SolsticeBench

Finished
TestTCEngineAuthorResult
9B5E3542A949DA1C SPRT diff
search: no check extension
4.0+0.04SolsticeDragonLLR: -0.47 (-2.94, 2.94) [0.00, 5.00]
Games: 692 W: 166 L: 183 D: 343
Ptnml(0-2): 15, 88, 149, 87, 7
1FA22E18C7C43F32 SPRT diff
search: draw jitter +-1
4.0+0.04SolsticeDragonLLR: -0.34 (-2.94, 2.94) [0.00, 5.00]
Games: 448 W: 101 L: 112 D: 235
Ptnml(0-2): 5, 55, 113, 48, 3
C0465EF85387A587 SPRT diff
search: fail-high blend (best*d+beta)/(d+1)
4.0+0.04SolsticeDragonLLR: -0.30 (-2.94, 2.94) [0.00, 5.00]
Games: 768 W: 166 L: 175 D: 427
Ptnml(0-2): 12, 92, 182, 89, 9
E794FF4E0EB25D11 SPRT diff
search: 4-ply conthist (half weight) in quiet ordering
4.0+0.04SolsticeDragonLLR: -0.22 (-2.94, 2.94) [0.00, 5.00]
Games: 362 W: 87 L: 95 D: 180
Ptnml(0-2): 7, 50, 72, 48, 4
48E9152746D05BF9 SPRT diff
qsearch: fail-high blend (score+beta)/2 [vs qs-failsoft]
4.0+0.04SolsticeDragonLLR: -0.87 (-2.94, 2.94) [0.00, 5.00]
Games: 1536 W: 344 L: 373 D: 819
Ptnml(0-2): 19, 204, 350, 177, 18
D159D54765BDDF52 SPRT diff
search: TT-cutoff quiet history bonus
4.0+0.04SolsticeDragonLLR: 0.34 (-2.94, 2.94) [0.00, 5.00]
Games: 6450 W: 1556 L: 1513 D: 3381
Ptnml(0-2): 93, 801, 1407, 818, 106
3E95002E7D24EFF3 SPRT diff
search: doDeeper/doShallower on LMR re-search
4.0+0.04SolsticeDragonLLR: -0.59 (-2.94, 2.94) [0.00, 5.00]
Games: 1296 W: 312 L: 331 D: 653
Ptnml(0-2): 20, 166, 293, 151, 18
D53EF2544CE16585 SPRT diff
tm: soft limit scaled by score drop over 2 iterations
4.0+0.04SolsticeDragonLLR: -0.55 (-2.94, 2.94) [0.00, 5.00]
Games: 1666 W: 400 L: 415 D: 851
Ptnml(0-2): 25, 189, 414, 186, 19
5A2C318D9B82C83B SPRT diff
qsearch: TT-adjusted stand pat
4.0+0.04SolsticeDragonLLR: 3.06 (-2.94, 2.94) [0.00, 5.00]
Games: 6664 W: 1698 L: 1539 D: 3427
Ptnml(0-2): 92, 765, 1488, 866, 121
D7FE0E882A3EDE10 SPRT diff
search: LMR for SEE-losing captures (quiet formula - 1)
4.0+0.04SolsticeDragonLLR: -0.86 (-2.94, 2.94) [0.00, 5.00]
Games: 980 W: 218 L: 251 D: 511
Ptnml(0-2): 18, 138, 208, 111, 15
57F1E628CA44E81A SPRT diff
search: depth-1 after alpha raise (depth 3..13)
4.0+0.04SolsticeDragonLLR: -0.86 (-2.94, 2.94) [0.00, 5.00]
Games: 1658 W: 383 L: 413 D: 862
Ptnml(0-2): 37, 203, 367, 197, 25
E75E63C05A1476AF SPRT diff
search: prior countermove bonus on fail low
4.0+0.04SolsticeDragonLLR: -1.05 (-2.94, 2.94) [0.00, 5.00]
Games: 1338 W: 319 L: 358 D: 661
Ptnml(0-2): 26, 177, 293, 156, 17
18E91AC1F8B141FF SPRT diff
search: persistent history across moves (halved per search)
4.0+0.04SolsticeDragonLLR: 3.05 (-2.94, 2.94) [0.00, 5.00]
Games: 3598 W: 919 L: 778 D: 1901
Ptnml(0-2): 42, 378, 833, 489, 57
B1945B4743DED73B SPRT diff
qsearch: fail-soft move loop
4.0+0.04SolsticeDragonLLR: 2.97 (-2.94, 2.94) [0.00, 5.00]
Games: 1520 W: 432 L: 303 D: 785
Ptnml(0-2): 9, 145, 352, 216, 38
6B6AA75B4F888BF1 SPRT diff
search: countermove picker stage after killers
4.0+0.04SolsticeDragonLLR: -0.45 (-2.94, 2.94) [0.00, 5.00]
Games: 7808 W: 1854 L: 1839 D: 4115
Ptnml(0-2): 114, 954, 1749, 977, 110
hindsight SPRT diff
search: hindsight depth adjustment from the parent's LMR reduction Co-Authored-By: Claude Opus 5.5 <[email protected]>
4.0+0.04SolsticeDragonLLR: 0.83 (-2.94, 2.94) [0.00, 5.00]
Games: 25288 W: 5991 L: 5850 D: 13447
Ptnml(0-2): 307, 3053, 5819, 3122, 343
small-probcut SPRT diff
search: small ProbCut, trust a TT lower bound far above beta Co-Authored-By: Claude Opus 5.5 <[email protected]>
4.0+0.04SolsticeDragonLLR: -2.37 (-2.94, 2.94) [0.00, 5.00]
Games: 20100 W: 4781 L: 4793 D: 10526
Ptnml(0-2): 294, 2368, 4695, 2442, 251
nmp-eval SPRT diff
search: NMP reduces more the further the eval is above beta Co-Authored-By: Claude Opus 5.5 <[email protected]>
4.0+0.04SolsticeDragonLLR: -1.46 (-2.94, 2.94) [0.00, 5.00]
Games: 3072 W: 694 L: 743 D: 1635
Ptnml(0-2): 51, 396, 696, 337, 56
eval-diff-hist SPRT diff
search: eval-difference bonus to the parent's quiet move history Co-Authored-By: Claude Opus 5.5 <[email protected]>
4.0+0.04SolsticeDragonLLR: -3.05 (-2.94, 2.94) [0.00, 5.00]
Games: 1948 W: 439 L: 563 D: 946
Ptnml(0-2): 42, 279, 445, 177, 31
net-kb4-sprt SPRT diff
net: gen45 kb4 (4 mirrored king buckets, 768x4-256-1, gen4+gen5 data, 40-epoch cosine) Co-Authored-By: Claude Opus 5.5 <[email protected]>
4.0+0.04SolsticeDragonLLR: -2.96 (-2.94, 2.94) [0.00, 5.00]
Games: 958 W: 204 L: 338 D: 416
Ptnml(0-2): 48, 147, 181, 97, 6
hindsight SPRT diff
search: hindsight depth adjustment from the parent's LMR reduction Co-Authored-By: Claude Opus 5.5 <[email protected]>
4.0+0.04SolsticeDragonLLR: 0.00 (-2.94, 2.94) [0.00, 5.00]
Games: 0 W: 0 L: 0 D: 0
Ptnml(0-2): 0, 0, 0, 0, 0
eval-diff-hist SPRT diff
search: eval-difference bonus to the parent's quiet move history Co-Authored-By: Claude Opus 5.5 <[email protected]>
4.0+0.04SolsticeDragonLLR: 0.00 (-2.94, 2.94) [0.00, 5.00]
Games: 0 W: 0 L: 0 D: 0
Ptnml(0-2): 0, 0, 0, 0, 0
small-probcut SPRT diff
search: small ProbCut, trust a TT lower bound far above beta Co-Authored-By: Claude Opus 5.5 <[email protected]>
4.0+0.04SolsticeDragonLLR: 0.00 (-2.94, 2.94) [0.00, 5.00]
Games: 0 W: 0 L: 0 D: 0
Ptnml(0-2): 0, 0, 0, 0, 0
nmp-eval SPRT diff
search: NMP reduces more the further the eval is above beta Co-Authored-By: Claude Opus 5.5 <[email protected]>
4.0+0.04SolsticeDragonLLR: 0.00 (-2.94, 2.94) [0.00, 5.00]
Games: 0 W: 0 L: 0 D: 0
Ptnml(0-2): 0, 0, 0, 0, 0
threat-history SPRT diff
Threat-aware quiet history: [color][from attacked*2 + to attacked][from][to]; attack map once per node.
4.0+0.04SolsticeDragonLLR: 0.41 (-2.94, 2.94) [0.00, 5.00]
Games: 27838 W: 6038 L: 5908 D: 15892
Ptnml(0-2): 281, 3298, 6641, 3408, 291