lmr diff
LLR: -2.96 (-2.94,2.94)
Total: 8915 W: 1830 L: 1913 D: 5172
sprt @ 15+0.05 th 1 Search deeper along variations that are improving: take 2
06-08-13 tv counter_cut diff
LLR: -2.97 (-2.94,2.94)
Total: 26454 W: 4782 L: 4752 D: 16920
sprt @ 60+0.05 th 1 Long TC for Gary: Reduce countermoves less in cut nodes