Stockfish Testing Queue

Finished - 40825 tests

15-08-16 lbr test diff
LLR: 3.11 (-2.94,2.94) [0.00,4.00]
Total: 71858 W: 13789 L: 13369 D: 44700
sprt @ 15+0.05 th 1 Less futility + more SE (10442788): - stop futility pruning one depth lower - start singular extension one depth lower - compensate by retuning futility_margin() => Hopefully this can be both an elo gain and an improvement in tactics.
15-08-16 tvi RRQ2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 25689 W: 4859 L: 4872 D: 15958
sprt @ 15+0.05 th 1 Try a simple RRQ passed bonus
15-08-16 Voy LMR_Tweak diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 815 W: 112 L: 216 D: 487
sprt @ 15+0.05 th 1 LMR Tweak
15-08-16 jos null_tweak diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 12155 W: 2268 L: 2383 D: 7504
sprt @ 15+0.05 th 1 Raise null-move depth limit by one ply.
15-08-15 Qua qsearch_bonus diff
LLR: 2.96 (-2.94,2.94) [-1.50,4.50]
Total: 82097 W: 15678 L: 15355 D: 51064
sprt @ 15+0.05 th 1 Small Bonus in qsearch if many good moves
15-08-16 lbr test diff
18710/25000 iterations
36846/55000 games played
55000 @ 15+0.05 th 1 quick tune
15-08-15 Voy ImproveHStats diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 22210 W: 4201 L: 4248 D: 13761
sprt @ 15+0.05 th 1 Getting the most out of our CMH array...
15-08-16 Voy ImproveHStats diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 5779 W: 1028 L: 1118 D: 3633
sprt @ 15+0.05 th 1 Take 2.
15-08-15 Roc SupportedKnight diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 22022 W: 4164 L: 4211 D: 13647
sprt @ 15+0.05 th 1 Bonus only for the Knight.
15-08-15 Fis TTreplace3 diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 9565 W: 1753 L: 1877 D: 5935
sprt @ 15+0.05 th 1 Tweak TT replacement policy to use depth divided by age. 2MB STC
15-08-15 Voy HistoryReduction diff
LLR: -1.63 (-2.94,2.94) [-1.50,4.50]
Total: 670 W: 98 L: 154 D: 418
sprt @ 15+0.05 th 1 Make reductions more dependent upon history stats.
15-08-15 Voy EG-Reduction diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 3767 W: 664 L: 760 D: 2343
sprt @ 15+0.05 th 1 Take 2: On improve formula..
15-08-15 Fis TTsaveTweak diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 9366 W: 1709 L: 1834 D: 5823
sprt @ 15+0.05 th 1 Tune a constant in TTEntry::save() 2MB STC
15-08-14 Dis passed_pawns_owned_bloc diff
38493/40000 iterations
58540/80000 games played
80000 @ 30+0.05 th 1 Tuning the "passed pawn when it is blocked by our piece" values (+Little fix x 2)
15-08-15 Voy EG-Reduction diff
LLR: -2.96 (-2.94,2.94) [0.00,6.00]
Total: 7806 W: 1199 L: 1265 D: 5342
sprt @ 60+0.05 th 1 LTC: Take 4: Improve formula.
15-08-14 Voy EG-Reduction diff
LLR: 2.95 (-2.94,2.94) [-1.50,4.50]
Total: 6254 W: 1245 L: 1123 D: 3886
sprt @ 15+0.05 th 1 Take 4: Improve formula.
15-08-14 Roc SupportedPieces diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 4408 W: 817 L: 912 D: 2679
sprt @ 15+0.05 th 1 Take 3: S(10,0), code simplified and bench fixed
15-08-14 Voy EG-Reduction diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 12145 W: 2292 L: 2365 D: 7488
sprt @ 15+0.05 th 1 Take 3: This time reduce more when searching at end game.
15-08-14 mbo qmob_and_ks diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 3800 W: 671 L: 767 D: 2362
sprt @ 15+0.05 th 1 Increase mobility for a queen which attacks the king, reduce when not attacking.
15-08-14 Roc SupportedPieces diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 17049 W: 3196 L: 3256 D: 10597
sprt @ 15+0.05 th 1 Prepare better squares for Minors
15-08-14 Roc SupportedPieces diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 7250 W: 1315 L: 1401 D: 4534
sprt @ 15+0.05 th 1 Take 2
15-08-13 lbr noCentralBind^ diff
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 60866 W: 11509 L: 11458 D: 37899
sprt @ 15+0.05 th 1 Remove CenterBind
15-08-08 lbr PawnPushSimpl diff
LLR: 5.30 (-2.94,2.94) [-3.00,1.00]
Total: 204194 W: 31698 L: 31734 D: 140762
sprt @ 60+0.05 th 1 LTC for Rocky640: Verify if the PawnSafePush bonus is still useful. Passed STC(-3,1) but LTC was never scheduled. Why not?
15-08-13 Dis adjust_delta_wScore diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 27746 W: 5232 L: 5264 D: 17250
sprt @ 15+0.05 th 1 Adjusting delta with score, take 2 (higher delta)
15-08-14 Voy EG-Reduction diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 15098 W: 2800 L: 2866 D: 9432
sprt @ 15+0.05 th 1 Take 2...
15-08-14 Voy ttStats diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 10581 W: 1978 L: 2056 D: 6547
sprt @ 15+0.05 th 1 Don't update stat bonuses for TT moves...as this may inflate bonuses.
15-08-13 Voy EG-Reduction diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 7045 W: 1294 L: 1381 D: 4370
sprt @ 15+0.05 th 1 Less reduction when search is exploring end game phase...
15-08-13 lbr noCentralBind diff
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 35227 W: 6715 L: 6619 D: 21893
sprt @ 15+0.05 th 1 Compensate in PSQT
15-08-13 Voy UnderPromotions diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 22297 W: 4250 L: 4296 D: 13751
sprt @ 15+0.05 th 1 Place under promotions at the bottom of our move list. If a Queen promotion fails to cause a cutoff or improve our position...it is very likely that under promotions will as well.
15-08-13 Dis adjust_delta_wScore diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 45065 W: 8549 L: 8535 D: 27981
sprt @ 15+0.05 th 1 Adujsting the delta of the aspiration window with the score, getting a little boost (wider window) when close to 0 value of evaluation score. (+Little fix, mod. values)
15-08-13 Voy LMR-Tweak diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 29935 W: 5677 L: 5729 D: 18529
sprt @ 15+0.05 th 1 LMR Tweak
15-08-12 jos asym_lmr diff
LLR: -2.94 (-2.94,2.94) [-1.50,4.50]
Total: 9135 W: 1703 L: 1784 D: 5648
sprt @ 15+0.05 th 1 Less LMR every 5 plies.
15-08-12 tvi MBP diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 35572 W: 6709 L: 6742 D: 22121
sprt @ 15+0.05 th 1 Test locally tuned MinorBehindPawn
15-08-10 jos latejoin_tweak2 diff
ELO: 2.19 +-3.6 (95%) LOS: 88.0%
Total: 10000 W: 1467 L: 1404 D: 7129
10000 @ 30+0.05 th 7 Simplification: latejoin to the split point with the highest depth. Compare with http://tests.stockfishchess.org/tests/view/55a394680ebc590abbe1ba42
15-08-11 cru tuned_param_mg_ev diff
LLR: -2.94 (-2.94,2.94) [0.00,4.00]
Total: 45424 W: 8826 L: 8821 D: 27777
sprt @ 15+0.05 th 1 Testing results of sequentially tuning the following parameters: ThreatenedByPawn, ThreatenedByHangingPawn, KingAttackWeights, Penalties for enemy's safe checks, and Assorted bonuses and penalties used by evaluation.
15-08-11 Fis TTreplace2 diff
LLR: 2.96 (-2.94,2.94) [-4.00,0.00]
Total: 34764 W: 6565 L: 6529 D: 21670
sprt @ 15+0.05 th 1 No regression w/ low hash pressure. 16MB STC
15-08-11 Voy QuietMoveSort diff
LLR: -3.05 (-2.94,2.94) [-1.50,4.50]
Total: 9054 W: 1663 L: 1748 D: 5643
sprt @ 15+0.05 th 1 Take 2...
15-08-11 jos asym_lmr diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 2595 W: 481 L: 581 D: 1533
sprt @ 15+0.05 th 1 Reduce less at odd plies.
15-08-11 Voy QuietMoveSort diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 2865 W: 486 L: 584 D: 1795
sprt @ 15+0.05 th 1 New way of sorting quiet moves...
15-08-09 Fis TTreplace2 diff
LLR: 2.95 (-2.94,2.94) [0.00,4.00]
Total: 56401 W: 8809 L: 8489 D: 39103
sprt @ 60+0.05 th 1 8MB LTC (If this passes I'll also schedule a no regression test with low hash pressure.)
15-08-10 Voy QuietMoveWeights diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 32359 W: 4972 L: 5026 D: 22361
sprt @ 60+0.05 th 1 LTC: Quiet weights tweak.
15-08-11 lbr combo diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 21380 W: 3293 L: 3334 D: 14753
sprt @ 60+0.05 th 1 LTC: combo of a couple of tuning patches that seem positive, but not strong enough to pass individually. Rebase on current master, just in case TT revert has any effect on SE.
15-08-09 cru MG_EV_values diff
50190/40000 iterations
89894/80000 games played
80000 @ 15+0.05 th 1 Continuing to tune various parameters, adjusted search ranges based on advice from Rocky640.
15-08-09 jos combo diff
LLR: -3.23 (-2.94,2.94) [0.00,4.00]
Total: 75165 W: 14334 L: 14241 D: 46590
sprt @ 15+0.05 th 1 Combo of 2 tuning patches. New queen values in the material imbalance tables, which almost passed LTC, and a lower space threshold.
15-08-10 tvi RRQ2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 23247 W: 4367 L: 4391 D: 14489
sprt @ 15+0.05 th 1 Test RRQ rebased passed pawn tuning, escore*80/64 <=> escore + escore/4
15-08-09 Roc EndgameTweaks diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 59260 W: 8931 L: 8901 D: 41428
sprt @ 15+0.05 th 1 Will probably be 0 elo, but if it works, will test separately. With 8moves_v3.pgn
15-08-09 lbr combo diff
LLR: 2.95 (-2.94,2.94) [0.00,5.00]
Total: 61742 W: 11679 L: 11284 D: 38779
sprt @ 15+0.05 th 1 combo of a couple of tuning patches that seem positive, but not strong enough to pass individually.
15-08-10 Voy QuietMoveWeights diff
LLR: 2.96 (-2.94,2.94) [0.00,4.00]
Total: 29449 W: 5680 L: 5418 D: 18351
sprt @ 15+0.05 th 1 Quiet weights tweak.
15-08-09 Voy UpdateStats diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 29689 W: 5667 L: 5694 D: 18328
sprt @ 15+0.05 th 1 New way of updating stat's values
15-08-09 jos probcut diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 2628 W: 471 L: 571 D: 1586
sprt @ 15+0.05 th 1 Final try.