Stockfish Testing Queue

Finished - 45441 tests

16-08-22 luc prob_square_dist^^^^ diff
LLR: -2.94 (-2.94,2.94) [0.00,5.00]
Total: 15092 W: 2768 L: 2828 D: 9496
sprt @ 10+0.1 th 1 Probabilistic square distance: take 2
16-08-22 Fis pawnAsymmetry diff
LLR: -2.98 (-2.94,2.94) [0.00,5.00]
Total: 16536 W: 3058 L: 3113 D: 10365
sprt @ 10+0.1 th 1 Change how we calculate asymmetry metric for pawns.
16-08-22 Elb pawn_dist diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 9126 W: 1625 L: 1751 D: 5750
sprt @ 10+0.1 th 1 Use pawn distance in opposite bishop scale factor calculation
16-08-22 pb0 null_move_hist diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 6892 W: 1231 L: 1327 D: 4334
sprt @ 10+0.1 th 1 Dedicated history for moves following a null-move
16-08-21 lbr tuned diff
LLR: -2.95 (-2.94,2.94) [-3.00,1.00]
Total: 81831 W: 11682 L: 11948 D: 58201
sprt @ 10+0.1 th 1 take 2
16-08-22 Voy rpCapture diff
LLR: 2.96 (-2.94,2.94) [0.00,5.00]
Total: 68191 W: 12872 L: 12449 D: 42870
sprt @ 10+0.1 th 1 Take 2
16-08-22 cib recursive_iid diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 36967 W: 6960 L: 6924 D: 23083
sprt @ 10+0.1 th 1 Allow Recursive IID.
16-08-22 Roc DrawHeuristic diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 3883 W: 513 L: 619 D: 2751
sprt @ 10+0.1 th 1 Experiment: return a draw score earlier in simple KBP, KNP or KP endings.
16-08-22 cru sn22_evaluate_space_cap diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 65416 W: 11955 L: 11887 D: 41574
sprt @ 10+0.1 th 1 [Try2: now going the other direction, increasing value by 500.] S. Nicolet idea 22: expand space area (changing the cap if necessary), or lower (by steps of 500) the material threshold for which evaluate_space() is called.
16-08-21 cru sn22_evaluate_space_cap diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 22776 W: 4124 L: 4203 D: 14449
sprt @ 10+0.1 th 1 S. Nicolet idea 22: expand space area (changing the cap if necessary), or lower (by steps of 500) the material threshold for which evaluate_space() is called.
16-08-21 luc prob_square_dist diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 11471 W: 2062 L: 2138 D: 7271
sprt @ 10+0.1 th 1 Probabilistic square distance: take 1
16-08-21 luc prob_square_dist diff
ELO: 2.08 +-4.1 (95%) LOS: 83.9%
Total: 10000 W: 1863 L: 1803 D: 6334
10000 @ 10+0.1 th 1 Sanity check: squares are in average farther apart when fewer paths exist (use for passed pawns)
16-08-21 Voy rpCapture diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 23601 W: 4332 L: 4356 D: 14913
sprt @ 10+0.1 th 1 Apply prior quiet-move penalty refutation on captures
16-08-21 SC ttValue diff
LLR: -2.95 (-2.94,2.94) [-3.00,1.00]
Total: 2337 W: 373 L: 536 D: 1428
sprt @ 10+0.1 th 1 Really rough attempt on non storing eval, more for checking that I understood it correctly. Low throughput.
16-08-21 luc repetition_value diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 15324 W: 2712 L: 2772 D: 9840
sprt @ 10+0.1 th 1 Try out Uri's idea: minimum non-zero value to help sort draws by repetition. (I'll leave it with low priority for anyone interested.)
16-08-21 SC smoothScaleFactor diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 51384 W: 9625 L: 9603 D: 32156
sprt @ 10+0.1 th 1 Try a simpler formula for scaling opposite bishops scale factor
16-08-21 Voy iidt diff
LLR: -2.94 (-2.94,2.94) [0.00,4.00]
Total: 29919 W: 5516 L: 5569 D: 18834
sprt @ 10+0.1 th 1 Take 2
16-08-21 SC pawnWeaknesses diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 7314 W: 1342 L: 1436 D: 4536
sprt @ 10+0.1 th 1 Take 3 and call it a day.
16-08-21 SC pawnWeaknesses diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 8024 W: 1487 L: 1578 D: 4959
sprt @ 10+0.1 th 1 Take 2.
16-08-21 SC pawnWeaknesses diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 29003 W: 5348 L: 5348 D: 18307
sprt @ 10+0.1 th 1 Quadratic penalty for multiple pawn weaknesses.
16-08-21 lbr tuned diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 13807 W: 1974 L: 2045 D: 9788
sprt @ 10+0.1 th 1 tuned
16-08-21 Elb lmr_captweak diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 5240 W: 946 L: 1050 D: 3244
sprt @ 10+0.1 th 1 Take 2
16-08-21 SC phalanxInitiative diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 17616 W: 3255 L: 3304 D: 11057
sprt @ 10+0.1 th 1 Phalanx does not help to convert. Are opponent's phalanxes a weakness?
16-08-21 cru timeman_tuning3 diff
56/30000 iterations
113/60000 games played
60000 @ 10+0.1 th 1 Time parameter tuning starting from deliberately overshot values in direction of previous yellow STC test.
16-08-20 lbr tune diff
29779/40000 iterations
60692/80000 games played
80000 @ 20+0.2 th 1 tune relative king distance scoring of passed pawns
16-08-20 mco oneply diff
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 95643 W: 17728 L: 17737 D: 60178
sprt @ 10+0.1 th 1 Regression test for ONE_PLY value independent
16-08-21 tvi SE diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 44610 W: 8206 L: 8209 D: 28195
sprt @ 10+0.1 th 1 Take 4
16-08-21 Voy iidt diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 42021 W: 7791 L: 7802 D: 26428
sprt @ 10+0.1 th 1 IID tweak
16-08-21 luc doubled_levers diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 32288 W: 6042 L: 6087 D: 20159
sprt @ 10+0.1 th 1 Most basic take 3 and call it a day...
16-08-21 SC phalanxInitiative diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 7018 W: 1260 L: 1356 D: 4402
sprt @ 10+0.1 th 1 Take 3. Pawn count compensation.
16-08-21 SC phalanxInitiative diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 2996 W: 512 L: 625 D: 1859
sprt @ 10+0.1 th 1 Take 2. Larger bonus.
16-08-21 mco fractional_lmr diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 7477 W: 1376 L: 1470 D: 4631
sprt @ 10+0.1 th 1 Fractional LMR: take 2
16-08-21 mco fractional_lmr^ diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 8702 W: 1570 L: 1658 D: 5474
sprt @ 10+0.1 th 1 Fractional LMR: take 3
16-08-18 lbr master diff
ELO: 52.59 +-1.7 (95%) LOS: 100.0%
Total: 40000 W: 8422 L: 2413 D: 29165
40000 @ 60+0.6 th 1 Regression test until e3af4921
16-08-21 cib tthit_iid2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 23275 W: 4348 L: 4372 D: 14555
sprt @ 10+0.1 th 1 Take 2: IID for missing TT Entry.
16-08-20 lbr passed_pawns diff
LLR: -2.95 (-2.94,2.94) [-3.00,1.00]
Total: 12842 W: 1686 L: 1852 D: 9304
sprt @ 60+0.6 th 1 LTC for sg (empty framework): Simplify by using additional evaluation also for passed pawns on rank 2 and 3.
16-08-20 SC LMR_clean diff
LLR: -2.95 (-2.94,2.94) [-3.00,1.00]
Total: 29479 W: 5481 L: 5691 D: 18307
sprt @ 10+0.1 th 1 Reorganiza a little bit LMR step after last LMR_captures.
16-08-21 luc smooth_initiative diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 19444 W: 3518 L: 3560 D: 12366
sprt @ 10+0.1 th 1 Apply initiative "smoothly", take 2 (with original capping)
16-08-21 lbr pawn diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 11979 W: 2166 L: 2240 D: 7573
sprt @ 10+0.1 th 1 Reduce blocked connected pawn bonus
16-08-20 cru timeman_tuned_2 diff
LLR: -2.94 (-2.94,2.94) [0.00,4.00]
Total: 13805 W: 2489 L: 2598 D: 8718
sprt @ 10+0.1 th 1 Another timeman attempt.
16-08-20 luc doubled_levers diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 31677 W: 5877 L: 5865 D: 19935
sprt @ 10+0.1 th 1 Take 2 (more radical)
16-08-20 SC phalanxInitiative diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 18790 W: 3522 L: 3566 D: 11702
sprt @ 10+0.1 th 1 Phalanx in initiative. Take 1.
16-08-20 mco fractional_lmr diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 6639 W: 1208 L: 1305 D: 4126
sprt @ 10+0.1 th 1 Use fracional LMR: take 1 (probably LMR is now too much reduced due to rounding up and an increase in LMR coefficents will be needed)
16-08-20 Voy rs diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 36210 W: 6743 L: 6774 D: 22693
sprt @ 10+0.1 th 1 Remove history "correction factor"
16-08-20 tvi SE diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 34445 W: 6351 L: 6389 D: 21705
sprt @ 10+0.1 th 1 Take 3
16-08-20 luc smooth_initiative diff
LLR: -2.99 (-2.94,2.94) [0.00,5.00]
Total: 12724 W: 2332 L: 2404 D: 7988
sprt @ 10+0.1 th 1 Apply initiative "smoothly", take 1 (without capping): see commit notes
16-08-19 Elb st_det diff
ELO: 0.07 +-2.6 (95%) LOS: 52.1%
Total: 20000 W: 2948 L: 2944 D: 14108
20000 @ 10+0.1 th 1 A quick check if it is possible to improve upon test http://tests.stockfishchess.org/tests/view/57b233d30ebc591c761f6405 for bishop endgames
16-08-20 Roc PPKingPos diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 10754 W: 1903 L: 1983 D: 6868
sprt @ 10+0.1 th 1 Passed pawn adjustment according to king position.
16-08-20 Roc PPKingPos2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 6198 W: 1134 L: 1233 D: 3831
sprt @ 10+0.1 th 1 Take 2: According to opponent king position.
16-08-18 sg passed_pawns diff
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 138560 W: 26235 L: 26322 D: 86003
sprt @ 10+0.1 th 1 Simplify by using additional evaluation also for passed pawns on rank 2 and 3.