Stockfish Testing Queue

Finished - 40825 tests

15-08-09 tvi RRQ diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 13889 W: 2588 L: 2657 D: 8644
sprt @ 15+0.05 th 1 Test Rocky640 suggestion before rebasing on tuned passed pawns
15-08-08 Fis TTreplace2 diff
LLR: 2.96 (-2.94,2.94) [0.00,4.00]
Total: 45258 W: 8595 L: 8279 D: 28384
sprt @ 15+0.05 th 1 A different way of weighing depth and generation to calculate a TT entry's replace value. See commit notes. It's a fix for this https://github.com/official-stockfish/Stockfish/pull/390 revert. If it should not be tested as a tuning please let me know and I'll reschedule. STC 2MB
15-08-09 Voy QuietSort diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 10287 W: 1886 L: 1964 D: 6437
sprt @ 15+0.05 th 1 New approach to sorting quiet moves.
15-08-09 Voy QuietStats diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 7723 W: 1424 L: 1509 D: 4790
sprt @ 15+0.05 th 1 New way of scoring quiet moves...
15-08-07 cru MG_EV_values diff
26582/20000 iterations
42000/40000 games played
40000 @ 15+0.05 th 1 Further tuning to ThreatenedByHangingPawn and KingAttackWeights
15-08-08 Dis passed_pawns_tuned_fina diff
LLR: 2.95 (-2.94,2.94) [0.00,4.00]
Total: 13620 W: 2216 L: 2023 D: 9381
sprt @ 60+0.05 th 1 Testing final tuned values for mid and end game passed pawns. (LTC and little fix thanks to Rocky and mstembera)
15-08-08 jos probcut diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 8436 W: 1549 L: 1632 D: 5255
sprt @ 15+0.05 th 1 Slower increase of margin. Take 3.
15-08-08 Dis passed_pawns_tuned_fina diff
LLR: 2.96 (-2.94,2.94) [0.00,4.00]
Total: 22691 W: 4468 L: 4228 D: 13995
sprt @ 15+0.05 th 1 Testing final tuned values for mid and end game passed pawns.
15-08-08 jos probcut diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 2895 W: 525 L: 624 D: 1746
sprt @ 15+0.05 th 1 Add probcut_margin(depth). Take 2.
15-08-08 Fis stableReply diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 17412 W: 3226 L: 3286 D: 10900
sprt @ 15+0.05 th 1 See commit notes.
15-08-08 Dis age_out_PV_tuned diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 14784 W: 2752 L: 2858 D: 9174
sprt @ 15+0.05 th 1 Testing the tuned age out PV parameter.
15-08-07 Voy HistoryStat diff
LLR: -2.94 (-2.94,2.94) [-1.50,4.50]
Total: 9882 W: 1840 L: 1919 D: 6123
sprt @ 15+0.05 th 1 Apply prior move refutation penalty for history stats as well
15-08-07 Voy ReductionTweak diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 9131 W: 1670 L: 1751 D: 5710
sprt @ 15+0.05 th 1 This is a sensitive parameter set at 1.5. 1.45 failed yellow in a past test. Try 1.475
15-08-04 Dis passed_pawns_tune diff
53532/20000 iterations
78753/80000 games played
80000 @ 30+0.05 th 1 Final SPSA test for passed pawns tuning, using the current values for mid and end game. Low priority.
15-08-07 Dis age_out_PV_tune diff
18269/20000 iterations
34606/40000 games played
40000 @ 30+0.05 th 1 Tuning the age out PV parameter.
15-08-07 cru MG_EV_values diff
19445/20000 iterations
39862/40000 games played
40000 @ 15+0.05 th 1 More tuning: ThreatenedByPawn
15-08-07 SC capture_tuned diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 7313 W: 1332 L: 1426 D: 4555
sprt @ 15+0.05 th 1 Another attempt on using piece value dependent rank penalty. This time values are chosen such that SEE is best approximated on my personal bench.
15-08-07 Voy RefutationBonusAdj diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 36277 W: 6872 L: 6902 D: 22503
sprt @ 15+0.05 th 1 Tweak Refutation Bonus
15-08-07 tvi RRQ diff
LLR: -2.94 (-2.94,2.94) [-1.50,4.50]
Total: 13645 W: 2546 L: 2615 D: 8484
sprt @ 15+0.05 th 1 RRQ bonus 8
15-08-07 tvi RRQ diff
LLR: -0.11 (-2.94,2.94) [-1.50,4.50]
Total: 29245 W: 5605 L: 5531 D: 18109
sprt @ 15+0.05 th 1 RRQ take 2
15-08-07 tvi RRQ diff
LLR: 3.52 (-2.94,2.94) [-1.50,4.50]
Total: 22413 W: 4327 L: 4142 D: 13944
sprt @ 15+0.05 th 1 RRQ bonus 9
15-08-06 Roc KingOnPawnV2 diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 76993 W: 9359 L: 9316 D: 58318
sprt @ 60+0.05 th 1 LTC
15-08-07 tvi RRQ diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 10521 W: 1968 L: 2046 D: 6507
sprt @ 15+0.05 th 1 Compensate passed pawn rank in RRQ ending
15-08-06 Voy StatsBonus diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 9688 W: 1748 L: 1828 D: 6112
sprt @ 15+0.05 th 1 Take 2...
15-08-07 Voy SE diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 2368 W: 406 L: 506 D: 1456
sprt @ 15+0.05 th 1 An additional condition for singular extension
15-08-06 Dis age_out_PV_tweak diff
ELO: -0.24 +-4.2 (95%) LOS: 45.5%
Total: 10000 W: 1897 L: 1904 D: 6199
10000 @ 15+0.05 th 1 Small test to see if tweaking age out PV parameter is a good idea, based on local testing (See https://imgur.com/TAj37vY)
15-08-06 cru MG_EV_values diff
9897/10000 iterations
20000/20000 games played
20000 @ 15+0.05 th 1 tune penalties for enemy's safe checks
15-08-06 Voy StatsBonus diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 21862 W: 4029 L: 4077 D: 13756
sprt @ 15+0.05 th 1 Add a third level. (fixed bench)
15-08-06 Roc KingOnPawnV2 diff
LLR: 4.06 (-2.94,2.94) [0.00,4.00]
Total: 86645 W: 13337 L: 12873 D: 60435
sprt @ 15+0.05 th 1 Same STC yellow [0,4] but using 8moves_v3.pgn "to steer toward endgame"
15-08-06 Fis excludedPawns diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 20466 W: 3756 L: 3842 D: 12868
sprt @ 15+0.05 th 1 Tune to consider pawns blocked by opponents pawns only. Final try.
15-08-05 Voy SE-Tweak diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 66144 W: 10363 L: 10314 D: 45467
sprt @ 60+0.05 th 1 LTC: Last shot for this SE parameter.
15-08-05 mco no_one_pawn diff
LLR: -2.95 (-2.94,2.94) [-3.00,1.00]
Total: 49370 W: 9125 L: 9369 D: 30876
sprt @ 15+0.05 th 1 Retire SCALE_FACTOR_ONEPAWN
15-08-04 Roc LogarithmicMob diff
ELO: -2.79 +-2.7 (95%) LOS: 2.0%
Total: 25000 W: 4682 L: 4883 D: 15435
25000 @ 15+0.05 th 1 Using a logarithmic model for each piece mobility. It might help simplify tuning. Let's try it first before any tuning.
15-08-04 Voy SE-Tweak diff
LLR: 2.91 (-2.94,2.94) [0.00,4.00]
Total: 129422 W: 24909 L: 24294 D: 80219
sprt @ 15+0.05 th 1 Last shot for this SE parameter.
15-08-05 Voy SeeValues diff
LLR: -2.99 (-2.94,2.94) [0.00,4.00]
Total: 12838 W: 2389 L: 2504 D: 7945
sprt @ 15+0.05 th 1 SEE values: Average of MG/EG piece value.
15-08-04 Roc LowMobBlockers diff
LLR: -3.58 (-2.94,2.94) [-1.50,4.50]
Total: 12078 W: 2239 L: 2335 D: 7504
sprt @ 15+0.05 th 1 Take 2
15-08-04 mco pop_lsb diff
LLR: -3.39 (-2.94,2.94) [-3.00,1.00]
Total: 50399 W: 9334 L: 9604 D: 31461
sprt @ 15+0.05 th 1 Retire piece lists
15-08-04 mco pop_lsb diff
ELO: -1.93 +-2.9 (95%) LOS: 9.3%
Total: 22000 W: 4192 L: 4314 D: 13494
20000 @ 15+0.05 th 1 Retire piece lists: measure ELO difference
15-08-04 SC capture_tuned diff
LLR: -3.26 (-2.94,2.94) [0.00,5.00]
Total: 6012 W: 1084 L: 1197 D: 3731
sprt @ 15+0.05 th 1 Previous version was 0 for pawns and high for kings. Now 0 for pawns, high for queens.
15-08-04 SC capture_tuned diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 7541 W: 1362 L: 1455 D: 4724
sprt @ 15+0.05 th 1 No rank penalty for pawn captures, high for queens.
15-08-03 lbr KingOnPawnV2 diff
LLR: -2.97 (-2.94,2.94) [0.00,4.00]
Total: 156211 W: 29758 L: 29365 D: 97088
sprt @ 15+0.05 th 1 Values from local tuning for advanced King supporting pawns. Retest with correct bounds *before* LTC
15-08-03 lbr SingularExtTweak diff
LLR: -3.14 (-2.94,2.94) [0.00,4.00]
Total: 87740 W: 16694 L: 16552 D: 54494
sprt @ 15+0.05 th 1 Take 2. Reschedule with correct bounds at STC.
15-08-04 Roc LowMobBlockers diff
LLR: -3.09 (-2.94,2.94) [-1.50,4.50]
Total: 9078 W: 1641 L: 1727 D: 5710
sprt @ 15+0.05 th 1 Reduce piece mobility when a low mobility piece is in the way.
15-08-03 jos passed_pawn_tweak2 diff
LLR: -3.07 (-2.94,2.94) [-1.50,4.50]
Total: 5526 W: 1000 L: 1095 D: 3431
sprt @ 15+0.05 th 1 Reworked defended and unsafe squares.
15-08-03 Voy RemoveSE diff
LLR: -2.95 (-2.94,2.94) [-3.00,1.00]
Total: 1340 W: 205 L: 369 D: 766
sprt @ 15+0.05 th 1 See if we are able to remove singular extension.
15-08-03 SC capture_promotion diff
LLR: -2.94 (-2.94,2.94) [0.00,5.00]
Total: 15489 W: 2876 L: 2934 D: 9679
sprt @ 15+0.05 th 1 Score promotions with capture over all other captures.
15-08-03 jos passed_pawn_tweak diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 19065 W: 3508 L: 3563 D: 11994
sprt @ 15+0.05 th 1 Assign a small bonus if the passer is defended. Take 2.
15-08-03 jos passed_pawn_tweak diff
LLR: -3.11 (-2.94,2.94) [-1.50,4.50]
Total: 22511 W: 4300 L: 4351 D: 13860
sprt @ 15+0.05 th 1 Does it make sense to include the pawn square in passer eval? Take 1.
15-08-03 Roc PasserOverprotection diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 7353 W: 1356 L: 1442 D: 4555
sprt @ 15+0.05 th 1 Take 2: Only for adv passer, and S(5, 5) Bench 9025634
15-08-02 lbr mobility diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 65720 W: 12394 L: 12321 D: 41005
sprt @ 15+0.05 th 1 Reduce mobility weight by 2.5%