FuryBench

FuryBench Testing Framework

Home to Integral, PlentyChess, Obsidian, Caissa, Quanticade, Serendipity, Spaghet, Schoenemann, Clarity, and Illumina

No logs for event exist

Active : 6 Machines / 326 Threads / 183.01 MNPS
Priority 0
Swedis	Integral	ti_tweaks	diff	4.0+0.04	LLR: 1.37 (-2.25, 2.89) [0.00, 2.50] Games: 9850 W: 2584 L: 2466 D: 4800 Ptnml(0-2): 53, 1067, 2575, 1169, 61 Elo: 4.16 +- 3.59 (95%)
DarkNe	Quanticade	net57	diff	60.0+0.60	LLR: 0.01 (-2.25, 2.89) [0.00, 3.00] Games: 28 W: 8 L: 7 D: 13 Ptnml(0-2): 0, 4, 5, 5, 0 Elo: 12.41 +- 82.38 (95%)	If it doesnt pass like this it will not pass ever
Priority -1
DarkNe	Quanticade	simp-ttpv-cutnode	diff	60.0+0.60	LLR: 0.43 (-2.94, 2.94) [-3.00, 1.00] Games: 6320 W: 1497 L: 1488 D: 3335 Ptnml(0-2): 1, 700, 1750, 707, 2 Elo: 0.49 +- 4.06 (95%)	search.c: Simplify ttpv && cutnode adjustment
Fury	Integral	lmr_depth_alpha_raises	diff	40.0+0.40	LLR: -1.62 (-2.25, 2.89) [0.00, 2.50] Games: 11434 W: 2714 L: 2803 D: 5917 Ptnml(0-2): 6, 1367, 3058, 1282, 4 Elo: -2.70 +- 3.09 (95%)

Finished
DarkNe	Quanticade	net57	diff	60.0+0.60	LLR: -1.46 (-2.25, 2.89) [0.00, 3.00] Games: 8376 W: 1951 L: 2015 D: 4410 Ptnml(0-2): 3, 994, 2260, 926, 5 Elo: -2.65 +- 3.59 (95%)	1024 L1
DarkNe	Quanticade	net57	diff	N=5000	Elo: 13.79 +- 2.53 (95%) [N=40000] Games: 40102 W: 13973 L: 12382 D: 13747 Ptnml(0-2): 1418, 4160, 7797, 4765, 1911	How much is it without the slowdown
Swedis	Integral	ti_tweaks	diff	4.0+0.04	LLR: -0.67 (-2.25, 2.89) [0.00, 2.50] Games: 23574 W: 6023 L: 6027 D: 11524 Ptnml(0-2): 159, 2527, 6405, 2551, 145 Elo: -0.06 +- 2.29 (95%)
Fury	Integral	nnue_threat_inputs *	diff	40.0+0.40	LLR: -1.20 (-2.25, 2.89) [0.00, 2.50] Games: 3198 W: 759 L: 836 D: 1603 Ptnml(0-2): 9, 402, 848, 337, 3 Elo: -8.37 +- 5.97 (95%)	Little optimization
DarkNe	Quanticade	lmr-rework-tune	diff	20.0+0.20	LLR: -0.20 (-2.25, 2.89) [0.00, 3.00] Games: 3956 W: 942 L: 945 D: 2069 Ptnml(0-2): 1, 427, 1125, 424, 1 Elo: -0.26 +- 5.05 (95%)	Will it gain ? Only one way to find out
DarkNe	Quanticade	lmr-rework	diff	20.0+0.20	Tuning 129 Parameters 3173/10000 Iterations 12692/40000 Games Played	lets tune the new LMR so it can beat master
DarkNe	Quanticade	fix-lmr-alpha-ttpv	diff	60.0+0.60	LLR: -2.96 (-2.94, 2.94) [-3.00, 1.00] Games: 17512 W: 3945 L: 4090 D: 9477 Ptnml(0-2): 3, 2006, 4882, 1863, 2 Elo: -2.88 +- 2.42 (95%)	search.c: Fix LMR ttpv alpha reduction
Fury	Integral	tile_threat_accum	diff	4.0+0.04	LLR: -2.28 (-2.25, 2.89) [0.00, 2.50] Games: 6948 W: 1685 L: 1843 D: 3420 Ptnml(0-2): 76, 815, 1830, 697, 56 Elo: -7.90 +- 4.42 (95%)
Fury	Integral	update_fmt	diff	4.0+0.04	LLR: 0.56 (-2.25, 2.89) [-3.00, 1.00] Games: 6054 W: 1607 L: 1591 D: 2856 Ptnml(0-2): 37, 641, 1650, 667, 32 Elo: 0.92 +- 4.48 (95%)
DarkNe	Quanticade	simp-ttpv-cutnode	diff	10.0+0.10	LLR: 2.96 (-2.94, 2.94) [-3.00, 1.00] Games: 24928 W: 5872 L: 5779 D: 13277 Ptnml(0-2): 34, 2950, 6406, 3037, 37 Elo: 1.30 +- 2.16 (95%)	search.c: Simplify ttpv && cutnode adjustment
DarkNe	Quanticade	lmr-unreduce-on-pv	diff	60.0+0.60	LLR: -2.96 (-2.94, 2.94) [-3.00, 1.00] Games: 12922 W: 2893 L: 3034 D: 6995 Ptnml(0-2): 1, 1550, 3500, 1409, 1 Elo: -3.79 +- 2.87 (95%)	Unreduce on PV in LMR instead
Fury	Integral	tile_threat_accum	diff	4.0+0.04	LLR: -2.27 (-2.25, 2.89) [0.00, 2.50] Games: 4606 W: 1133 L: 1291 D: 2182 Ptnml(0-2): 41, 573, 1218, 445, 26 Elo: -11.92 +- 5.29 (95%)
Fury	Integral	nnue_threat_inputs	diff	10.0+0.10	LLR: -2.34 (-2.25, 2.89) [0.00, 2.50] Games: 4348 W: 1049 L: 1216 D: 2083 Ptnml(0-2): 36, 590, 1073, 455, 20 Elo: -13.35 +- 5.57 (95%)	Remove assume alignment actually
Fury	Integral	faster_thrinputs	diff	4.0+0.04	LLR: -1.85 (-2.25, 2.89) [0.00, 2.50] Games: 3558 W: 890 L: 1026 D: 1642 Ptnml(0-2): 47, 443, 920, 337, 32 Elo: -13.29 +- 6.32 (95%)
Fury	Integral	faster_thrinputs	diff	10.0+0.10	LLR: -0.68 (-2.25, 2.89) [0.00, 4.00] Games: 348 W: 75 L: 109 D: 164 Ptnml(0-2): 5, 61, 73, 33, 2 Elo: -34.05 +- 21.53 (95%)
Fury	Integral	faster_thrinputs	diff	4.0+0.04	LLR: -2.26 (-2.25, 2.89) [0.00, 2.50] Games: 2482 W: 563 L: 736 D: 1183 Ptnml(0-2): 29, 371, 603, 220, 18 Elo: -24.26 +- 7.60 (95%)	Bruh
Fury	Integral	faster_thrinputs	diff	4.0+0.04	LLR: -2.30 (-2.25, 2.89) [0.00, 2.50] Games: 3052 W: 724 L: 904 D: 1424 Ptnml(0-2): 55, 410, 752, 278, 31 Elo: -20.51 +- 7.13 (95%)
DarkNe	Quanticade	fix-cutoff-lmr	diff	10.0+0.10	LLR: -3.03 (-2.94, 2.94) [-3.00, 1.00] Games: 22382 W: 5146 L: 5310 D: 11926 Ptnml(0-2): 40, 2722, 5840, 2540, 49 Elo: -2.55 +- 2.28 (95%)	search.c: Switch cutoff lmr to 1 ply ahead
Fury	Integral	nnue_threat_inputs	diff	N=20000	LLR: 2.94 (-2.25, 2.89) [0.00, 2.50] Games: 2550 W: 919 L: 633 D: 998 Ptnml(0-2): 43, 214, 536, 378, 104 Elo: 39.13 +- 9.07 (95%)	Incremental threat updates
Fury	Integral	nnue_threat_inputs	diff	10.0+0.10	LLR: -0.94 (-2.25, 2.89) [0.00, 2.50] Games: 574 W: 123 L: 196 D: 255 Ptnml(0-2): 14, 96, 126, 51, 0 Elo: -44.43 +- 16.56 (95%)	Incremental threat updates
Fury	Integral	nnue_threat_inputs	diff	N=20000	LLR: 2.99 (-2.25, 2.89) [0.00, 2.50] Games: 3076 W: 1089 L: 802 D: 1185 Ptnml(0-2): 55, 272, 646, 461, 104 Elo: 32.51 +- 8.13 (95%)	Merge branch 'main' into nnue_threat_inputs
Fury	Integral	nnue_threat_inputs	diff	N=20000	LLR: 1.27 (-2.25, 2.89) [0.00, 2.50] Games: 1240 W: 444 L: 318 D: 478 Ptnml(0-2): 24, 113, 242, 195, 46 Elo: 35.43 +- 13.24 (95%)	Fix up inference
DarkNe	Quanticade	lmr-unreduce-on-pv	diff	10.0+0.10	LLR: 0.94 (-2.94, 2.94) [-3.00, 1.00] Games: 10920 W: 2582 L: 2557 D: 5781 Ptnml(0-2): 25, 1250, 2882, 1281, 22 Elo: 0.80 +- 3.25 (95%)	search.c: Unreduce on pv in LMR instead
DarkNe	Quanticade	lmr-remove-check	diff	10.0+0.10	LLR: 2.96 (-2.94, 2.94) [-3.00, 1.00] Games: 12986 W: 3128 L: 3018 D: 6840 Ptnml(0-2): 20, 1487, 3378, 1579, 29 Elo: 2.94 +- 3.00 (95%)	search.c: Remove was in check lmr
Fury	Integral	lmr_depth_alpha_raises	diff	10.0+0.10	LLR: 2.90 (-2.25, 2.89) [0.00, 2.50] Games: 86612 W: 21807 L: 21440 D: 43365 Ptnml(0-2): 266, 10177, 22070, 10510, 283 Elo: 1.47 +- 1.19 (95%)

1 2 3 216 217 218