FuryBench Testing Framework

home toIntegralPlentyChessObsidianCaissaQuanticadeSerendipitySpaghetSchoenemannClarityIlluminaPawnocchioVineAnura
Finished
Pawnocchio SwedishChef
mixed_data_longer_finetune_s1.nnue
diff10.0+0.10
ELO    -12.56 +- 5.93 (95%)
LLR    -2.25 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 3542 W: 798 L: 926 D: 1818
PENTA  [11, 499, 878, 373, 10]
Pawnocchio SwedishChef
mixed_data_longer_finetune.nnue
diff10.0+0.10
ELO    -0.76 +- 2.00 (95%)
LLR    -2.27 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 30644 W: 7411 L: 7478 D: 15755
PENTA  [86, 3750, 7718, 3681, 87]
Vine SwedishChef
soa
diffN=1000
ELO    -0.00 +- 0.00 (95%)
LLR    0.63 (-2.25, 2.89) [-5.00, 0.00]
GAMES  N: 3072 W: 997 L: 997 D: 1078
PENTA  [0, 0, 1536, 0, 0]
Vine SwedishChef
soa
crash diffN=1000
ELO    -1199.83 +- 0.00 (95%)
LLR    -10.31 (-2.25, 2.89) [-5.00, 0.00]
GAMES  N: 1024 W: 0 L: 1024 D: 0
PENTA  [512, 0, 0, 0, 0]
Vine SwedishChef
policy_refactor
crash diffN=10000
ELO    -1199.83 +- 0.00 (95%)
LLR    -10.31 (-2.25, 2.89) [-5.00, 0.00]
GAMES  N: 1024 W: 0 L: 1024 D: 0
PENTA  [512, 0, 0, 0, 0]
Pawnocchio SwedishChef
2048hl
diff40.0+0.40
ELO    -1.26 +- 2.44 (95%)
LLR    -1.90 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 18750 W: 4391 L: 4459 D: 9900
PENTA  [15, 2224, 4966, 2154, 16]
Pawnocchio SwedishChef
2048hl
diff10.0+0.10
ELO    -0.58 +- 1.86 (95%)
LLR    -2.32 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 35568 W: 8614 L: 8673 D: 18281
PENTA  [103, 4332, 8976, 4267, 106]
Pawnocchio SwedishChef
2048hl
diffN=20000
ELO    23.50 +- 7.74 (95%)
LLR    2.97 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 3702 W: 1275 L: 1025 D: 1402
PENTA  [84, 375, 736, 519, 137]
Pawnocchio SwedishChef
mixed_data_01_03.nnue
diff10.0+0.10
ELO    10.56 +- 4.43 (95%)
LLR    2.97 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 6286 W: 1585 L: 1394 D: 3307
PENTA  [6, 695, 1574, 838, 30]
Pawnocchio SwedishChef
mixed_data_01_03.nnue
diff40.0+0.40
ELO    4.13 +- 2.56 (95%)
LLR    2.92 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 16986 W: 4100 L: 3898 D: 8988
PENTA  [4, 1917, 4461, 2095, 16]
Pawnocchio SwedishChef
512_testnet_vine_01_02_data75_piece_relabel.nnue
diff40.0+0.40
ELO    7.99 +- 3.89 (95%)
LLR    2.90 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 8526 W: 2250 L: 2054 D: 4222
PENTA  [13, 993, 2098, 1103, 56]
Pawnocchio SwedishChef
vine_data_test_adjustible_scale
diff10.0+0.10
ELO    -6.29 +- 4.82 (95%)
LLR    -1.85 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 5522 W: 1279 L: 1379 D: 2864
PENTA  [28, 709, 1381, 621, 22]
Pawnocchio SwedishChef
512_testnet_vine_01_02_data75_piece_relabel.nnue
diff10.0+0.10
ELO    19.34 +- 6.61 (95%)
LLR    2.96 (-2.25, 2.89) [0.00, 3.00]
GAMES  N: 3830 W: 1160 L: 947 D: 1723
PENTA  [19, 439, 860, 504, 93]
Vine SwedishChef
faster_inverse_sigmoid
diff2.0+0.02
ELO    -1.69 +- 3.85 (95%)
LLR    -2.25 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 9888 W: 2061 L: 2109 D: 5718
PENTA  [99, 1180, 2441, 1118, 106]
Vine SwedishChef
pinputs
diff10.0+0.10
ELO    -1.72 +- 4.03 (95%)
LLR    -2.29 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 10330 W: 2439 L: 2490 D: 5401
PENTA  [147, 1297, 2333, 1236, 152]
Vine SwedishChef
pinputs
diff10.0+0.10
ELO    -171.31 +- 27.73 (95%)
LLR    -2.38 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 346 W: 36 L: 194 D: 116
PENTA  [40, 87, 38, 7, 1]
Vine SwedishChef
combined_v55_p20
diff10.0+0.10
ELO    7.06 +- 4.63 (95%)
LLR    2.92 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 7332 W: 1754 L: 1605 D: 3973
PENTA  [77, 853, 1666, 984, 86]
Vine SwedishChef
simd_puct_32bit
diff2.0+0.02
ELO    -4.26 +- 4.90 (95%)
LLR    -2.59 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 6286 W: 1410 L: 1487 D: 3389
PENTA  [88, 731, 1561, 696, 67]
Vine SwedishChef
simd_puct
diff2.0+0.02
ELO    -1.18 +- 3.41 (95%)
LLR    -2.54 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 13596 W: 3113 L: 3159 D: 7324
PENTA  [179, 1621, 3251, 1561, 186]
Vine SwedishChef
simd_puct
diff2.0+0.02
ELO    4.18 +- 3.16 (95%)
LLR    3.02 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 15300 W: 3699 L: 3515 D: 8086
PENTA  [174, 1704, 3746, 1816, 210]
Vine SwedishChef
simd_puct
diff2.0+0.02
ELO    -3.33 +- 4.26 (95%)
LLR    -2.93 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 8554 W: 1991 L: 2073 D: 4490
PENTA  [106, 1044, 2061, 958, 108]
Vine SwedishChef
simd_puct
diff5.0+0.05
ELO    1.51 +- 3.19 (95%)
LLR    0.11 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 14490 W: 3326 L: 3263 D: 7901
PENTA  [140, 1693, 3533, 1722, 157]
Vine SwedishChef
simple_puct_speedup
diff5.0+0.05
ELO    8.61 +- 5.11 (95%)
LLR    3.00 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 5728 W: 1390 L: 1248 D: 3090
PENTA  [61, 605, 1394, 739, 65]
Vine SwedishChef
combined_v55_p19_test3
diff40.0+0.40
ELO    11.17 +- 5.80 (95%)
LLR    3.00 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 3982 W: 854 L: 726 D: 2402
PENTA  [26, 406, 1001, 530, 28]
Vine SwedishChef
combined_v55_p19_test3
diff10.0+0.10
ELO    17.72 +- 7.81 (95%)
LLR    2.94 (-2.25, 2.89) [0.00, 5.00]
GAMES  N: 2512 W: 626 L: 498 D: 1388
PENTA  [18, 262, 587, 352, 37]