FuryBench Testing Framework

home toIntegralPlentyChessObsidianCaissaQuanticadeSerendipitySpaghetSchoenemannClarityIlluminaPawnocchioVineAnura
Finished
Illumina Mergener
90-epochs
90 epochs net
diff8.0+0.08
ELO    4.17 +- 3.18 (95%)
LLR    2.91 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 13496 W: 3055 L: 2893 D: 7548
PENTA  [86, 1546, 3332, 1688, 96]
Caissa Witek
tt_replace_gen_2
diff8.0+0.08
ELO    1.69 +- 1.26 (95%)
LLR    2.95 (-2.94, 2.94) [0.00, 2.00]
GAMES  N: 78070 W: 18568 L: 18188 D: 41314
PENTA  [292, 9016, 20045, 9384, 298]
Caissa Witek
eval-ml-6-160B
eval-ml-6-160B
diff10.0+0.10
ELO    3.30 +- 1.92 (95%)
LLR    3.00 (-2.94, 2.94) [0.00, 2.00]
GAMES  N: 33440 W: 8016 L: 7698 D: 17726
PENTA  [109, 3846, 8487, 4174, 104]
PlentyChess Yoshie2000
0189r
diffN=20000
ELO    11.76 +- 4.66 (95%)
LLR    3.00 (-2.25, 2.89) [0.00, 2.50]
GAMES  N: 8128 W: 2441 L: 2166 D: 3521
PENTA  [108, 920, 1796, 1069, 171]
Caissa Witek
multilayer
Add missing AVX-512 and NEON inference
frc diff10.0+0.10
ELO    4.95 +- 3.02 (95%)
LLR    2.95 (-2.94, 2.94) [-0.50, 2.50]
GAMES  N: 14182 W: 2766 L: 2564 D: 8852
PENTA  [131, 1369, 3873, 1603, 115]
Caissa Witek
multilayer
Add missing AVX-512 and NEON inference
diff20.0+0.10
ELO    53.67 +- 7.28 (95%)
LLR    2.95 (-2.94, 2.94) [0.00, 2.00]
GAMES  N: 1866 W: 628 L: 342 D: 896
PENTA  [0, 89, 469, 375, 0]
Caissa Witek
multilayer
Switch to multilayer net
diff60.0+0.60
ELO    20.58 +- 6.13 (95%)
LLR    2.96 (-2.94, 2.94) [0.00, 3.00]
GAMES  N: 3042 W: 788 L: 608 D: 1646
PENTA  [7, 269, 784, 459, 2]
Caissa Witek
multilayer
Switch to multilayer net
diffN=25000
ELO    29.93 +- 7.14 (95%)
LLR    2.97 (-2.94, 2.94) [0.00, 2.00]
GAMES  N: 4282 W: 1506 L: 1138 D: 1638
PENTA  [95, 393, 860, 635, 158]
Caissa Witek
multilayer
Switch to multilayer net
diff10.0+0.10
ELO    18.21 +- 4.88 (95%)
LLR    2.95 (-2.94, 2.94) [0.00, 2.00]
GAMES  N: 5310 W: 1427 L: 1149 D: 2734
PENTA  [17, 516, 1322, 772, 28]
Integral Fury
main
Add support for DFRC (#249)---------Co-authored-by: Quinniboi10 <69563433+Quinniboi10@users.noreply.github.com>
diff40.0+0.40
ELO    109.63 +- 4.28 (95%)
GAMES  N: 5002/5000 W: 2014 L: 486 D: 2502
PENTA  [2, 82, 821, 1578, 18]
Integral Fury
main
Add support for DFRC (#249)---------Co-authored-by: Quinniboi10 <69563433+Quinniboi10@users.noreply.github.com>
diff10.0+0.10
ELO    99.16 +- 4.91 (95%)
GAMES  N: 5002/5000 W: 2016 L: 626 D: 2360
PENTA  [3, 164, 842, 1424, 68]
Integral Fury
smp_spsa
smp diff20.0+0.20
ELO    5.64 +- 3.45 (95%)
LLR    2.90 (-2.25, 2.89) [0.00, 4.00]
GAMES  N: 8928 W: 2349 L: 2204 D: 4375
PENTA  [3, 937, 2439, 1082, 3]
Integral Fury
smp_spsa
smp diff20.0+0.20
ELO    16.37 +- 5.43 (95%)
LLR    2.94 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 3674 W: 1021 L: 848 D: 1805
PENTA  [2, 334, 997, 497, 7]
Illumina Mergener
gen7b-60epochs
diff8.0+0.08
ELO    4.37 +- 3.31 (95%)
LLR    2.90 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 12968 W: 3111 L: 2948 D: 6909
PENTA  [97, 1515, 3095, 1682, 95]
Integral Fury
shared_correction_history
smp diff10.0+0.10
ELO    6.88 +- 3.43 (95%)
LLR    2.91 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 9198 W: 2338 L: 2156 D: 4704
PENTA  [6, 957, 2496, 1129, 11]
Illumina Mergener
neg-extensions
Negative extensions
diff40.0+0.40
ELO    2.94 +- 2.36 (95%)
LLR    2.91 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 20662 W: 4274 L: 4099 D: 12289
PENTA  [39, 2325, 5429, 2498, 40]
Integral Fury
smp_spsa
smp diff20.0+0.20
ELO    8.98 +- 3.95 (95%)
LLR    2.91 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 6772 W: 1787 L: 1612 D: 3373
PENTA  [3, 670, 1870, 835, 8]
Quanticade DarkNeutrino
se-tt-move-exists
search.c: Verify that TT move actually exists before going into SE WTF is wrong with me
diff10.0+0.10
ELO    0.79 +- 1.77 (95%)
LLR    3.00 (-2.94, 2.94) [-3.00, 1.00]
GAMES  N: 34242 W: 8166 L: 8088 D: 17988
PENTA  [59, 3700, 9510, 3808, 44]
Integral Fury
hmc_inputs_no_kb
diff40.0+0.40
ELO    4.66 +- 3.65 (95%)
LLR    2.91 (-2.25, 2.89) [-3.00, 1.00]
GAMES  N: 8356 W: 2091 L: 1979 D: 4286
PENTA  [5, 922, 2219, 1020, 12]
Integral Fury
hmc_inputs_no_kb
diffN=25000
ELO    0.10 +- 1.53 (95%)
LLR    3.02 (-2.25, 2.89) [-3.00, 1.00]
GAMES  N: 78326 W: 22638 L: 22615 D: 33073
PENTA  [1448, 9559, 17253, 9328, 1575]
Integral Anematode
woozy
woozy
diff10.0+0.10
ELO    4.21 +- 2.55 (95%)
LLR    2.94 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 15926 W: 4075 L: 3882 D: 7969
PENTA  [25, 1564, 4606, 1729, 39]
Caissa Witek
eval-85-203B
Move Orderer: Fix lack of history bonus for overflown best move ELO 0.24 +- 1.43 (95%) SPRT 10.0+0.10s Threads=1 Hash=16MB LLR 2.96 (-2.94, 2.94) [-3.00, 0.00] GAMES N: 60648 W: 14199 L: 14157 D: 32292 PENTA [222, 7191, 15421, 7303, 187]
diff8.0+0.08
ELO    6.56 +- 2.76 (95%)
LLR    2.95 (-2.94, 2.94) [0.00, 2.00]
GAMES  N: 14200 W: 3810 L: 3542 D: 6848
PENTA  [10, 1494, 3818, 1774, 4]
Integral Fury
hmc_inputs_and_scaling
diff10.0+0.10
ELO    3.66 +- 1.93 (95%)
LLR    2.96 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 10908 W: 1981 L: 1866 D: 7061
PENTA  [6, 384, 4572, 473, 19]
Quanticade DarkNeutrino
pp
nnue.c: Change eval scale back
diff60.0+0.60
ELO    2.42 +- 1.81 (95%)
LLR    2.90 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 33250 W: 7981 L: 7749 D: 17520
PENTA  [16, 3701, 8963, 3925, 20]
Caissa Witek
hist-bonus-overflow
diff10.0+0.10
ELO    0.24 +- 1.43 (95%)
LLR    2.96 (-2.94, 2.94) [-3.00, 0.00]
GAMES  N: 60648 W: 14199 L: 14157 D: 32292
PENTA  [222, 7191, 15421, 7303, 187]