FuryBench Testing Framework

home toIntegralPlentyChessObsidianCaissaQuanticadeSerendipitySpaghetSchoenemannClarityIlluminaPawnocchioVineAnura
No logs for event exist
Active : 12 Machines / 210 Threads / 124.55 MNPS
Priority 0
Caissa Witek
root-sing-tweak
Time Manager: Tweak root singularity time fraction
diff60.0+0.60
ELO    2.20 +- 3.13 (95%)
LLR    0.98 (-2.94, 2.94) [-0.50, 2.50]
GAMES  N: 11048 W: 2624 L: 2554 D: 5870
PENTA  [1, 1244, 2967, 1308, 4]
Illumina Mergener
shared-corrhist
Turns out non-pawn corrhist was broken and probably losing so let's test illu without it
diff8.0+0.08
ELO    6.85 +- 19.56 (95%)
LLR    0.16 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 406 W: 94 L: 86 D: 226
PENTA  [7, 42, 90, 64, 0]
Integral Fury
teehee
diff10.0+0.10
ELO    -8.67 +- 11.37 (95%)
LLR    -0.42 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 882 W: 209 L: 231 D: 442
PENTA  [1, 115, 231, 93, 1]
Quanticade DarkNeutrino
alpha-raises-lmr
search.c: Reduce more based on alpha raises
diff60.0+0.60
ELO    -1.57 +- 2.97 (95%)
LLR    -1.41 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 11704 W: 2636 L: 2689 D: 6379
PENTA  [2, 1324, 3252, 1273, 1]
Priority -1
Illumina Mergener
qsearch-tt-cutoff
QSearch TT cutoff
diff8.0+0.08
ELO    1.65 +- 3.31 (95%)
LLR    0.28 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 12452 W: 2733 L: 2674 D: 7045
PENTA  [75, 1491, 3048, 1524, 88]
Illumina Mergener
nmp-cutnode
Only do NMP when cut_node
diff8.0+0.08
ELO    -0.67 +- 4.51 (95%)
LLR    -1.04 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 6770 W: 1483 L: 1496 D: 3791
PENTA  [45, 835, 1633, 832, 40]
Caissa Witek
qs-conthist-fix
QSearch: Fix uninitialized cont history pointers
diff10.0+0.10
ELO    -0.47 +- 1.77 (95%)
LLR    -1.22 (-2.94, 2.94) [0.00, 2.00]
GAMES  N: 38226 W: 8968 L: 9020 D: 20238
PENTA  [103, 4569, 9821, 4517, 103]
Priority -2
Illumina Mergener
capture-pruning
Capture pruning
diff8.0+0.08
ELO    -1.82 +- 8.82 (95%)
LLR    -0.42 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 1718 W: 366 L: 375 D: 977
PENTA  [10, 210, 429, 199, 11]
Illumina Mergener
reorder-qsearch
Reorder qsearch
diff8.0+0.08
ELO    -0.32 +- 4.28 (95%)
LLR    -0.97 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 7566 W: 1645 L: 1652 D: 4269
PENTA  [52, 921, 1846, 910, 54]
Finished
Integral Fury
teehee
diffN=20000
ELO    6.11 +- 3.53 (95%)
LLR    2.95 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 13980 W: 4083 L: 3837 D: 6060
PENTA  [215, 1594, 3173, 1746, 262]
Integral Fury
1024ti
smp diff20.0+0.20
ELO    1.37 +- 1.85 (95%)
LLR    1.05 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 30460 W: 7545 L: 7425 D: 15490
PENTA  [10, 3321, 8450, 3437, 12]
Integral Fury
pawnocchio_ppbig11
diffN=20000
ELO    7.17 +- 4.19 (95%)
GAMES  N: 10076/10000 W: 2980 L: 2772 D: 4324
PENTA  [136, 1178, 2302, 1186, 236]
Integral Fury
pawnocchio_ppbig11
diffN=20000
ELO    21.63 +- 21.27 (95%)
LLR    0.32 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 386 W: 116 L: 92 D: 178
PENTA  [7, 36, 79, 68, 3]
Illumina Mergener
shared-corrhist
This branch accidentally fixed a bug with non pawn corrhist and I need to check whether it gains
diff8.0+0.08
ELO    -4.73 +- 5.36 (95%)
LLR    -2.28 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 4998 W: 1120 L: 1188 D: 2690
PENTA  [54, 611, 1217, 583, 34]
Caissa Witek
root-sing-tweak
Time Manager: Tweak root singularity time fraction
crash diff10.0+0.10
ELO    -0.19 +- 1.25 (95%)
LLR    -1.79 (-2.94, 2.94) [0.00, 2.00]
GAMES  N: 78518 W: 18686 L: 18730 D: 41102
PENTA  [251, 9444, 19954, 9318, 292]
Illumina Mergener
higher-rfp-depth
Higher RFP max depth
diff8.0+0.08
ELO    -4.21 +- 5.01 (95%)
LLR    -2.29 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 5364 W: 1206 L: 1271 D: 2887
PENTA  [36, 670, 1331, 613, 32]
Integral Fury
1024ti
diff60.0+0.60
ELO    -3.05 +- 5.88 (95%)
LLR    -0.59 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 2966 W: 708 L: 734 D: 1524
PENTA  [1, 338, 830, 314, 0]
Integral Fury
1024ti
diff10.0+0.10
ELO    -7.28 +- 5.69 (95%)
LLR    -1.44 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 3628 W: 850 L: 926 D: 1852
PENTA  [9, 467, 933, 401, 4]
Integral Fury
1024ti
diffN=20000
ELO    16.45 +- 6.16 (95%)
LLR    2.95 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 4882 W: 1546 L: 1315 D: 2021
PENTA  [92, 488, 1058, 703, 100]
Quanticade DarkNeutrino
alpha-raises-lmr
search.c: Reduce more based on alpha raises
diff10.0+0.10
ELO    0.46 +- 1.17 (95%)
LLR    -1.18 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 83348 W: 19740 L: 19629 D: 43979
PENTA  [134, 9719, 21850, 9844, 127]
Illumina Mergener
fix-increments
Tuned values
diff40.0+0.40
ELO    7.00 +- 4.40 (95%)
LLR    2.90 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 6202 W: 1371 L: 1246 D: 3585
PENTA  [7, 699, 1578, 796, 21]
Caissa Witek
root-complexity-tm
diff60.0+0.60
ELO    1.47 +- 1.68 (95%)
LLR    1.93 (-2.94, 2.94) [-0.50, 2.50]
GAMES  N: 37396 W: 8794 L: 8636 D: 19966
PENTA  [21, 4085, 10328, 4243, 21]
Quanticade DarkNeutrino
mo-checks-good
search.c: Give move ordering bonus to quiet moves that give check
diff10.0+0.10
ELO    -2.67 +- 2.94 (95%)
LLR    -2.28 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 13384 W: 3088 L: 3191 D: 7105
PENTA  [23, 1628, 3491, 1529, 21]
Illumina Mergener
fix-increments
Tuned values
diff40.0+0.40
ELO    0.00 +- 0.00 (95%)
LLR    0.00 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 0 W: 0 L: 0 D: 0
PENTA  [0, 0, 0, 0, 0]