Stockfish Testing Framework

14 min read Original article ↗

Workers - 176 machines

Pending approval - 1 tests

Paused - 12 tests

25-12-06 sg lmr_cutnode_movecount3 diff 
  
  
    LLR: -0.77 (-2.94,2.94) <0.00,2.00>
      

    Total: 108608 W: 28195 L: 28103 D: 52310
      

    Ptnml(0-2): 370, 12739, 28020, 12779, 396
sprt @ 10+0.1 th 1

cores: 0 (0)

Double effect: More move count dependent reduction at CUT nodes.
25-12-10 sg lmr_cutnode_movecount7 diff 
  
  
    LLR: -1.07 (-2.94,2.94) <0.00,2.00>
      

    Total: 20064 W: 5148 L: 5215 D: 9701
      

    Ptnml(0-2): 82, 2452, 5021, 2405, 72
sprt @ 10+0.1 th 1

cores: 0 (0)

8 times effect: More move count dependent reduction at CUT nodes.
25-12-11 pb0 extendPromo diff 
  
  
    LLR: -0.28 (-2.94,2.94) <0.00,2.00>
      

    Total: 9056 W: 2309 L: 2320 D: 4427
      

    Ptnml(0-2): 21, 989, 2526, 964, 28
sprt @ 10+0.1 th 1

cores: 0 (0)

Take 1: Decrease reduction on successful Queen promotions at (reduced) null-move subtrees & verification searches (stopped because likely to small change)
25-12-11 pb0 weakMoveScore diff 
  
  
    LLR: -0.88 (-2.94,2.94) <0.00,2.00>
      

    Total: 5504 W: 1398 L: 1467 D: 2639
      

    Ptnml(0-2): 20, 666, 1439, 617, 10
sprt @ 10+0.1 th 1

cores: 0 (0)

Take 1: if weakMove -> nullMove -> arbitraryMove leads to nm-pruning then subsequently arbitraryMove -> nullMove -> weakMove should lead to unsatisfying positions too, thus down-rank/score this weak Move.
25-12-09 0x5 opt-div2 diff 
  
  
    LLR: -2.09 (-2.94,2.94) <0.00,2.00>
      

    Total: 110304 W: 28608 L: 28631 D: 53065
      

    Ptnml(0-2): 345, 12151, 30217, 12060, 379
sprt @ 10+0.1 th 1

cores: 0 (0)

disgusting micro-opts
25-12-04 pb0 breakDraw diff 
  
  
    LLR: -0.52 (-2.94,2.94) <0.50,2.50>
      

    Total: 3960 W: 972 L: 1005 D: 1983
      

    Ptnml(0-2): 3, 405, 1194, 378, 0
sprt @ 60+0.6 th 1

cores: 0 (0)

LTC Take 3: Take 2 is not performing so bad after all, so even try to loosen up the conditions a bit. Solves hard positions like 8/6pp/1K6/N5P1/3N4/8/npn1P3/k7 w - - bm Nab3+; 8/8/8/1B6/6p1/8/4KPpp/3N2kr w - - bm Kd3;
25-12-12 sg lmr_cutnode_movecount10 diff 
  
  
    LLR: -0.00 (-2.94,2.94) <0.50,2.50>
      

    Total: 115560 W: 29455 L: 29220 D: 56885
      

    Ptnml(0-2): 75, 12425, 32535, 12680, 65
sprt @ 60+0.6 th 1

cores: 0 (0)

LTC: Quadruple effect: Less move count dependent reduction at CUT nodes.
25-12-15 sg lmr_cutnode_movecount9d diff 
  
  
    LLR: -0.84 (-2.94,2.94) <0.50,2.50>
      

    Total: 77646 W: 19836 L: 19747 D: 38063
      

    Ptnml(0-2): 31, 8464, 21758, 8525, 45
sprt @ 60+0.6 th 1

cores: 0 (0)

LTC: Reduce reductions at CUT nodes sigmoid like dependent on move count and root depth (Effect increases with moveCount and decreases with rootDepth).
25-12-16 sg lmr_cutnode_movecount9c diff 
  
  
    LLR: -0.97 (-2.94,2.94) <0.50,2.50>
      

    Total: 37686 W: 9619 L: 9622 D: 18445
      

    Ptnml(0-2): 14, 4091, 10638, 4084, 16
sprt @ 60+0.6 th 1

cores: 0 (0)

LTC: Reduce reductions at CUT nodes sigmoid like dependent on move count and depth (Effect increases with moveCount and decreases with depth).
25-12-16 erg sc2 diff 
  
  
    LLR: -1.11 (-2.94,2.94) <0.00,2.00>
      

    Total: 48120 W: 12284 L: 12310 D: 23526
      

    Ptnml(0-2): 71, 5428, 13070, 5438, 53
sprt @ 5+0.05 th 8

cores: 0 (0)

take 2
25-12-16 erg sc3 diff 
  
  
    LLR: -0.77 (-2.94,2.94) <0.00,2.00>
      

    Total: 32288 W: 8267 L: 8287 D: 15734
      

    Ptnml(0-2): 42, 3655, 8770, 3635, 42
sprt @ 5+0.05 th 8

cores: 0 (0)

take 3
25-12-09 Viz histMext1 diff 
  
  
    LLR: -0.07 (-2.94,2.94) <0.50,2.50>
      

    Total: 47162 W: 12154 L: 12072 D: 22936
      

    Ptnml(0-2): 4, 4229, 15034, 4309, 5
sprt @ 60+0.6 th 8

cores: 0 (0)

LTC: Take 1

Failed - 0 tests

No failed tests on this page

Active - 39 tests

25-12-15 Ali Simp_fut_prun_2 diff 
  
  
    LLR: 2.12 (-2.94,2.94) <-1.75,0.25>
      

    Total: 198222 W: 50667 L: 50695 D: 96860
      

    Ptnml(0-2): 118, 21542, 55804, 21544, 103
sprt @ 60+0.6 th 1

cores: 46 (3)

LTC: Combine first, then multiply and divide later. A bit shorter logic.
25-12-18 sg lmr_scale_allnode2 diff 
  
  
    LLR: 2.01 (-2.94,2.94) <0.50,2.50>
      

    Total: 56442 W: 14531 L: 14251 D: 27660
      

    Ptnml(0-2): 35, 6000, 15873, 6276, 37
sprt @ 60+0.6 th 1

cores: 151 (13)

LTC: At ALL nodes: Scale up reduction with factor 9/8.
25-12-18 erg 17dec25e diff 
  
  
    LLR: 1.46 (-2.94,2.94) <0.00,2.00>
      

    Total: 98208 W: 25599 L: 25323 D: 47286
      

    Ptnml(0-2): 341, 11466, 25229, 11712, 356
sprt @ 10+0.1 th 1

cores: 38 (5)

take 5
25-12-17 ane rook-bird diff 
  
  
    LLR: 1.39 (-2.94,2.94) <0.00,2.00>
      

    Total: 144448 W: 37516 L: 37191 D: 69741
      

    Ptnml(0-2): 470, 15622, 39738, 15901, 493
sprt @ 10+0.1 th 1

cores: 48 (6)

pdep
25-12-18 Dis remote-sense diff 
  
  
    LLR: 1.18 (-2.94,2.94) <0.00,2.00>
      

    Total: 87392 W: 22642 L: 22406 D: 42344
      

    Ptnml(0-2): 322, 10260, 22274, 10540, 300
sprt @ 10+0.1 th 1

cores: 56 (3)

mhh
25-12-18 Viz trimSP2 diff 
  
  
    LLR: 0.83 (-2.94,2.94) <0.00,2.00>
      

    Total: 65376 W: 16956 L: 16784 D: 31636
      

    Ptnml(0-2): 250, 7628, 16761, 7798, 251
sprt @ 10+0.1 th 1

cores: 22 (3)

Take 2
25-12-16 sg lmr_scale_allnode diff 
  
  
    LLR: 0.75 (-2.94,2.94) <0.50,2.50>
      

    Total: 115752 W: 29762 L: 29467 D: 56523
      

    Ptnml(0-2): 53, 12324, 32847, 12579, 73
sprt @ 60+0.6 th 1

cores: 307 (10)

LTC: At ALL nodes: Scale up reduction with factor 17/16.
25-12-18 Viz trimSP1 diff 
  
  
    LLR: 0.55 (-2.94,2.94) <0.00,2.00>
      

    Total: 52224 W: 13656 L: 13529 D: 25039
      

    Ptnml(0-2): 197, 6135, 13316, 6272, 192
sprt @ 10+0.1 th 1

cores: 22 (4)

Take 1
25-12-19 fau insie2 diff 
  
  
    LLR: 0.51 (-2.94,2.94) <0.00,2.00>
      

    Total: 2400 W: 646 L: 600 D: 1154
      

    Ptnml(0-2): 3, 247, 655, 291, 4
sprt @ 10+0.1 th 1

cores: 8 (1)

insie2
25-12-13 sg lmr_cutnode_movecount9b diff 
  
  
    LLR: 0.35 (-2.94,2.94) <0.50,2.50>
      

    Total: 215136 W: 55004 L: 54538 D: 105594
      

    Ptnml(0-2): 115, 23141, 60608, 23571, 133
sprt @ 60+0.6 th 1

cores: 132 (8)

LTC: Reduce reductions at CUT nodes sigmoid like dependent on move count. Take 3.
25-12-19 fau insie3 diff 
  
  
    LLR: 0.34 (-2.94,2.94) <0.00,2.00>
      

    Total: 1728 W: 452 L: 420 D: 856
      

    Ptnml(0-2): 4, 188, 449, 218, 5
sprt @ 10+0.1 th 1

cores: 10 (2)

insie3
25-12-19 ane swan diff 
  
  
    LLR: 0.29 (-2.94,2.94) <0.00,2.00>
      

    Total: 29856 W: 7712 L: 7642 D: 14502
      

    Ptnml(0-2): 95, 3461, 7769, 3485, 118
sprt @ 10+0.1 th 1

cores: 30 (4)

weight (ss-2)->continuationCorrectionHistory more
25-12-19 Dis relaxed-atomics diff 
  
  
    LLR: 0.26 (-2.94,2.94) <-1.75,0.25>
      

    Total: 10240 W: 2668 L: 2657 D: 4915
      

    Ptnml(0-2): 35, 1106, 2830, 1111, 38
sprt @ 10+0.1 th 1

cores: 32 (4)

relaxed
25-12-19 fau insie5 diff 
  
  
    LLR: 0.22 (-2.94,2.94) <0.00,2.00>
      

    Total: 1856 W: 481 L: 458 D: 917
      

    Ptnml(0-2): 6, 215, 463, 238, 6
sprt @ 10+0.1 th 1

cores: 12 (3)

insie5
25-12-19 Tar bonus_previous diff 
  
  
    LLR: 0.13 (-2.94,2.94) <0.00,2.00>
      

    Total: 2400 W: 619 L: 604 D: 1177
      

    Ptnml(0-2): 7, 264, 644, 277, 8
sprt @ 10+0.1 th 1

cores: 24 (5)

take 3
25-12-19 rn5 dontReplaceNonQSEntries diff 
  
  
    LLR: 0.05 (-2.94,2.94) <0.00,2.00>
      

    Total: 5120 W: 1314 L: 1302 D: 2504
      

    Ptnml(0-2): 14, 615, 1293, 621, 17
sprt @ 10+0.1 th 1

cores: 22 (5)

High hash pressure
25-12-19 Dan dyn_cc_red_1 diff 
  
  
    LLR: -0.03 (-2.94,2.94) <0.00,2.00>
      

    Total: 11168 W: 2899 L: 2885 D: 5384
      

    Ptnml(0-2): 33, 1325, 2867, 1313, 46
sprt @ 10+0.1 th 1

cores: 42 (3)

take 1
25-12-19 fau insie7 diff 
  
  
    LLR: -0.05 (-2.94,2.94) <0.00,2.00>
      

    Total: 1440 W: 366 L: 368 D: 706
      

    Ptnml(0-2): 4, 176, 361, 176, 3
sprt @ 10+0.1 th 1

cores: 9 (3)

insie7
25-12-19 fau insie6 diff 
  
  
    LLR: -0.07 (-2.94,2.94) <0.00,2.00>
      

    Total: 1504 W: 376 L: 380 D: 748
      

    Ptnml(0-2): 3, 178, 397, 168, 6
sprt @ 10+0.1 th 1

cores: 8 (1)

insie6
25-12-17 Viz nmpRefBns3 diff 
  
  
    LLR: -0.12 (-2.94,2.94) <0.00,2.00>
      

    Total: 97344 W: 25296 L: 25163 D: 46885
      

    Ptnml(0-2): 312, 11453, 25048, 11508, 351
sprt @ 10+0.1 th 1

cores: 18 (4)

Take 3
25-12-19 fau insie4 diff 
  
  
    LLR: -0.13 (-2.94,2.94) <0.00,2.00>
      

    Total: 896 W: 227 L: 237 D: 432
      

    Ptnml(0-2): 3, 110, 232, 100, 3
sprt @ 10+0.1 th 1

cores: 8 (1)

insie4
25-12-15 fau nvds diff 
  
  
    LLR: -0.32 (-2.94,2.94) <-1.75,0.25>
      

    Total: 147990 W: 37708 L: 37885 D: 72397
      

    Ptnml(0-2): 66, 16260, 41543, 16037, 89
sprt @ 60+0.6 th 1

cores: 32 (4)

LTC: nvds
25-12-19 Ali Move_speed diff 
  
  
    LLR: -0.36 (-2.94,2.94) <0.50,2.50>
      

    Total: 5952 W: 1518 L: 1536 D: 2898
      

    Ptnml(0-2): 3, 683, 1626, 657, 7
sprt @ 60+0.6 th 1

cores: 128 (14)

LTC: STC finally passed. Hopefully it scales.
25-12-10 sg lmr_cutnode_movecount8 diff 
  
  
    LLR: -0.47 (-2.94,2.94) <0.00,2.00>
      

    Total: 136416 W: 35308 L: 35148 D: 65960
      

    Ptnml(0-2): 461, 16093, 34953, 16227, 474
sprt @ 10+0.1 th 1

cores: 16 (2)

More move count dependent reduction at CUT nodes (quadratic formula).
25-12-19 tso complex_sibling_reducti diff 
  
  
    LLR: -0.54 (-2.94,2.94) <0.00,2.00>
      

    Total: 12416 W: 3192 L: 3222 D: 6002
      

    Ptnml(0-2): 47, 1489, 3161, 1469, 42
sprt @ 10+0.1 th 1

cores: 31 (6)

Reducing siblings of score improvements depending on various factors (LLM generated)
25-12-19 rn5 miniTTEntry diff 
  
  
    LLR: -0.57 (-2.94,2.94) <0.00,2.00>
      

    Total: 8384 W: 2101 L: 2139 D: 4144
      

    Ptnml(0-2): 27, 1010, 2153, 978, 24
sprt @ 10+0.1 th 1

cores: 23 (2)

High hash pressure because thats where this would matter
25-12-12 fau insie1 diff 
  
  
    LLR: -0.73 (-2.94,2.94) <0.50,2.50>
      

    Total: 250416 W: 63931 L: 63482 D: 123003
      

    Ptnml(0-2): 134, 26923, 70641, 27380, 130
sprt @ 60+0.6 th 1

cores: 48 (4)

LTC: insie1
25-12-18 Viz trimSP3 diff 
  
  
    LLR: -0.77 (-2.94,2.94) <0.00,2.00>
      

    Total: 56000 W: 14486 L: 14472 D: 27042
      

    Ptnml(0-2): 206, 6589, 14381, 6633, 191
sprt @ 10+0.1 th 1

cores: 20 (4)

Take 3
25-12-19 Dan dyn_cc_red_2 diff 
  
  
    LLR: -0.78 (-2.94,2.94) <0.00,2.00>
      

    Total: 10720 W: 2724 L: 2777 D: 5219
      

    Ptnml(0-2): 28, 1301, 2761, 1236, 34
sprt @ 10+0.1 th 1

cores: 31 (4)

take 2 - without additional reduction in allnodes
25-12-18 Tar bonus_previous diff 
  
  
    LLR: -0.88 (-2.94,2.94) <0.00,2.00>
      

    Total: 59424 W: 15367 L: 15358 D: 28699
      

    Ptnml(0-2): 218, 6971, 15301, 7028, 194
sprt @ 10+0.1 th 1

cores: 20 (3)

take 1
25-12-19 rn5 ageTTShortlyBeforeOptim diff 
  
  
    LLR: -0.89 (-2.94,2.94) <0.00,2.00>
      

    Total: 3288 W: 790 L: 861 D: 1637
      

    Ptnml(0-2): 1, 417, 882, 340, 4
sprt @ 5+0.05 th 8

cores: 117 (7)

Also test this at SMP with little higher hash pressure since with more threads there may be more random entries from random tree parts sitting around that are no longer used
25-12-15 sg lmr_cutnode_movecount9c diff 
  
  
    LLR: -0.90 (-2.94,2.94) <0.00,2.00>
      

    Total: 22304 W: 5794 L: 5841 D: 10669
      

    Ptnml(0-2): 72, 2661, 5741, 2598, 80
sprt @ 10+0.1 th 1

cores: 20 (3)

Reduce reductions at CUT nodes sigmoid like dependent on move count and depth (depth offset = 5).
25-12-14 Viz nmpRefBns1 diff 
  
  
    LLR: -1.12 (-2.94,2.94) <0.00,2.00>
      

    Total: 184192 W: 47689 L: 47517 D: 88986
      

    Ptnml(0-2): 630, 21527, 47608, 21703, 628
sprt @ 10+0.1 th 1

cores: 17 (3)

Take 1
25-12-18 Tar bonus_previous diff 
  
  
    LLR: -1.21 (-2.94,2.94) <0.00,2.00>
      

    Total: 61792 W: 15977 L: 15994 D: 29821
      

    Ptnml(0-2): 252, 7280, 15819, 7323, 222
sprt @ 10+0.1 th 1

cores: 28 (4)

take 2
25-12-15 pb0 weakMoveScore2 diff 
  
  
    LLR: -1.45 (-2.94,2.94) <0.00,2.00>
      

    Total: 203328 W: 52698 L: 52527 D: 98103
      

    Ptnml(0-2): 705, 23866, 52361, 24017, 715
sprt @ 10+0.1 th 1

cores: 29 (3)

Take 7: like take 5 but use vector instead of array[4]
25-12-19 rn5 ageTTShortlyBeforeOptim diff 
  
  
    LLR: -1.84 (-2.94,2.94) <0.00,2.00>
      

    Total: 5696 W: 1396 L: 1554 D: 2746
      

    Ptnml(0-2): 26, 743, 1466, 589, 24
sprt @ 10+0.1 th 1

cores: 23 (5)

High hash pressure
25-12-15 pb0 weakMoveScore2 diff 
  
  
    LLR: -2.19 (-2.94,2.94) <0.00,2.00>
      

    Total: 232000 W: 60178 L: 60032 D: 111790
      

    Ptnml(0-2): 807, 26844, 60523, 27048, 778
sprt @ 10+0.1 th 1

cores: 30 (4)

Take 6: apply when nm-search was done with qsearch (depth - R <= 0)
25-12-11 fau insie diff 
  
  
    LLR: -2.28 (-2.94,2.94) <0.50,2.50>
      

    Total: 231006 W: 58861 L: 58578 D: 113567
      

    Ptnml(0-2): 140, 24800, 65357, 25049, 157
sprt @ 60+0.6 th 1

cores: 64 (6)

LTC: insie
25-12-03 fau wave5 diff 
  
  
    70275/100000 iterations
      

    140550/200000 games played
200000 @ 60+0.6 th 8

cores: 95 (7)

wave5

Finished - 179217 tests

25-12-19 ane relaxed-atomics diff 
  
  
    LLR: 0.09 (-2.94,2.94) <-1.75,0.25>
      

    Total: 800 W: 213 L: 206 D: 381
      

    Ptnml(0-2): 3, 81, 222, 94, 0
sprt @ 10+0.1 th 1 Placate TSAN? (non-regression bounds intended oops)
25-12-17 Ali Move_speed diff 
  
  
    LLR: 2.96 (-2.94,2.94) <0.00,2.00>
      

    Total: 151456 W: 39468 L: 38979 D: 73009
      

    Ptnml(0-2): 529, 17667, 38846, 18158, 528
sprt @ 10+0.1 th 1 Now, count for probcut and null moves.
25-12-19 ane relaxed-atomics diff 
  
  
    LLR: 0.00 (-2.94,2.94) <0.00,2.00>
      

    Total: 0 W: 0 L: 0 D: 0
      

    Ptnml(0-2): 0, 0, 0, 0, 0
sprt @ 10+0.1 th 1 Placate TSAN?
25-12-19 ane relaxed-atomics diff 
  
  
    LLR: 0.00 (-2.94,2.94) <-1.75,0.25>
      

    Total: 0 W: 0 L: 0 D: 0
      

    Ptnml(0-2): 0, 0, 0, 0, 0
sprt @ 10+0.1 th 1 Placate TSAN?
25-12-17 fau insiet diff 
  
  
    40000/40000 iterations
      

    80000/80000 games played
80000 @ 10+0.1 th 1 insiet
25-12-19 ane common-snipe diff 
  
  
    LLR: 0.00 (-2.94,2.94) <0.00,2.00>
      

    Total: 640 W: 171 L: 170 D: 299
      

    Ptnml(0-2): 0, 66, 188, 65, 1
sprt @ 10+0.1 th 1 prevent tte key16 from being cached
25-12-17 Viz nmpRefBns2 diff 
  
  
    LLR: -2.93 (-2.94,2.94) <0.00,2.00>
      

    Total: 79264 W: 20488 L: 20632 D: 38144
      

    Ptnml(0-2): 242, 9505, 20311, 9303, 271
sprt @ 10+0.1 th 1 Take 2
25-12-18 ane coot diff 
  
  
    LLR: -2.17 (-2.94,2.94) <0.00,2.00>
      

    Total: 68416 W: 17599 L: 17688 D: 33129
      

    Ptnml(0-2): 224, 7556, 18743, 7455, 230
sprt @ 10+0.1 th 1 bitbase only for wrong colored pawns
25-12-19 ane loon-pr diff 
  
  
    LLR: -0.53 (-2.94,2.94) <0.50,2.50>
      

    Total: 12540 W: 3191 L: 3208 D: 6141
      

    Ptnml(0-2): 2, 1294, 3696, 1275, 3
sprt @ 60+0.6 th 2 LTC: See benefit or lack thereof with only 2 threads
25-12-14 sg lmr_cutnode_movecount9c diff 
  
  
    LLR: 2.93 (-2.94,2.94) <0.00,2.00>
      

    Total: 90976 W: 23687 L: 23293 D: 43996
      

    Ptnml(0-2): 301, 10499, 23504, 10873, 311
sprt @ 10+0.1 th 1 Reduce reductions at CUT nodes sigmoid like dependent on move count and depth (depth offset = 3).
25-12-18 Dan cn_h_3 diff 
  
  
    LLR: -2.00 (-2.94,2.94) <0.50,2.50>
      

    Total: 16236 W: 4101 L: 4231 D: 7904
      

    Ptnml(0-2): 9, 1817, 4592, 1695, 5
sprt @ 60+0.6 th 1 LTC: take 3 - speculative
25-12-15 pb0 weakMoveScore2 diff 
  
  
    LLR: -2.94 (-2.94,2.94) <0.00,2.00>
      

    Total: 178272 W: 46218 L: 46217 D: 85837
      

    Ptnml(0-2): 598, 20994, 45940, 21017, 587
sprt @ 10+0.1 th 1 Take 5: inverse logic of take 4
25-12-18 Dan e_d_mo_2 diff 
  
  
    LLR: -2.95 (-2.94,2.94) <0.00,2.00>
      

    Total: 44864 W: 11437 L: 11635 D: 21792
      

    Ptnml(0-2): 162, 5398, 11529, 5162, 181
sprt @ 10+0.1 th 1 take 2
25-12-17 erg sc1 diff 
  
  
    LLR: -2.25 (-2.94,2.94) <0.50,2.50>
      

    Total: 38538 W: 9766 L: 9868 D: 18904
      

    Ptnml(0-2): 13, 3965, 11421, 3851, 19
sprt @ 20+0.2 th 8 LTC: take 1
25-12-19 ane loon-pr diff 
  
  
    LLR: 0.00 (-2.94,2.94) <0.50,2.50>
      

    Total: 0 W: 0 L: 0 D: 0
      

    Ptnml(0-2): 0, 0, 0, 0, 0
sprt @ 60+0.6 th 2 LTC: See benefit or lack thereof with only two threads
25-12-15 sg lmr_cutnode_movecount9c diff 
  
  
    LLR: -2.95 (-2.94,2.94) <0.00,2.00>
      

    Total: 97184 W: 25187 L: 25306 D: 46691
      

    Ptnml(0-2): 325, 11510, 25050, 11373, 334
sprt @ 10+0.1 th 1 Reduce reductions at CUT nodes sigmoid like dependent on move count and depth (depth offset = 1).
25-12-18 erg 17dec25b diff 
  
  
    LLR: -2.76 (-2.94,2.94) <0.00,2.00>
      

    Total: 30848 W: 7826 L: 8026 D: 14996
      

    Ptnml(0-2): 110, 3706, 7987, 3516, 105
sprt @ 10+0.1 th 1 take 2
25-12-18 erg 17dec25a diff 
  
  
    LLR: -2.07 (-2.94,2.94) <0.00,2.00>
      

    Total: 36480 W: 9347 L: 9477 D: 17656
      

    Ptnml(0-2): 117, 4351, 9437, 4215, 120
sprt @ 10+0.1 th 1 take 1
25-12-18 Dan rm_sn diff 
  
  
    LLR: -2.94 (-2.94,2.94) <-1.75,0.25>
      

    Total: 22752 W: 5717 L: 6008 D: 11027
      

    Ptnml(0-2): 106, 2823, 5780, 2590, 77
sprt @ 10+0.1 th 1 try to remove smallnet
25-12-18 fau tncc1 diff 
  
  
    LLR: -1.55 (-2.94,2.94) <0.00,2.00>
      

    Total: 11584 W: 2927 L: 3048 D: 5609
      

    Ptnml(0-2): 35, 1434, 2978, 1307, 38
sprt @ 10+0.1 th 1 tncc1
25-12-17 fau tncc diff 
  
  
    LLR: -2.96 (-2.94,2.94) <0.00,2.00>
      

    Total: 39712 W: 10197 L: 10403 D: 19112
      

    Ptnml(0-2): 129, 4839, 10148, 4589, 151
sprt @ 10+0.1 th 1 tncc
25-12-18 jay classify diff 
  
  
    LLR: 0.00 (-2.94,2.94) <0.00,2.00>
      

    Total: 0 W: 0 L: 0 D: 0
      

    Ptnml(0-2): 0, 0, 0, 0, 0
sprt @ 10+0.1 th 1 Quick check of complexity/futility idea
25-12-18 jay movepick-opt2 diff 
  
  
    LLR: -0.46 (-2.94,2.94) <0.00,2.00>
      

    Total: 4096 W: 1038 L: 1073 D: 1985
      

    Ptnml(0-2): 14, 493, 1069, 458, 14
sprt @ 10+0.1 th 1 This boosts NPS on my local Intel Alder Lake, see how it performs.
25-12-12 Dan cn_h_3 diff 
  
  
    LLR: -2.97 (-2.94,2.94) <0.00,2.00>
      

    Total: 360608 W: 93601 L: 93333 D: 173674
      

    Ptnml(0-2): 1181, 42581, 92558, 42757, 1227
sprt @ 10+0.1 th 1 take 3
25-12-16 ane verbatim-tables diff 
  
  
    LLR: 2.93 (-2.94,2.94) <0.00,2.00>
      

    Total: 136832 W: 35509 L: 35060 D: 66263
      

    Ptnml(0-2): 445, 14885, 37341, 15266, 479
sprt @ 10+0.1 th 1 Verbatim RookAttacks and BishopAttacks bc why not