Stockfish Testing Queue

Pending - 13 tests 0.0 hrs

18-08-14 sni game_phase4 diff
LLR: -1.16 (-2.94,2.94) [0.00,5.00]
Total: 8914 W: 1940 L: 1948 D: 5026
sprt @ 10+0.1 th 1 Use pawns in game phase: pawns/2 - 5
18-08-14 sni game_phase5 diff
LLR: -2.33 (-2.94,2.94) [0.00,5.00]
Total: 8896 W: 1860 L: 1920 D: 5116
sprt @ 10+0.1 th 1 Use pawns in game phase: pawns/2 - 7
18-08-14 sni game_phase4 diff
LLR: -1.19 (-2.94,2.94) [0.00,5.00]
Total: 12260 W: 2644 L: 2637 D: 6979
sprt @ 10+0.1 th 1 Pretend SF knows the future: shift by 3
18-08-14 sni bad_capture4 diff
LLR: -1.36 (-2.94,2.94) [0.00,4.00]
Total: 8831 W: 1918 L: 1959 D: 4954
sprt @ 10+0.1 th 1 Softer use of capture history for bad captures
18-08-14 sni bad_capture4 diff
LLR: -1.81 (-2.94,2.94) [0.00,4.00]
Total: 9103 W: 1929 L: 1994 D: 5180
sprt @ 10+0.1 th 1 Even softer use of capture history for bad captures
18-08-14 sni bad_capture4~1 diff
Pending... (0.4 hrs)
[0.00,5.00]
sprt @ 10+0.1 th 1 Take 3
18-08-14 sni bad_capture4 diff
Pending... (0.4 hrs)
[0.00,5.00]
sprt @ 10+0.1 th 1 Take 4
18-08-14 sg prune_draw_new diff
LLR: -2.60 (-2.94,2.94) [0.00,5.00]
Total: 19000 W: 4043 L: 4066 D: 10891
sprt @ 10+0.1 th 1 Check now not only for exact draws but drawish positions (abs(value) <= 10). Idea from snicolet.
18-08-14 sg prune_draw_new diff
LLR: -1.74 (-2.94,2.94) [0.00,5.00]
Total: 18872 W: 3098 L: 3096 D: 12678
sprt @ 60+0.6 th 1 LTC: Extend effect only to possible PV nodes.. Rebased to current master
18-08-14 sg prune_best_move diff
LLR: -1.95 (-2.94,2.94) [0.00,5.00]
Total: 14000 W: 3023 L: 3041 D: 7936
sprt @ 10+0.1 th 1 More general version based on an idea of SC. No move count pruning if best move > static eval. (Park with prio -1 to wait for my LTC)
18-08-14 sg lmr_best_move diff
LLR: -1.67 (-2.94,2.94) [0.00,5.00]
Total: 9000 W: 1932 L: 1962 D: 5106
sprt @ 10+0.1 th 1 Less reduction if value of best move < static eval (Park with prio -1)
18-08-14 sg lmr_best_move diff
LLR: -2.50 (-2.94,2.94) [0.00,5.00]
Total: 18200 W: 3930 L: 3952 D: 10318
sprt @ 10+0.1 th 1 Less reduction if value of best move < static eval and < draw.
18-08-14 SC pawnTune diff
127886/250000 iterations
284640/500000 games played
500000 @ 30+0.3 th 1 Wild guesswork: can we strongly lower the value of pawns and adapt all other pawn related stuff? Low throughput experiment.

Active - 60 machines 282 cores 1.30M nps (366.07M total nps) 634 games/minute

Machine Cores MNps System Version Running on Last updated
gvreuls
15
0.00 Linux 4.17.12-100.fc27.x86_64 67:2 mob_initiative6 seconds ago
ChessDBCN
8
0.00 Linux 3.10.0-862.9.1.el7.x86_64 66:2 mob_initiative6 seconds ago
ChessDBCN
8
0.00 Linux 3.10.0-862.9.1.el7.x86_64 66:2 mob_initiative6 seconds ago
mhoram
3
2.19 Linux 4.9.0-7-amd64 67:2 mob_initiative6 seconds ago
bcross
15
1.28 Windows 10 64 mob_initiative6 seconds ago
ChessDBCN
8
1.54 Linux 3.10.0-862.9.1.el7.x86_64 66:2 mob_initiative6 seconds ago
dew
2
0.94 Linux 4.12.12-gentoo 63 mob_initiative6 seconds ago
ctoks
15
1.28 Linux 4.13.0-41-generic 67:3 mob_initiative6 seconds ago
ChessDBCN
8
1.53 Linux 3.10.0-862.9.1.el7.x86_64 66:2 mob_initiative6 seconds ago
TueRens
3
1.39 Windows 7 67:2 mob_initiative6 seconds ago
ChessDBCN
8
1.55 Linux 3.10.0-862.9.1.el7.x86_64 66:2 mob_initiative6 seconds ago
robal
2
2.23 Windows 10 63 mob_initiative6 seconds ago
ChessDBCN
8
1.53 Linux 3.10.0-862.9.1.el7.x86_64 66:2 mob_initiative6 seconds ago
Zaplanc
3
1.51 Linux 4.4.0-43-Microsoft 67:2 mob_initiative6 seconds ago
dew
1
1.28 Linux 3.18.7-gentoo 63 mob_initiative6 seconds ago
ChessDBCN
8
1.54 Linux 3.10.0-862.9.1.el7.x86_64 66:2 mob_initiative6 seconds ago
ChessDBCN
8
1.53 Linux 3.10.0-862.9.1.el7.x86_64 66:2 mob_initiative6 seconds ago
ChessDBCN
8
1.41 Linux 3.10.0-862.9.1.el7.x86_64 66:2 mob_initiative6 seconds ago
ChessDBCN
8
1.50 Linux 3.10.0-862.9.1.el7.x86_64 66:2 mob_initiative6 seconds ago
ChessDBCN
8
1.53 Linux 3.10.0-862.9.1.el7.x86_64 66:2 mob_initiative6 seconds ago
drabel
1
1.79 Windows 10 63 mob_initiative6 seconds ago
ChessDBCN
8
1.53 Linux 3.10.0-862.9.1.el7.x86_64 66:2 mob_initiative6 seconds ago
ChessDBCN
8
1.40 Linux 3.10.0-862.9.1.el7.x86_64 66:2 mob_initiative6 seconds ago
ChessDBCN
8
1.53 Linux 3.10.0-862.9.1.el7.x86_64 66:2 mob_initiative6 seconds ago
brabos
4
1.79 Windows 10 63 mob_initiative6 seconds ago
ChessDBCN
8
1.65 Linux 3.10.0-862.9.1.el7.x86_64 66:2 mob_initiative6 seconds ago
ChessDBCN
8
1.51 Linux 3.10.0-862.9.1.el7.x86_64 66:2 mob_initiative6 seconds ago
ChessDBCN
8
1.53 Linux 3.10.0-862.9.1.el7.x86_64 66:2 mob_initiative6 seconds ago
ChessDBCN
8
1.53 Linux 3.10.0-862.9.1.el7.x86_64 66:2 mob_initiative6 seconds ago
JojoM
3
1.13 Linux 3.16.0-6-amd64 67:2 mob_initiative6 seconds ago
sqrt2
3
2.19 Linux 3.16.0-6-amd64 63 mob_initiative6 seconds ago
jbwiebe
3
2.06 Linux 4.9.0-7-amd64 64 mob_initiative6 seconds ago
Adrian.Schmidt123
4
2.62 Windows 10 63 mob_initiative6 14 minutes ago
horst.prack
1
1.16 Linux 4.17.0-2-amd64 63 mob_initiative6 seconds ago
JojoM
3
1.13 Linux 3.16.0-6-amd64 67:2 mob_initiative6 seconds ago
JojoM
3
1.12 Linux 3.16.0-6-amd64 67:2 mob_initiative6 seconds ago
choklas81
3
1.88 Windows 10 67:2 mob_initiative6 seconds ago
dew
4
1.34 Linux 4.8.17-hardened-r2 63 mob_initiative6 seconds ago
leszek
7
0.96 Windows 7 63 tune_king_shelter seconds ago
velislav
1
2.24 Linux 4.4.0-112-generic 63 tune_king_shelter 2 minutes ago
JojoM
2
1.73 Linux 4.4.0-17134-Microsoft 67:2 tune_king_shelter seconds ago
JojoM
3
1.12 Linux 3.16.0-6-amd64 67:2 tune_king_shelter seconds ago
mgrabiak
3
1.60 Windows 7 64 tune_king_shelter seconds ago
dew
1
1.14 Linux 4.9.76-gentoo-r1 63 tune_king_shelter seconds ago
JojoM
3
1.12 Linux 3.16.0-6-amd64 67:2 tune_king_shelter 2 minutes ago
ctoks
3
0.96 Linux 4.4.0-17134-Microsoft 67:2 ps_safety2 3 minutes ago
sunu
2
1.22 Linux 4.16.0-9.slh.1-aptosid-amd64 63 ps_safety2 seconds ago
nordlandia
3
1.16 Windows 10 64 ps_safety2 seconds ago
dew
1
1.14 Linux 4.0.5-gentoo 63 ps_safety2 seconds ago
mgrabiak
3
0.83 Windows 10 64 ps_safety2 seconds ago
dew
1
1.10 Linux 4.9.76-gentoo-r1 63 ps_safety2 5 minutes ago
Thanar
3
1.15 Windows 7 63 ps_safety2 seconds ago
dew
3
1.06 Linux 3.18.9-hardened 63 ps_safety2 seconds ago
gri
1
2.10 Linux 4.9.0-7-amd64 63 ps_safety2 seconds ago
dew
1
1.20 Linux 3.12.13-gentoo 63 ps_safety2 2 minutes ago
dew
1
1.11 Linux 4.9.76-gentoo-r1 63 ps_safety2 3 minutes ago
geoffeg
1
0.73 Linux 3.10.105 63 ps_safety2 10 minutes ago
dew
1
1.13 Linux 4.9.76-gentoo-r1 63 ps_safety2 2 minutes ago
mhoram
1
0.96 Linux 4.9.0-7-amd64 63 ps_safety2 2 minutes ago
velislav
2
1.55 Linux 3.10.0-123.8.1.el7.x86_64 63 master seconds ago
18-08-15 sni mob_initiative6 diff
LLR: -0.47 (-2.94,2.94) [0.00,4.00]
Total: 23962 W: 5168 L: 5101 D: 13693
sprt @ 10+0.1 th 1 Just the connection and the centrality (no mobility)
18-08-14 pro ps_safety2 diff
LLR: -0.64 (-2.94,2.94) [-3.00,1.00]
Total: 49334 W: 8046 L: 8157 D: 33131
sprt @ 60+0.6 th 1 LTC: -5, and +5 in safety initialization.
18-08-14 sg tune_king_shelter diff
LLR: -1.85 (-2.94,2.94) [0.00,4.00]
Total: 48721 W: 8030 L: 7972 D: 32719
sprt @ 60+0.6 th 1 LTC: Try from protonsprings simplification only the king shelter values as tuning.
18-08-14 sni master diff
ELO: 41.70 +-1.9 (95%) LOS: 100.0%
Total: 39907 W: 8704 L: 3937 D: 27266
40000 @ 60+0.6 th 1 Regression/progression test against SF9 after "Double weight of capture history" of August, 14th

Finished - 33101 tests

18-08-15 sg lmr_best_move2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 10568 W: 1690 L: 1771 D: 7107
sprt @ 60+0.6 th 1 LTC: At PV nodes do less reduction if best value < draw <= static eval.
18-08-15 sni mob_initiative6 diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 18988 W: 4069 L: 4108 D: 10811
sprt @ 10+0.1 th 1 With central pawns bonus (A,B)=(3,3)
18-08-15 Gua Tweaks diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 13478 W: 2818 L: 2929 D: 7731
sprt @ 10+0.1 th 1 Several parameter tweaks based on recent tests
18-08-14 fau FauMob3 diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 48620 W: 7992 L: 7992 D: 32636
sprt @ 60+0.6 th 1 LTC: mobility tuned values after all games, half the changes, with a few manual tweaks to some values
18-08-15 sni mob_initiative6^ diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 9649 W: 2045 L: 2129 D: 5475
sprt @ 10+0.1 th 1 Normal bonus for mobile pawns (not in initiative)
18-08-14 sni game_phase5 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 11269 W: 1798 L: 1876 D: 7595
sprt @ 60+0.6 th 1 LTC: Use pawns in game phase: pawns/2 - 6
18-08-15 pro ps_safety3 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 43042 W: 9336 L: 9257 D: 24449
sprt @ 10+0.1 th 1 bonus for king blocking enemy pawn on edge files, ranks 1 and 2 only.
18-08-15 pro ps_safety3 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 29028 W: 6327 L: 6316 D: 16385
sprt @ 10+0.1 th 1 king blocking pawns on any rank and edge files.
18-08-14 sni mob_initiative6^ diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 18864 W: 4066 L: 4105 D: 10693
sprt @ 10+0.1 th 1 Pawn mobility
18-08-14 sni mob_initiative6 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 15516 W: 3394 L: 3449 D: 8673
sprt @ 10+0.1 th 1 Take 2, without the centrality bonus
18-08-14 sg lmr_best_move2 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 35317 W: 5781 L: 5765 D: 23771
sprt @ 60+0.6 th 1 LTC: Fixed version. Less reduction if value of best move < draw <= static eval.
18-08-14 sni connected11 diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 29231 W: 4694 L: 4756 D: 19781
sprt @ 60+0.6 th 1 LTC: Tweak connected pawns seed[] array : +5
18-08-14 sg lmr_best_move2 diff
LLR: 2.95 (-2.94,2.94) [0.00,5.00]
Total: 90804 W: 20151 L: 19573 D: 51080
sprt @ 10+0.1 th 1 At PV nodes do less reduction if best value < draw <= static eval.
18-08-14 sg prune_best_move2 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 18647 W: 4044 L: 4084 D: 10519
sprt @ 10+0.1 th 1 No move count pruning if best value > static eval at PV nodes. (park with prio -1).
18-08-14 MJZ Knight_OP diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 5316 W: 1131 L: 1237 D: 2948
sprt @ 10+0.1 th 1 More bonus for Knight oupost if not bishop on same square color, Rerun after correction.
18-08-14 MJZ Knight_OP diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 13867 W: 3020 L: 3083 D: 7764
sprt @ 10+0.1 th 1 More bonus for Knight oupost if not bishop on same square color, first try to see it it works well.
18-08-14 sni connected11 diff
LLR: 2.95 (-2.94,2.94) [0.00,4.00]
Total: 16486 W: 3758 L: 3528 D: 9200
sprt @ 10+0.1 th 1 Tweak connected pawns seed[] array : +5
18-08-14 sni game_phase5 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 16967 W: 2745 L: 2801 D: 11421
sprt @ 60+0.6 th 1 LTC: Cap the phase to [0..128] (suggested by Stefan). Submitted with priority -1.
18-08-14 IIv king_psqt diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 14214 W: 2230 L: 2339 D: 9645
sprt @ 60+0.6 th 1 LTC: Tuned values.
18-08-14 SC scaleHistory diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 13089 W: 2822 L: 2889 D: 7378
sprt @ 10+0.1 th 1 Scale history comparison, take 1.
18-08-14 SC scaleHistory diff
LLR: -2.96 (-2.94,2.94) [-3.00,1.00]
Total: 6336 W: 1328 L: 1506 D: 3502
sprt @ 10+0.1 th 1 Scale history comparison, take 3.
18-08-14 SC scaleHistory diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 3640 W: 723 L: 836 D: 2081
sprt @ 10+0.1 th 1 Scale history comparison, take 2.
18-08-14 sg tune_king_shelter diff
LLR: 2.96 (-2.94,2.94) [0.00,4.00]
Total: 21074 W: 4724 L: 4477 D: 11873
sprt @ 10+0.1 th 1 Try from protonsprings simplification only the king shelter values as tuning.
18-08-14 sg tune_king_shelter2 diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 16904 W: 3696 L: 3794 D: 9414
sprt @ 10+0.1 th 1 Add some compensation.
18-08-14 fau FauMob1 diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 162417 W: 35713 L: 35239 D: 91465
sprt @ 10+0.1 th 1 mobility tuned values after all games, as they are (Fixed)
18-08-14 pro ps_safety2 diff
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 16883 W: 3708 L: 3577 D: 9598
sprt @ 10+0.1 th 1 -5, and +5 in safety initialization.
18-08-14 pro ps_safety2 diff
LLR: -2.95 (-2.94,2.94) [-3.00,1.00]
Total: 34636 W: 7445 L: 7677 D: 19514
sprt @ 10+0.1 th 1 remove safety initialization using -9 offset.
18-08-14 pro ps_safety2 diff
LLR: -2.95 (-2.94,2.94) [-3.00,1.00]
Total: 20464 W: 4316 L: 4519 D: 11629
sprt @ 10+0.1 th 1 remove safety initialization (again) using -7 offset. Not sure if this is a simplification or not.
18-08-14 sni game_phase5 diff
LLR: 2.95 (-2.94,2.94) [0.00,5.00]
Total: 49839 W: 11185 L: 10807 D: 27847
sprt @ 10+0.1 th 1 Cap the phase to [0..128] (suggested by Stefan)
18-08-14 IIv king_psqt diff
LLR: 2.95 (-2.94,2.94) [0.00,4.00]
Total: 26220 W: 5823 L: 5557 D: 14840
sprt @ 10+0.1 th 1 Tuned values.
18-08-14 fau FauMob3 diff
LLR: 2.95 (-2.94,2.94) [0.00,4.00]
Total: 45776 W: 10129 L: 9786 D: 25861
sprt @ 10+0.1 th 1 mobility tuned values after all games, half the changes, with a few manual tweaks to some values
18-08-14 sni game_phase5 diff
LLR: 2.96 (-2.94,2.94) [0.00,5.00]
Total: 42572 W: 9432 L: 9092 D: 24048
sprt @ 10+0.1 th 1 Use pawns in game phase: pawns/2 - 6
18-08-14 sni game_phase4 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 38923 W: 8465 L: 8406 D: 22052
sprt @ 10+0.1 th 1 Pretend that SF knows the future: anticipating the endgame values by shifting the game phase.
18-08-14 sg lmr_best_move2 diff
LLR: 2.95 (-2.94,2.94) [0.00,5.00]
Total: 36700 W: 8148 L: 7837 D: 20715
sprt @ 10+0.1 th 1 Fixed version. Less reduction if value of best move < draw <= static eval.
18-08-14 SC uselessPawns diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 40075 W: 8655 L: 8662 D: 22758
sprt @ 10+0.1 th 1 PMG 132.
18-08-14 SC uselessPawns diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 32880 W: 7144 L: 7179 D: 18557
sprt @ 10+0.1 th 1 PMG 132, 210.
18-08-14 sg lmr_best_move2 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 4219 W: 907 L: 1019 D: 2293
sprt @ 10+0.1 th 1 Less reduction if best value < draw <= static eval.
18-08-14 fau FauMob2 diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 26290 W: 5713 L: 5774 D: 14803
sprt @ 10+0.1 th 1 mobility tuned values after all games, with a few manual tweaks to some values
18-08-14 xor fbTweak^ diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 31426 W: 7002 L: 7042 D: 17382
sprt @ 10+0.1 th 1 Subtract 16 from this constant.
18-08-14 Anb lbm diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 17555 W: 3758 L: 3804 D: 9993
sprt @ 10+0.1 th 1 STC: take 3, add suggestions from SG
18-08-14 sg lmr_best_move2 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 13152 W: 2832 L: 2899 D: 7421
sprt @ 10+0.1 th 1 Less reduction if value of best move < static eval < draw.
18-08-14 sg lmr_best_move2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 4267 W: 865 L: 975 D: 2427
sprt @ 10+0.1 th 1 Less reduction if value of best move < draw <= static eval.
18-08-14 SC uselessPawns diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 8034 W: 1655 L: 1787 D: 4592
sprt @ 10+0.1 th 1 132, 204.
18-08-14 Anb doubled-blocked diff
LLR: -2.94 (-2.94,2.94) [0.00,5.00]
Total: 4378 W: 885 L: 994 D: 2499
sprt @ 10+0.1 th 1 STC: differentiate between blocked and unblocked doubled pawns
18-08-14 sg prune_best_move diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 18172 W: 3904 L: 3946 D: 10322
sprt @ 10+0.1 th 1 Try the opposite. No move count pruning if best move < static eval.
18-08-14 Anb lbm diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 13358 W: 2911 L: 2977 D: 7470
sprt @ 10+0.1 th 1 STC: take 2, mob<=3
18-08-14 fau FauMob1 diff
LLR: -0.22 (-2.94,2.94) [0.00,4.00]
Total: 383 W: 80 L: 91 D: 212
sprt @ 10+0.1 th 1 mobility tuned values after all games, as they are
18-08-14 SC qHistory diff
LLR: -2.94 (-2.94,2.94) [0.00,5.00]
Total: 17009 W: 3588 L: 3636 D: 9785
sprt @ 10+0.1 th 1 Move count based reductions in qsearch.
18-08-14 sni bad_capture4 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 7710 W: 1590 L: 1683 D: 4437
sprt @ 10+0.1 th 1 bad captures: SEE < max(0, value/2048)
18-08-14 Anb lbm diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 6369 W: 1345 L: 1445 D: 3579
sprt @ 10+0.1 th 1 STC: if low bishop mobility is due to enemy pawns blocking the way give additional penalty