Stockfish Testing Queue

Finished - 30472 tests

15-09-25 sg passed_pawns diff
LLR: 3.33 (-2.94,2.94) [0.00,5.00]
Total: 36889 W: 6805 L: 6507 D: 23577
sprt @ 15+0.05 th 1 try tuned values. The are only small changes, but lets see. Perhaps a further tuning session is needed.
15-09-25 Roc KnightMob diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 12603 W: 1891 L: 1965 D: 8747
sprt @ 60+0.05 th 1 Restore previous Knight Mobility definition which excluded mobile pawns. @ LTC
15-09-25 Voy cmhStatB diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 8680 W: 1537 L: 1625 D: 5518
sprt @ 15+0.05 th 1 Take 2...
15-09-25 SC multipv_search diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 15840 W: 2849 L: 2907 D: 10084
sprt @ 15+0.05 th 1 Search with multipv at low depth When we have plenty of time available, let us search in multipv. Maybe it help sorting moves. Scale multipv proportionally to sqrt of remaining time. Furiousfish ignored me for some reason.
15-09-25 Mys smp1 diff
LLR: -1.00 (-2.94,2.94) [0.00,5.00]
Total: 2400 W: 360 L: 392 D: 1648
sprt @ 15+0.05 th 7 Implement an 'option' splitpoint concept using depth. Verified the conditions are reached this time.
15-09-25 sg tune_passed_pawns diff
19658/20000 iterations
40000/40000 games played
40000 @ 30+0.05 th 1 Tune file based passed pawn bonus starting from my first values (i prefer not to break symmetry, so 8 values tuned)
15-09-25 mbo lazy_smp diff
ELO: -13.21 +-5.5 (95%) LOS: 0.0%
Total: 5000 W: 738 L: 928 D: 3334
5000 @ 15+0.05 th 3 Get another rough estimate for Lazy SMP. This version has some time management fixes. Previous run was aborted by "wrong bench".
15-09-23 Roc KnightMob diff
LLR: 2.95 (-2.94,2.94) [0.00,5.00]
Total: 74183 W: 14157 L: 13705 D: 46321
sprt @ 15+0.05 th 1 Restore previous Knight Mobility definition which excluded mobile pawns.
15-09-24 sni king_safety2 diff
LLR: -1.93 (-2.94,2.94) [0.00,5.00]
Total: 23067 W: 4226 L: 4209 D: 14632
sprt @ 15+0.05 th 1 Alter psqt table for centered king in endgame
15-09-24 SC imbalance_gp2 diff
LLR: -4.05 (-2.94,2.94) [0.00,5.00]
Total: 37303 W: 6876 L: 6887 D: 23540
sprt @ 15+0.05 th 1 Faster and tuned version.
15-09-25 sni second_push diff
LLR: -2.96 (-2.94,2.94) [-3.00,1.00]
Total: 3712 W: 636 L: 802 D: 2274
sprt @ 15+0.05 th 1 Suppress second push bonus
15-09-25 Voy cmhStatB diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 21901 W: 3915 L: 3947 D: 14039
sprt @ 15+0.05 th 1 Version B.
15-09-24 sg passed_pawns diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 33676 W: 6380 L: 6358 D: 20938
sprt @ 15+0.05 th 1 Take 1 struggles at LTC. So try double bonus. Take 3
15-09-23 sni king_separation2 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 45255 W: 7114 L: 7064 D: 31077
sprt @ 60+0.05 th 1 LTC: Try king separation bonus, take 3
15-09-24 IIv reductions_check diff
ELO: -25.69 +-5.8 (95%) LOS: 0.0%
Total: 5000 W: 719 L: 1088 D: 3193
5000 @ 15+0.05 th 1 A quick check in which direction the tuning is going. The used function is almost identical with log for mc>15, and it is the best quadratic fit of tuned values for mc>8.
15-09-24 Voy cmhStat diff
LLR: -3.09 (-2.94,2.94) [0.00,5.00]
Total: 15269 W: 2290 L: 2360 D: 10619
sprt @ 60+0.05 th 1 We meet again LTC...I shall seek vengeance for my fallen brothers. Attack!!!
15-09-24 mbo lazy_smp diff
ELO: -24.86 +-17.3 (95%) LOS: 0.2%
Total: 448 W: 49 L: 81 D: 318
5000 @ 15+0.05 th 3 Get another rough estimate for Lazy SMP. This version has some time management fixes.
15-09-24 IIv reductions_tune diff
9958/10000 iterations
20000/20000 games played
20000 @ 30+0.05 th 1 Tuning mc reductions, last try before conclusion, using (not too?) agressive ck values inspired by tuneDelta by Fisherman - it seems that first 10-20 moves should be more finely tuned, and the rest is slowly going from log to a linear function
15-09-24 mbo lazy_smp diff
ELO: -15.63 +-7.3 (95%) LOS: 0.0%
Total: 3002 W: 452 L: 587 D: 1963
5000 @ 15+0.05 th 3 Get a rough estimate for Lazy SMP. More comments on github.
15-09-24 Voy cmhStat diff
LLR: 2.95 (-2.94,2.94) [0.00,5.00]
Total: 17517 W: 3314 L: 3113 D: 11090
sprt @ 15+0.05 th 1 Update CMH and History stats differently.
15-09-24 tvi BadPawnSupport diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 19637 W: 3632 L: 3673 D: 12332
sprt @ 15+0.05 th 1 Penalty for passers with bad bishop support (thanks to snicolet for detecting coding error)
15-09-20 jos mobility3 diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 115321 W: 17726 L: 17534 D: 80061
sprt @ 60+0.05 th 1 LTC: New mobility tuning session, proposed by Rocky640. All endgame values (except knight) are noticeably increased.
15-09-21 SC imbalance_gp_tuning diff
19495/20000 iterations
38394/40000 games played
40000 @ 60+0.05 th 1 Tuning imbalance_gp after including ideas from Stephane. Since it failed LTC, I tune LTC at throughput 500. As suggested by Rocky640 reschedule with different settings (and less games).
15-09-23 Fis tuneDelta diff
20052/20000 iterations
40000/40000 games played
40000 @ 30+0.05 th 1 Tune new delta update with increment.
15-09-23 IIv reductions_tune diff
9916/10000 iterations
20000/20000 games played
20000 @ 30+0.05 th 1 Tuning mc reductions with refined ck values
15-09-17 Roc tune_search3 diff
LLR: 3.12 (-2.94,2.94) [0.00,4.00]
Total: 69395 W: 13201 L: 12791 D: 43403
sprt @ 15+0.05 th 1 Testing tuned auxiliary parameters after 43M games
15-09-23 Voy StatTweak diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 17308 W: 3113 L: 3165 D: 11030
sprt @ 15+0.05 th 1 Stat Update Tweak
15-09-23 sni king_separation2 diff
LLR: 2.95 (-2.94,2.94) [0.00,5.00]
Total: 22134 W: 4248 L: 4025 D: 13861
sprt @ 15+0.05 th 1 Try king separation bonus, take 3
15-09-23 sni stalemate5 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 6010 W: 1078 L: 1178 D: 3754
sprt @ 15+0.05 th 1 A strange idea: stalemate positions often result from capture(s) in the last few moves, but can often be avoided by the strong side if he/she doesn't capture the last defensive piece and play a quiet move instead. However, we don't consider quiet moves in qsearch, and a draw stalemate evaluation there may well be pessimistic for the strong side. See what happen if we score a stalemate as a "small win" for the strong side.
15-09-23 sni pawn_asymmetry3 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 17069 W: 3114 L: 3166 D: 10789
sprt @ 15+0.05 th 1 Try pawn asymmetry, take 4
15-09-22 sni pawn_asymmetry3 diff
LLR: -2.63 (-2.94,2.94) [0.00,5.00]
Total: 38906 W: 7178 L: 7122 D: 24606
sprt @ 15+0.05 th 1 Try pawn asymmetry, take 3
15-09-21 sg material4 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 62117 W: 11400 L: 11259 D: 39458
sprt @ 15+0.05 th 1 Endgame pawn scaling. Use also passed pawns. Lower weight=1. Take 3
15-09-22 sg opposite_bishops diff
LLR: -2.95 (-2.94,2.94) [-3.00,1.00]
Total: 26984 W: 4894 L: 5098 D: 16992
sprt @ 15+0.05 th 1 Simplification: replace opposite bishops scaling by term in new endgame scaling formula. Lower weight=9. Take 2
15-09-22 tvi BadPawnSupport diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 6810 W: 1207 L: 1303 D: 4300
sprt @ 15+0.05 th 1 Penalty for bishop with bad passer support, take 2
15-09-22 tvi BadPawnSupport diff
LLR: -3.08 (-2.94,2.94) [0.00,5.00]
Total: 31324 W: 5736 L: 5732 D: 19856
sprt @ 15+0.05 th 1 Penalty for bishop with bad passer support (restart after bad bench from JojoM)
15-09-22 Voy lmrLogic diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 15966 W: 2968 L: 3024 D: 9974
sprt @ 15+0.05 th 1 Take 2...
15-09-22 jos delta_tweak diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 17885 W: 3223 L: 3318 D: 11344
sprt @ 15+0.05 th 1 Increase delta faster at higher depths.
15-09-22 Voy lmrLogic diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 19606 W: 3595 L: 3636 D: 12375
sprt @ 15+0.05 th 1 Take 3
15-09-22 sni pawn_asymmetry3 diff
LLR: -2.40 (-2.94,2.94) [0.00,5.00]
Total: 14238 W: 2547 L: 2588 D: 9103
sprt @ 15+0.05 th 1 Try pawn asymmetry, take 2: asymmetry attacking bonus increases with number of pawns.
15-09-22 tvi BadPawnSupport diff
LLR: 0.67 (-2.94,2.94) [0.00,5.00]
Total: 1639 W: 304 L: 269 D: 1066
sprt @ 15+0.05 th 1 Penalty for bishop with bad passer support
15-09-21 sg opposite_bishops diff
LLR: -2.95 (-2.94,2.94) [-3.00,1.00]
Total: 39309 W: 7115 L: 7340 D: 24854
sprt @ 15+0.05 th 1 Simplification: replace opposite bishops scaling by term in new endgame scaling formula.
15-09-21 Fis deltaUpdate diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 22873 W: 4209 L: 4236 D: 14428
sprt @ 15+0.05 th 1 New update for delta.
15-09-21 Voy lmrLogic diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 13293 W: 1956 L: 2028 D: 9309
sprt @ 60+0.05 th 1 LTC: LMR Logic Tweak
15-09-21 sni blocked_pawns diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 14239 W: 2613 L: 2677 D: 8949
sprt @ 15+0.05 th 1 Try blocked pawns malus for initiative.
15-09-21 adp shelterweakness diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 6519 W: 1172 L: 1307 D: 4040
sprt @ 15+0.05 th 1 Shelter Weakness Take 2.
15-09-21 sni stalemate5 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 5860 W: 1013 L: 1113 D: 3734
sprt @ 15+0.05 th 1 Take 2: add the case where the king is blocked by its own pawns (Stefan Geschwentner)
15-09-20 SC imbalance_gp_tuning diff
11969/30000 iterations
23767/60000 games played
60000 @ 60+0.05 th 1 Tuning imbalance_gp after including ideas from Stephane. Since it failed LTC, I tune LTC at throughput 500. Bugfix (forgot to commit before pushing).
15-09-19 adp Razoring diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 68052 W: 12425 L: 12260 D: 43367
sprt @ 15+0.05 th 1 Razor Tuned Values
15-09-21 sni stalemate5 diff
LLR: -0.44 (-2.94,2.94) [0.00,5.00]
Total: 1137 W: 216 L: 230 D: 691
sprt @ 15+0.05 th 1 Fast stalemate detection: isolated king, and isolated king + pawn(s)
15-09-20 Roc Islands diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 14208 W: 2623 L: 2687 D: 8898
sprt @ 15+0.05 th 1 Take 1: S(10, 20)