Stockfish Testing Queue

Pending - 0 tests 0.0 hrs

None

Active - 1 tests

26-03-17 Ro MultiChecksQN diff
LLR: -2.04 (-2.94,2.94) [0.00,5.00]
Total: 19736 W: 3531 L: 3534 D: 12671
sprt @ 10+0.1 th 1 Score safe checks which can be given from many directions (only for queen and knight)

Finished - 944 tests

26-03-17 Ro WeakQueen&Rook diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 7071 W: 1202 L: 1297 D: 4572
sprt @ 10+0.1 th 1 Do not penalize a queen - rook or queen bishop battery
25-03-17 Ro OutpostOptions diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 28946 W: 5145 L: 5203 D: 18598
sprt @ 10+0.1 th 1 Take 2
21-03-17 Ro ControlledMajors diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 104735 W: 13625 L: 13404 D: 77706
sprt @ 60+0.6 th 1 Proceed to LTC since tuning shows that we have good parameter S(5, 10) which applies minimally in all cases.
24-03-17 Ro OutpostOptions diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 13351 W: 2329 L: 2440 D: 8582
sprt @ 10+0.1 th 1 Outpost experiment
20-03-17 Ro ControlledMajors diff
LLR: 2.95 (-2.94,2.94) [0.00,5.00]
Total: 40601 W: 7429 L: 7132 D: 26040
sprt @ 10+0.1 th 1 Adjust for controlled squares in the mobilityarea of major pieces.
19-03-17 Ro ControlledMobility diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 6759 W: 1165 L: 1262 D: 4332
sprt @ 10+0.1 th 1 Apply the "controlled" idea to mobilityArea
19-03-17 Ro SpaceControlled diff
LLR: -2.94 (-2.94,2.94) [0.00,5.00]
Total: 19490 W: 3484 L: 3526 D: 12480
sprt @ 10+0.1 th 1 Applying the controlled idea elsewhere in evaluate_threat did not help. So let's try applying it only in evaluate_space.
19-03-17 Ro Controlled1 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 5548 W: 979 L: 1081 D: 3488
sprt @ 10+0.1 th 1 Take 2 Consider a square "safe" for an attacking pawn if not "controlled by them"
19-03-17 Ro Controlled1 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 2222 W: 365 L: 482 D: 1375
sprt @ 10+0.1 th 1 Consider a pawn "safe" if protected by own pawn, or not controlled by them
18-03-17 Ro SimpleSafe diff
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 64001 W: 11471 L: 11423 D: 41107
sprt @ 10+0.1 th 1 Extend the "strongly protected" (== "controlled") idea so that it can be re-used at many places and make the code more readable.
18-03-17 Ro NoQueenSupport diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 5187 W: 850 L: 953 D: 3384
sprt @ 10+0.1 th 1 Do not consider a square "strongly protected" (attackedBy2) if defended by a queen.
18-03-17 Ro QueenSupport diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 6296 W: 1041 L: 1139 D: 4116
sprt @ 10+0.1 th 1 Adjustments to the attackedBy2 bitboard
17-03-17 Ro PawnImbalanceTuned diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 12050 W: 2114 L: 2230 D: 7706
sprt @ 10+0.1 th 1 Testing 32 param tuned model against Joerg 9 param green model
17-03-17 Ro PawnImbalanceTunedSimpl diff
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 22169 W: 3831 L: 3913 D: 14425
sprt @ 10+0.1 th 1 Testing 32 param tuned model simplified back to 9 params, against Joerg 9 param green model
16-03-17 Ro PawnImbalanceTune diff
38551/40000 iterations
79850/80000 games played
80000 @ 20+0.2 th 1 Previous run was stopped because "bestvalues" were not updating. Let's see if it will work better without the "UPDATE_ON_LAST();" (since I'm not using a postupdate function)
17-03-17 Ro ReachableOutpostTweak diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 35794 W: 6292 L: 6328 D: 23174
sprt @ 10+0.1 th 1 Allow squares occupied by unblocked pawns, and allows squares occupied by own pieces except king
16-03-17 Ro PawnImbalanceTune diff
99/40000 iterations
217/80000 games played
80000 @ 20+0.2 th 1 PawnSet based on joerg green values, but modelled with 36 parameters, hoping for a simplification after that.
15-03-17 Ro AttackedQueenMob diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 18047 W: 3175 L: 3224 D: 11648
sprt @ 10+0.1 th 1 Adjust queen mob when under attack by a non queen
15-03-17 Ro PassiveLever diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 19791 W: 3447 L: 3489 D: 12855
sprt @ 10+0.1 th 1 More precise version, do not consider a square strongly defended by a pawn if it is levered.
15-03-17 Ro PassiveLever diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 50116 W: 8947 L: 8861 D: 32308
sprt @ 10+0.1 th 1 penalty for a levered pawn which should not take because would leave another pawn unprotected
15-03-17 Ro ActiveLever diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 21425 W: 3757 L: 3792 D: 13876
sprt @ 10+0.1 th 1 Bonus for a lever when a take by the enemy would isolate the taker.
13-03-17 Ro PawnPushTweak diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 24687 W: 4324 L: 4397 D: 15966
sprt @ 10+0.1 th 1 Using the new stronglyProtected idea for the ThreatByPawnPush
13-03-17 Ro weakTweak diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 27978 W: 4909 L: 4917 D: 18152
sprt @ 10+0.1 th 1 Take 2
13-03-17 Ro weakTweak diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 11235 W: 1917 L: 1995 D: 7323
sprt @ 10+0.1 th 1 Small rewrite of evaluate_threats and consider weak the pawn attacked enemies
25-02-17 Ro MoveOrdering3 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 5951 W: 993 L: 1093 D: 3865
sprt @ 10+0.1 th 1 Using from_sq instead (suggested by SC)
25-02-17 Ro MoveOrdering2 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 3875 W: 651 L: 760 D: 2464
sprt @ 10+0.1 th 1 In evaluate, we do not apply a ThreatByRank on pawns. Try to do the same in movepicker
23-02-17 Ro MoveOrdering diff
LLR: -2.95 (-2.94,2.94) [-3.00,1.00]
Total: 65283 W: 8327 L: 8559 D: 48397
sprt @ 60+0.6 th 1 LTC Now that we have a ThreatByRank bonus, do we still need the rank factor when we sort captures ?
23-02-17 Ro ThreatsByKing2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 16952 W: 2923 L: 2977 D: 11052
sprt @ 10+0.1 th 1 Bonus for threats by king which can be maintained from another square
23-02-17 Ro OverloadedKing diff
LLR: -2.94 (-2.94,2.94) [0.00,5.00]
Total: 6037 W: 1017 L: 1116 D: 3904
sprt @ 10+0.1 th 1 Take 3: Apply a smaller penalty, and only once.
23-02-17 Ro OverloadedKing diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 7715 W: 1318 L: 1410 D: 4987
sprt @ 10+0.1 th 1 Take 2: logical fix
23-02-17 Ro OverloadedKing diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 7968 W: 1366 L: 1457 D: 5145
sprt @ 10+0.1 th 1 Evaluate squares with pieces which are defended only by the king
23-02-17 Ro MoveOrdering diff
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 41295 W: 7472 L: 7386 D: 26437
sprt @ 10+0.1 th 1 Now that we have a ThreatByRank bonus, do we still need the rank factor when we sort captures ?
21-02-17 Ro XRayMob diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 12242 W: 2186 L: 2259 D: 7797
sprt @ 10+0.1 th 1 And what if we simply exclude only the rook home rank xrays through queen ?
20-02-17 Ro XRayMob diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 59840 W: 10780 L: 10652 D: 38408
sprt @ 10+0.1 th 1 Take 4, suggested by snicolet, applied in the bishop case only.
20-02-17 Ro XRayMob diff
LLR: -2.99 (-2.94,2.94) [0.00,5.00]
Total: 2870 W: 449 L: 563 D: 1858
sprt @ 10+0.1 th 1 Take 5: Applying a similar change as in test 4, this time measuring the rook case only
19-02-17 Ro XRayMob diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 59225 W: 10634 L: 10509 D: 38082
sprt @ 10+0.1 th 1 Last try, remove some xrays for bishop only
19-02-17 Ro XRayMob diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 10667 W: 1847 L: 1927 D: 6893
sprt @ 10+0.1 th 1 take 2: Just for the rook
19-02-17 Ro XRayMob diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 19192 W: 3413 L: 3457 D: 12322
sprt @ 10+0.1 th 1 Consider x-ray through queen only on special occasions.
18-02-17 Ro SecondaryAttacks diff
LLR: -2.33 (-2.94,2.94) [0.00,5.00]
Total: 55558 W: 10183 L: 10044 D: 35331
sprt @ 10+0.1 th 1 1b ) Just for knights, but adding also to ei.attackedBy[][KNIGHT]
18-02-17 Ro SecondaryAttacksByQueen diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 9077 W: 1600 L: 1687 D: 5790
sprt @ 10+0.1 th 1 Similar idea, with the queen.
18-02-17 Ro KingDecoy diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 20133 W: 3565 L: 3605 D: 12963
sprt @ 10+0.1 th 1 Take 2: only decoy by capture, as originally intended
17-02-17 Ro SecondaryAttacks diff
LLR: -2.94 (-2.94,2.94) [0.00,5.00]
Total: 42156 W: 7594 L: 7540 D: 27022
sprt @ 10+0.1 th 1 Add attacks by a Knight after a check to the attackedBy[][ALL_PIECES] Purpose is mainly to change the refine the hanging evaluation.
17-02-17 Ro KingDecoy diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 10557 W: 1885 L: 1965 D: 6707
sprt @ 10+0.1 th 1 Analyse king decoy captures followed by knight check
17-02-17 Ro PawnPushSupport2 diff
LLR: -2.35 (-2.94,2.94) [0.00,5.00]
Total: 18980 W: 2445 L: 2477 D: 14058
sprt @ 60+0.6 th 1 Speculative LTC for this Jan 21 2017 yellow test http://tests.stockfishchess.org/tests/view/5882e6fe0ebc5915193f78f0
12-02-17 Ro pp_verify_npm diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 28338 W: 3676 L: 3702 D: 20960
sprt @ 60+0.6 th 1 Sanity check: Retry vintage version of unstoppable (before it went through 3 simplifications !)
11-02-17 Ro pp_verify_npm diff
LLR: 2.95 (-2.94,2.94) [0.00,5.00]
Total: 55932 W: 10182 L: 9821 D: 35929
sprt @ 10+0.1 th 1 Sanity check: Retry vintage version of unstoppable (before it went through 3 simplifications !)
11-02-17 Ro pp_EG diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 46992 W: 8416 L: 8413 D: 30163
sprt @ 10+0.1 th 1 Increase endgame pp by constant 20
06-02-17 Ro PP_Resolution2 diff
LLR: -2.96 (-2.94,2.94) [-3.00,1.00]
Total: 172887 W: 30934 L: 31385 D: 110568
sprt @ 10+0.1 th 1 Take 2, see git notes.
06-02-17 Ro PassedFileSimpl diff
LLR: -2.95 (-2.94,2.94) [-3.00,1.00]
Total: 89577 W: 16118 L: 16428 D: 57031
sprt @ 10+0.1 th 1 Do we need the PassedFile array ?
05-02-17 Ro PP_Resolution diff
LLR: -2.96 (-2.94,2.94) [-3.00,1.00]
Total: 77057 W: 13760 L: 14048 D: 49249
sprt @ 10+0.1 th 1 removing the passed array, using a simple 13 * rr instead and some compensations.