Stockfish Testing Queue

Pending - 0 tests 0.0 hrs

None

Active - 3 tests

19-11-13 pb0 captureKiller5 diff
Pending... (0.1 hrs)
[-1.50,4.50]
sprt @ 10+0.1 th 1
Take 3
19-11-13 pb0 captureKiller5 diff
LLR: -2.28 (-2.94,2.94) [-1.50,4.50]
Total: 4807 W: 998 L: 1068 D: 2741
sprt @ 10+0.1 th 1
Take 2
19-11-13 pb0 captureKiller5^ diff
LLR: -2.35 (-2.94,2.94) [-1.50,4.50]
Total: 4146 W: 873 L: 948 D: 2325
sprt @ 10+0.1 th 1
Take 1

Finished - 1498 tests

19-11-12 pb0 futility3 diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 11385 W: 2435 L: 2511 D: 6439
sprt @ 10+0.1 th 1
Take 4 (improving tweak)
19-11-12 pb0 futility3 diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 14340 W: 3078 L: 3145 D: 8117
sprt @ 10+0.1 th 1
Take 3
19-11-12 pb0 futility3 diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 11964 W: 2538 L: 2612 D: 6814
sprt @ 10+0.1 th 1
Take 2
19-11-12 pb0 futility3 diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 7299 W: 1589 L: 1677 D: 4033
sprt @ 10+0.1 th 1
Some "empty-framework" monkey experiments on futility formula
19-11-12 pb0 capt_killer_score diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 28581 W: 6144 L: 6170 D: 16267
sprt @ 10+0.1 th 1
Take 1 (bugfixed)
19-11-11 pb0 capture_killer_repl_ttm diff
LLR: -2.95 (-2.94,2.94) [0.00,3.50]
Total: 30764 W: 4896 L: 4987 D: 20881
sprt @ 60+0.6 th 1
Take 2: use for probcut only (LTC)
19-11-11 pb0 capt_killer_repl_ttm_va diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 8661 W: 1810 L: 1894 D: 4957
sprt @ 10+0.1 th 1
Variant: validate capture-move candidate in search.cpp not in probcut movepick constructor
19-11-11 pb0 capture_killer_repl_ttm diff
LLR: 2.97 (-2.94,2.94) [-1.50,4.50]
Total: 27926 W: 6103 L: 5912 D: 15911
sprt @ 10+0.1 th 1
Take 2: use for probcut only
19-11-11 pb0 capture_killer_repl_ttm diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 11062 W: 2317 L: 2394 D: 6351
sprt @ 10+0.1 th 1
Capture-killer idea for replacing useles ttm's in movepicker construcors
19-11-11 pb0 quiet_probcut_ttm2 diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 9787 W: 2075 L: 2156 D: 5556
sprt @ 10+0.1 th 1
With threshold <=0 consider checking tt-move & don't incr. probCutCount
19-11-11 pb0 quiet_probcut_ttm2 diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 4871 W: 1003 L: 1098 D: 2770
sprt @ 10+0.1 th 1
With threshold <=0 consider also checking tt-moves
19-11-11 pb0 probCutNegSee diff
LLR: -2.94 (-2.94,2.94) [-1.50,4.50]
Total: 17687 W: 3806 L: 3863 D: 10018
sprt @ 10+0.1 th 1
Take 3
19-11-11 pb0 probCutNegSee diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 33571 W: 7176 L: 7187 D: 19208
sprt @ 10+0.1 th 1
Take 2: apply also on captures whith small gain so (more kick-ins)
19-11-11 pb0 probCutNegSee diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 5814 W: 1223 L: 1315 D: 3276
sprt @ 10+0.1 th 1
Probcut: verification search with depth 1 instad of 0 when negative SEE
19-11-08 pb0 probCutDepth5 diff
LLR: -2.95 (-2.94,2.94) [0.00,3.50]
Total: 46074 W: 7382 L: 7431 D: 31261
sprt @ 60+0.6 th 1
Experiment (empty queue), like to see how this performs at LTC since it might be tc sensitive.(lower throughput)
19-11-08 pb0 probCutQsearch diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 66069 W: 14359 L: 14275 D: 37435
sprt @ 10+0.1 th 1
Probcut: abandon earlier when first moves fail qsearch verification
19-11-08 pb0 probCutQsearch diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 5736 W: 1177 L: 1269 D: 3290
sprt @ 10+0.1 th 1
Take 3
19-11-08 pb0 probCutDepth5 diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 17625 W: 3776 L: 3834 D: 10015
sprt @ 10+0.1 th 1
Experiment (empty queue)
19-11-08 pb0 probCutQsearch diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 14690 W: 3095 L: 3161 D: 8434
sprt @ 10+0.1 th 1
Take 2 the opposite: search some more capture if first ones pass qsearch
19-11-08 pb0 quiet_probcut_ttm2 diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 7254 W: 1540 L: 1628 D: 4086
sprt @ 10+0.1 th 1
Another try
19-11-08 pb0 probCutVar diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 24733 W: 5274 L: 5311 D: 14148
sprt @ 10+0.1 th 1
Take 4 (correct bench)
19-11-08 pb0 probCutVar diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 18638 W: 4024 L: 4079 D: 10535
sprt @ 10+0.1 th 1
Trying another variant
19-11-08 pb0 probCutVar diff
LLR: -2.95 (-2.94,2.94) [0.00,3.50]
Total: 27699 W: 4458 L: 4557 D: 18684
sprt @ 60+0.6 th 1
Take 2: more cautious, just adapting raisedBeta (LTC)
19-11-08 pb0 probCutVar2 diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 9557 W: 2021 L: 2102 D: 5434
sprt @ 10+0.1 th 1
Take 5
19-11-08 pb0 probCutVar diff
LLR: 2.96 (-2.94,2.94) [-1.50,4.50]
Total: 21612 W: 4787 L: 4614 D: 12211
sprt @ 10+0.1 th 1
Take 2: more cautious, just adapting raisedBete
19-11-08 pb0 probCutVar diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 11931 W: 2531 L: 2605 D: 6795
sprt @ 10+0.1 th 1
Take 3
19-11-07 pb0 probCutQueen diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 29555 W: 6305 L: 6328 D: 16922
sprt @ 10+0.1 th 1
Take 2
19-11-07 pb0 probCutQueen diff
LLR: -2.94 (-2.94,2.94) [-1.50,4.50]
Total: 12578 W: 2703 L: 2775 D: 7100
sprt @ 10+0.1 th 1
Take 1
19-11-07 pb0 probCutContPenalty diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 12062 W: 2530 L: 2604 D: 6928
sprt @ 10+0.1 th 1
Take 2
19-11-07 pb0 probCutContPenalty diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 3115 W: 614 L: 714 D: 1787
sprt @ 10+0.1 th 1
Take 1
19-11-07 pb0 pinners6 diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 4874 W: 990 L: 1085 D: 2799
sprt @ 10+0.1 th 1
Small bonus for attacking weak opponent pinner with a smaller piece
19-11-06 pb0 pinners6 diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 23748 W: 5149 L: 5189 D: 13410
sprt @ 10+0.1 th 1
Small bonus for attacking opponent pinner (correct commit)
19-11-06 pb0 ttHitcutOff diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 16664 W: 3598 L: 3658 D: 9408
sprt @ 10+0.1 th 1
Take 3
19-11-06 pb0 ttHitcutOff diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 12753 W: 2729 L: 2801 D: 7223
sprt @ 10+0.1 th 1
Take 2
19-11-06 pb0 ttHitcutOff diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 398 W: 53 L: 166 D: 179
sprt @ 10+0.1 th 1
Early tt-cutoff also w/ lower tt-depth but value > beta by some degree
19-11-05 pb0 quiet_probcut_ttm diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 6726 W: 1400 L: 1489 D: 3837
sprt @ 10+0.1 th 1
Take 3: allow quiet ttm move in probcut with further raised beta
19-11-05 pb0 quiet_probcut_ttm diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 4429 W: 934 L: 1031 D: 2464
sprt @ 10+0.1 th 1
Take 2: allow quiet ttm move in probcut with further raised beta
19-11-05 pb0 quiet_probcut_ttm diff
LLR: -2.97 (-2.94,2.94) [-1.50,4.50]
Total: 2325 W: 458 L: 561 D: 1306
sprt @ 10+0.1 th 1
Allow quiet ttm move in probcut, take 1
19-11-05 pb0 badCaptureBoost diff
LLR: -1.81 (-2.94,2.94) [-1.50,4.50]
Total: 10489 W: 2285 L: 2321 D: 5883
sprt @ 10+0.1 th 1
Extra boost for 'bad'-captures after becoming bestmove
19-11-05 pb0 badCaptureBoost diff
LLR: -2.05 (-2.94,2.94) [-1.50,4.50]
Total: 9124 W: 1967 L: 2016 D: 5141
sprt @ 10+0.1 th 1
Take 2
19-10-21 pb0 fff diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 43598 W: 9387 L: 9339 D: 24872
sprt @ 10+0.1 th 1
Try the opposite: another follow-Move layer
19-10-24 pb0 recaptureMove diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 29241 W: 6277 L: 6301 D: 16663
sprt @ 10+0.1 th 1
Countercapture stats, take 2
19-10-24 pb0 recaptureMove diff
LLR: -2.96 (-2.94,2.94) [-1.50,4.50]
Total: 26012 W: 5697 L: 5730 D: 14585
sprt @ 10+0.1 th 1
Countercapture stats, take 1
19-10-24 pb0 recaptureMove diff
LLR: -2.95 (-2.94,2.94) [-1.50,4.50]
Total: 5501 W: 1144 L: 1237 D: 3120
sprt @ 10+0.1 th 1
Take 3 (and probably last on this idea)
19-10-23 pb0 exp55 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 24314 W: 5276 L: 5321 D: 13717
sprt @ 10+0.1 th 1
Second countermove, handle like killers are handled
19-10-23 pb0 exp55 diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 9545 W: 2003 L: 2120 D: 5422
sprt @ 10+0.1 th 1
Take 2
19-10-21 pb0 fff diff
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 19901 W: 4294 L: 4360 D: 11247
sprt @ 10+0.1 th 1
Take 2: experimental just read the values from the layer without writing
19-10-21 pb0 noFollowFollow diff
LLR: -2.95 (-2.94,2.94) [-3.00,1.00]
Total: 24234 W: 5135 L: 5346 D: 13753
sprt @ 10+0.1 th 1
Is follow follow move history still useful?
19-10-18 pb0 clearanceBonus diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 6842 W: 1386 L: 1516 D: 3940
sprt @ 10+0.1 th 1
Take 1
19-10-17 pb0 checkExt10 diff
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 19452 W: 4221 L: 4290 D: 10941
sprt @ 10+0.1 th 1
Take 2