Stockfish Testing Queue

Pending - 0 tests 0.0 hrs

None

Active - 0 tests

Finished - 1901 tests

18-12-17 Roc DeadSquare diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 21850 W: 4709 L: 4733 D: 12408
sprt @ 10+0.1 th 1 Trying to remove even more squares from the bishop mobility area
18-12-17 Roc BishopEdgeMobility diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 33646 W: 7304 L: 7271 D: 19071
sprt @ 10+0.1 th 1 BEM Remove some squares in the bishop mobility area For example: White b4 black b5 a3 is removed from the bishop mobility area.
18-12-17 Roc OverloadedPawns3 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 13107 W: 2780 L: 2847 D: 7480
sprt @ 10+0.1 th 1 OP3 Now that stronglyProtected is used in many place, try to improve it.
18-12-17 Roc PawnAttackBackUp diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 7573 W: 1611 L: 1705 D: 4257
sprt @ 10+0.1 th 1 PA_1 White Bc3 f6 Include g7 in the attackedBy2[Us] bitboard
18-12-16 Roc Restriction2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 12877 W: 2801 L: 2869 D: 7207
sprt @ 10+0.1 th 1 R2b
18-12-16 Roc Restriction3 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 7664 W: 1622 L: 1716 D: 4326
sprt @ 10+0.1 th 1 R3
18-12-16 Roc Restriction2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 5375 W: 1137 L: 1242 D: 2996
sprt @ 10+0.1 th 1 R2 Consider restrcited when attacking opponent lowranks Take 1: first 3 ranks
18-12-16 Roc UndefendedQ diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 5126 W: 1080 L: 1186 D: 2860
sprt @ 10+0.1 th 1 UQ1 Handle the undefended WeakQueen case when pinned my opponent queen (of course, if both queens are not defended, this will cancel out)
18-12-16 Roc NotSoWeakQueen diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 30482 W: 6644 L: 6626 D: 17212
sprt @ 10+0.1 th 1 NSW_1
18-12-16 Roc KingRingComputation diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 22572 W: 4923 L: 4998 D: 12651
sprt @ 10+0.1 th 1 Try a different king ring computation method (and as master include the 9th square which makes a difference since some x rays through our queen will "attack" the ksq)
18-12-16 Roc 26008de diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 8473 W: 1807 L: 1938 D: 4728
sprt @ 10+0.1 th 1 Try a different king ring calculation method (and always 8 squares)
18-12-16 Roc KRSP_2 diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 17753 W: 3804 L: 3849 D: 10100
sprt @ 10+0.1 th 1 KR2 Always keep the squares attacked by pawn in the king ring.
18-12-15 Roc KRSimplify2 diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 25655 W: 5563 L: 5626 D: 14466
sprt @ 10+0.1 th 1 Find out if this fonctional change is better
18-12-15 Roc KRSimplify diff
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 9494 W: 2191 L: 2045 D: 5258
sprt @ 10+0.1 th 1 KRS-2 Fixed code Run as a simplification, since it reduces the number of processor instructions (line 313 is called for each piece)
18-12-14 Roc PawnPushSupportR2 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 14139 W: 3036 L: 3098 D: 8005
sprt @ 10+0.1 th 1 PPR2-2 Take 3: Consider a pawn push safe if supported from behind by us but not attackedby2 by them
18-12-14 Roc PawnPushSupportR2 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 8090 W: 1714 L: 1806 D: 4570
sprt @ 10+0.1 th 1 PP_2 Take 2: Consider a pawn push safe when supported from behind and not defended from front
18-12-14 Roc KindDangerByRank2 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 6458 W: 1366 L: 1466 D: 3626
sprt @ 10+0.1 th 1 KD2
18-12-14 Roc KindDangerByRank diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 2934 W: 601 L: 719 D: 1614
sprt @ 10+0.1 th 1 KD_1 increase king danger for adventurous kings
18-12-14 Roc PawnPushSupportR2 diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 10965 W: 2318 L: 2396 D: 6251
sprt @ 10+0.1 th 1 PP_1 Consider an advanced pawn push safe if supported from behind
18-12-13 Roc TunedBishopMobPerSquare diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 12393 W: 2653 L: 2724 D: 7016
sprt @ 10+0.1 th 1 TB_1 From my last SPSA tuning
18-12-08 Roc BishopMobPerSquare diff
47651/50000 iterations
100000/100000 games played
100000 @ 20+0.2 th 1 Fixed bench
18-12-08 Roc BishopMobSquareTuned diff
LLR: -1.51 (-2.94,2.94) [0.00,5.00]
Total: 100 W: 4 L: 92 D: 4
sprt @ 10+0.1 th 1 BMST-1 Raw Tuned values 100000 games
18-12-08 Roc BishopMobPerSquare diff
47665/50000 iterations
100000/100000 games played
100000 @ 20+0.2 th 1 Tune bishop mobility according to the bishop square, using log formulas
18-12-02 Roc XrayAndAttackedBy diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 6127 W: 1296 L: 1397 D: 3434
sprt @ 10+0.1 th 1 Fixed version
18-12-02 Roc XrayAndAttackedBy diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 4682 W: 1005 L: 1114 D: 2563
sprt @ 10+0.1 th 1 XRc Last try
18-12-02 Roc KRBlockedPawns diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 25316 W: 5482 L: 5489 D: 14345
sprt @ 10+0.1 th 1 KR2
18-12-02 Roc KRPawnDefense diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 15563 W: 3324 L: 3379 D: 8860
sprt @ 10+0.1 th 1 KRP
18-12-02 Roc XrayAndAttackedBy diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 3332 W: 693 L: 809 D: 1830
sprt @ 10+0.1 th 1 Try some adjustments to attackedBy2, leading to more accurate safe checks
18-12-02 Roc BishopBlockers5 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 11702 W: 2513 L: 2587 D: 6602
sprt @ 10+0.1 th 1 BB5 Don't cont our pawns blocked by our our pieces as "blocked" for the BishopPawn penalty
18-12-02 Roc TropismDoublePawnDefens diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 15618 W: 3393 L: 3448 D: 8777
sprt @ 10+0.1 th 1 VT_1 Exclude doubly protected squares (by pawn) for tropism too.
18-12-02 Roc VizKingSafety2 diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 12793 W: 2745 L: 2814 D: 7234
sprt @ 10+0.1 th 1 VKS2b Save the double pawn attacks in the pawn hash table. Same bench, looking for a speedup.
18-12-02 Roc ExtendKingRing diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 26545 W: 5754 L: 5755 D: 15036
sprt @ 10+0.1 th 1 EKR
18-12-02 Roc KSPinnedPawns diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 44205 W: 9620 L: 9535 D: 25050
sprt @ 10+0.1 th 1 KSP Exclude pinned pawns from the double defended calculation
18-12-02 Roc VizKingSafety2 diff
LLR: -2.94 (-2.94,2.94) [0.00,5.00]
Total: 20901 W: 4577 L: 4605 D: 11719
sprt @ 10+0.1 th 1 Test a speed up on Viz king safety patch
18-11-17 Roc DomKnight diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 13238 W: 2799 L: 2866 D: 7573
sprt @ 10+0.1 th 1 DK_2 rebased on latest master only for knight not in largecenter bench; 3386676
18-11-17 Roc DomKnight diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 15152 W: 3220 L: 3277 D: 8655
sprt @ 10+0.1 th 1 DK_Viz Retake on Viz idea
18-11-12 Roc BishopSkewer diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 20835 W: 4418 L: 4448 D: 11969
sprt @ 10+0.1 th 1 BSK_2 xray through rooks only when not pawn protected.
18-11-12 Roc LowRankTuned1 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 31266 W: 6789 L: 6767 D: 17710
sprt @ 10+0.1 th 1 LRT1 Tuned PSQT mg values for pieces on first 3 ranks. For King, first 2 rank mg values with assymetry.
18-11-11 Roc LowRankTune diff
38362/40000 iterations
80000/80000 games played
80000 @ 20+0.2 th 1 add missing mB[7] parameters
18-11-11 Roc BishopSkewer diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 7842 W: 1629 L: 1722 D: 4491
sprt @ 10+0.1 th 1 BSK_R Allow a bishop to x-ray through opponent rook.
18-11-11 Roc KingOptions diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 22665 W: 4860 L: 4881 D: 12924
sprt @ 10+0.1 th 1 Take 2
18-11-11 Roc KingOptions diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 17539 W: 3679 L: 3725 D: 10135
sprt @ 10+0.1 th 1 mg bonus for each Ke1 legal options on first rank (0-0-0, Rd1, Rf1 or 0-0)
18-11-11 Roc LowRankTune diff
19696/40000 iterations
41153/80000 games played
80000 @ 20+0.2 th 1 Tuning mg values for PSQT first 2 ranks for pieces, and asymmetric king
18-11-08 Roc AssymetricKing1 diff
LLR: -0.92 (-2.94,2.94) [0.00,5.00]
Total: 36122 W: 5857 L: 5755 D: 24510
sprt @ 60+0.6 th 1 LTC for info only. Try only first 2 ranks assymetric for king.
18-11-08 Roc AssymetricKing1 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 18003 W: 3857 L: 3900 D: 10246
sprt @ 10+0.1 th 1 Keep only first 2 ranks assymetric for king. Tested [0,5] against last master.
18-11-07 Roc SymmetricKing1 diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 14179 W: 2989 L: 3097 D: 8093
sprt @ 10+0.1 th 1 Sanity test # 1. Test avg of asymmetric king values
18-11-06 Roc KingInCenter3 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 8990 W: 1919 L: 2006 D: 5065
sprt @ 10+0.1 th 1 KC3 Larger penalty
18-11-06 Roc KingInCenter4 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 23059 W: 4859 L: 4879 D: 13321
sprt @ 10+0.1 th 1 KC4 More penalty for weak kingDanger squares when king in center
18-11-06 Roc KingInCenter3 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 33112 W: 7117 L: 7087 D: 18908
sprt @ 10+0.1 th 1 KC3 Add constant kingDnager penalty for king in center
18-11-06 Roc KingInCenter diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 6181 W: 1349 L: 1451 D: 3381
sprt @ 10+0.1 th 1 KC_2 increase closeenemy score when king on files c/d/e/f