Stockfish Testing Queue

Finished - 2445 tests

19-03-23 MJZ Eval-Risk3 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 33117 W: 7337 L: 7337 D: 18443
sprt @ 10+0.1 th 1
Extension for last leafs of PV if position is risky
19-03-23 MJZ Eval-Risk3 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 16698 W: 3629 L: 3711 D: 9358
sprt @ 10+0.1 th 1
Extension for last leafs if position is unsafe - rerun after correction (NMP & excludedMove).
19-03-23 MJZ Eval-Risk2 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 7981 W: 1714 L: 1839 D: 4428
sprt @ 10+0.1 th 1
Decrease reduction for all unsafe positions - rerun after correction (NMP & excludedMove).
19-03-22 MJZ Eval-Risk2 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 56206 W: 12492 L: 12378 D: 31336
sprt @ 10+0.1 th 1
Decrease reduction for unsafe positions of CST.
19-03-22 MJZ Eval-Risk3 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 16762 W: 3655 L: 3737 D: 9370
sprt @ 10+0.1 th 1
Extension for last leafs if position is unsafe.
19-03-22 MJZ Eval-Risk3 diff
LLR: -2.94 (-2.94,2.94) [0.50,4.50]
Total: 28268 W: 6248 L: 6272 D: 15748
sprt @ 10+0.1 th 1
Decrease reduction for unsafe positions in CST - threshold 300, only MG (NPM > 9000).
19-03-22 MJZ Eval-Risk3 diff
LLR: -2.32 (-2.94,2.94) [0.50,4.50]
Total: 22293 W: 4928 L: 4947 D: 12418
sprt @ 10+0.1 th 1
Decrease reduction for unsafe positions in CST - threshold 260.
19-03-22 MJZ Eval-Risk3 diff
LLR: -1.72 (-2.94,2.94) [0.50,4.50]
Total: 19464 W: 4337 L: 4336 D: 10791
sprt @ 10+0.1 th 1
Decrease reduction for unsafe positions in CST - threshold 340.
19-03-22 MJZ Eval-Risk3 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 21160 W: 4687 L: 4747 D: 11726
sprt @ 10+0.1 th 1
Decrease reduction for all unsafe positions.
19-03-22 MJZ Eval-Risk diff
LLR: -2.94 (-2.94,2.94) [0.50,4.50]
Total: 10956 W: 2401 L: 2511 D: 6044
sprt @ 10+0.1 th 1
Odd eval for unsafe positions. Just checking the ELO loss for first changes.
19-03-21 MJZ Eval-Risk diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 3924 W: 812 L: 958 D: 2154
sprt @ 10+0.1 th 1
Trying to make more interaction between evaluate and search procedure. First try to check sanity (skip futility pruning for risky positions).
19-03-21 MJZ BlockerForHanging diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 4103 W: 851 L: 996 D: 2256
sprt @ 10+0.1 th 1
Bonus if we pin a piece/pawn with hanging piece behind. (5, 5), only non pawn enemies.
19-03-21 MJZ BlockerForHanging diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 5118 W: 1054 L: 1193 D: 2871
sprt @ 10+0.1 th 1
Bonus if we pin a piece/pawn with hanging piece behind. (5, 5), try 2 after correction
19-03-21 MJZ BlockerForHanging diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 7078 W: 1500 L: 1630 D: 3948
sprt @ 10+0.1 th 1
Bonus if we pin a piece/pawn with hanging piece behind. (5, 5)
19-03-21 MJZ Shuffle-TT2d diff
LLR: -0.34 (-2.94,2.94) [0.00,3.50]
Total: 5451 W: 937 L: 942 D: 3572
sprt @ 60+0.6 th 1
Last test (TT2c) showed some time trouble. I think it is caused by long extensions. This is a try to limit extension lengths.
19-03-20 MJZ Shuffle-TT2c diff
ELO: -0.64 +-15.9 (95%) LOS: 46.8%
Total: 540 W: 79 L: 80 D: 381
10000 @ 60+0.6 th 4
Second verification for this commit : higher depth and multi threads (usage of TT). Fixed number of games to save ressources.
19-03-20 MJZ Probcut-WS diff
LLR: -2.94 (-2.94,2.94) [0.50,4.50]
Total: 23377 W: 5192 L: 5240 D: 12945
sprt @ 10+0.1 th 1
Limit probcut for winning side. Try opposite only beta > -300.
19-03-19 MJZ Shuffle-TT2c diff
LLR: 2.95 (-2.94,2.94) [0.00,3.50]
Total: 51249 W: 8807 L: 8486 D: 33956
sprt @ 60+0.6 th 1
Last commit passed spec LTC but I am still afraid that the shuffle limit of 32 half moves is not enough is some special positions. New try with 36 half moves and extension at move 14. spec LTC.
19-03-20 MJZ Probcut-WS diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 12290 W: 2674 L: 2778 D: 6838
sprt @ 10+0.1 th 1
Limit probcut for winning side. Only beta < 300.
19-03-19 MJZ Shuffle-TT2c diff
LLR: -2.94 (-2.94,2.94) [0.50,4.50]
Total: 32595 W: 7273 L: 7275 D: 18047
sprt @ 10+0.1 th 1
Last commit passed spec LTC but I am still afraid that the shuffle limit of 32 half moves is not enough is some special positions. New try with 36 half moves and extension at move 14.
19-03-18 MJZ Shuffle-TT2 diff
LLR: 0.16 (-2.94,2.94) [0.00,3.50]
Total: 23601 W: 4056 L: 3979 D: 15566
sprt @ 60+0.6 th 1
Return draw if we are shuffling and the position has been already searched (TT). Shuffle limit = 32, extension limit = 18 (LTC with low thp)
19-03-18 MJZ Shuffle-TT2b diff
LLR: 2.95 (-2.94,2.94) [0.00,3.50]
Total: 21723 W: 3815 L: 3577 D: 14331
sprt @ 60+0.6 th 1
Return draw if we are shuffling and the position has been already searched (TT). Shuffle limit = 32, extension limit = 16 & only for PV (LTC with low thp)
19-03-14 MJZ Shuffle-Reduction diff
ELO: -2.99 +-5.1 (95%) LOS: 12.6%
Total: 5000 W: 683 L: 726 D: 3591
5000 @ 60+0.6 th 8
Try to detect shuffling positions. Draw limit = 32, reduction limit = 12 if alpha > 20 (Limited LTC with 8 threads to reach higher depths)
19-03-15 MJZ Shuffle-TT2 diff
LLR: -2.95 (-2.94,2.94) [0.00,3.50]
Total: 10278 W: 1652 L: 1799 D: 6827
sprt @ 60+0.6 th 1
Return draw if we are shuffling and the position has been already searched (TT). Shuffle limit = 18, depth < 3. Speculative LTC
19-03-15 MJZ Shuffle-TT2 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 63126 W: 14000 L: 13852 D: 35274
sprt @ 10+0.1 th 1
Return draw if we are shuffling and the position has been already searched (TT). Shuffle limit = 18, depth < 3. Try STC, even if the goal is to improve LTC.
19-03-15 MJZ Shuffle-TT diff
ELO: -0.98 +-7.1 (95%) LOS: 39.3%
Total: 3199 W: 550 L: 559 D: 2090
5000 @ 60+0.6 th 1
Return draw if we are shuffling and the position has been already searched (TT). Shuffle limit = 16, depth < 3
19-03-15 MJZ Shuffle-TT2 diff
ELO: 3.39 +-5.7 (95%) LOS: 87.6%
Total: 4719 W: 815 L: 769 D: 3135
5000 @ 60+0.6 th 1
Return draw if we are shuffling and the position has been already searched (TT). Shuffle limit = 18, depth < 3
19-03-15 MJZ Shuffle-TT diff
ELO: 0.28 +-5.5 (95%) LOS: 53.9%
Total: 5000 W: 818 L: 814 D: 3368
5000 @ 60+0.6 th 1
Return draw if we are shuffling and the position has been already searched (TT). Shuffle limit = 20, depth < 3
19-03-14 MJZ Shuffle-Reduction2 diff
ELO: -1.96 +-5.6 (95%) LOS: 24.7%
Total: 4951 W: 825 L: 853 D: 3273
5000 @ 60+0.6 th 1
Try to detect shuffling positions. Draw limit = 28, reduction limit = 12 (Limited LTC)
19-03-14 MJZ Shuffle-Reduction diff
ELO: 0.97 +-5.5 (95%) LOS: 63.5%
Total: 4995 W: 835 L: 821 D: 3339
5000 @ 60+0.6 th 1
New try to detect shuffling positions. Draw limit = 32, reduction limit = 12 (Limited LTC)
19-03-14 MJZ King_path diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 18035 W: 3961 L: 4036 D: 10038
sprt @ 10+0.1 th 1
Bonus if King can reach quicky central squares just before endgame. (0, 20), NPM < 8000
19-03-14 MJZ King_path diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 13607 W: 2987 L: 3084 D: 7536
sprt @ 10+0.1 th 1
Bonus if King can reach quicky central squares just before endgame. (0, 40), NPM < 9000
19-03-14 MJZ PV_extension2 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 58839 W: 13151 L: 13023 D: 32665
sprt @ 10+0.1 th 1
Testing the effect of extending last leafs of PV line under certain conditions. 180 < abs(alpha) < 600 && rule50 > 6 && ply > 6
19-03-14 MJZ King_path diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 13944 W: 3055 L: 3151 D: 7738
sprt @ 10+0.1 th 1
Bonus if King can reach quicky central squares just before endgame. (0, 40)
19-03-13 MJZ PV_extension3 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 23454 W: 5175 L: 5223 D: 13056
sprt @ 10+0.1 th 1
Testing the effect of extending last leafs of PV line under certain conditions. 240 < abs(alpha) < 600 && rule50 > 8 && ply > 8
19-03-13 MJZ PV_extension2 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 15853 W: 3467 L: 3553 D: 8833
sprt @ 10+0.1 th 1
Testing the effect of extending last leafs of PV line under certain conditions. 200 < abs(alpha) < 600 && rule50 > 9 && ply > 9
19-03-13 MJZ PV_extension2 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 60814 W: 13496 L: 13359 D: 33959
sprt @ 10+0.1 th 1
Testing the effect of extending last leafs of PV line under certain conditions. 200 < abs(alpha) < 500 && rule50 > 8 && ply > 8
19-03-13 MJZ PV_extension diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 62528 W: 13979 L: 13833 D: 34716
sprt @ 10+0.1 th 1
Testing the effect of extending last leafs of PV line under certain conditions. 200 < abs(alpha) < 600 && rule50 > 8 && ply > 8
19-03-13 MJZ PV_extension3 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 7649 W: 1668 L: 1795 D: 4186
sprt @ 10+0.1 th 1
Testing the effect of extending last leafs of PV line under certain conditions. 200 < abs(alpha) < 400 && rule50 > 8 && ply > 8
19-03-13 MJZ PV_extension2 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 26154 W: 5807 L: 5842 D: 14505
sprt @ 10+0.1 th 1
Testing the effect of extending last leafs of PV line under certain conditions. 200 < abs(alpha) < 700 && rule50 > 8 && ply > 8
19-03-13 MJZ PV_extension diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 19838 W: 4343 L: 4409 D: 11086
sprt @ 10+0.1 th 1
Testing the effect of extending last leafs of PV line under certain conditions. 100 < abs(alpha) < 700 && rule50 > 8 && ply > 8
19-03-13 MJZ Blocked_posNM2 diff
ELO: -2.28 +-6.3 (95%) LOS: 23.9%
Total: 3961 W: 657 L: 683 D: 2621
5000 @ 60+0.6 th 1
Limited test at LTC to check sanity
19-02-23 MJZ Shuffle_extension2 diff
LLR: -2.95 (-2.94,2.94) [0.00,3.50]
Total: 225406 W: 37770 L: 37315 D: 150321
sprt @ 60+0.6 th 1
Shuffle extension : extend PV if rule50 > 20 && depth < 2, - LTC (needs high depth search)
19-03-03 MJZ TTmove2 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 2785 W: 544 L: 695 D: 1546
sprt @ 10+0.1 th 1
Decrease reduction if the moved piece is same than ttMove, add ttMove condition and filter !ttcapture and positiveSEE, one last try for NPM > 12000
19-03-03 MJZ TTmove diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 19919 W: 4391 L: 4457 D: 11071
sprt @ 10+0.1 th 1
Decrease reduction if the moved piece is same than ttMove, add ttMove condition and filter !ttcapture
19-03-03 MJZ TTmove2 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 3299 W: 666 L: 815 D: 1818
sprt @ 10+0.1 th 1
Decrease reduction if the moved piece is same than ttMove, add ttMove condition and filter !ttcapture and positiveSEE
19-03-03 MJZ TTmove2 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 3148 W: 638 L: 788 D: 1722
sprt @ 10+0.1 th 1
Decrease reduction if the moved piece is same than ttMove, add ttMove condition and filter positive SEE
19-03-03 MJZ Move_sacrifice diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 3855 W: 765 L: 910 D: 2180
sprt @ 10+0.1 th 1
Decrease reduction if move is a sacrifice
19-03-03 MJZ TTmove2 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 5568 W: 1140 L: 1277 D: 3151
sprt @ 10+0.1 th 1
Decrease reduction if the moved piece is same than ttMove, add ttMove condition and filter sacrifices
19-03-03 MJZ Move_sacrifice diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 5390 W: 1149 L: 1288 D: 2953
sprt @ 10+0.1 th 1
Increase reduction if move is a sacrifice