FuryBench

Active : 15 Machines / 194 Threads / 112.63 MNPS
Priority 0
Integral Fury	spsa_6000	diff	40.0+0.40	ELO 1.57 +- 2.36 (95%) LLR 0.85 (-2.25, 2.89) [0.00, 3.00] GAMES N: 19644 W: 4840 L: 4751 D: 10053 PENTA [14, 2223, 5259, 2312, 14]
Quanticade DarkNeutrino	master 25K node datagen.	diff	N=25000	ELO -0.04 +- 0.21 (95%) GAMES N: 7452004/30000000 W: 2614966 L: 2615757 D: 2221281
Quanticade DarkNeutrino	faster_indexing nnue.c: Replace indexing with lookup table	diff	4.0+0.04	ELO 0.42 +- 1.36 (95%) LLR -0.17 (-2.25, 2.89) [0.00, 2.00] GAMES N: 64654 W: 15897 L: 15818 D: 32939 PENTA [295, 7083, 17490, 7166, 293]
Caissa Witek	singular-beta-history	diff	10.0+0.10	ELO 0.14 +- 1.02 (95%) LLR -1.43 (-2.94, 2.94) [0.00, 2.00] GAMES N: 118888 W: 28109 L: 28061 D: 62718 PENTA [352, 14276, 30180, 14244, 392]
Priority -1
Caissa Witek	rfp-qsearch-verify	diff	10.0+0.10	ELO 0.16 +- 1.01 (95%) LLR -1.38 (-2.94, 2.94) [0.00, 2.00] GAMES N: 120220 W: 28278 L: 28223 D: 63719 PENTA [365, 14410, 30510, 14455, 370]
Priority -2
Illumina Mergener	neg-extensions Negative extensions	diff	40.0+0.40	ELO 3.06 +- 7.50 (95%) LLR 0.31 (-2.25, 2.89) [0.00, 5.00] GAMES N: 2044 W: 423 L: 405 D: 1216 PENTA [6, 221, 549, 241, 5]
Illumina Mergener	neg-extensions Negative extensions	diff	8.0+0.08	ELO 0.59 +- 6.81 (95%) LLR -0.17 (-2.25, 2.89) [0.00, 5.00] GAMES N: 2934 W: 651 L: 646 D: 1637 PENTA [13, 366, 713, 353, 22]
Illumina Mergener	fix-non-pawn-corrhist Fix non pawn corrhist	diff	8.0+0.08	ELO -4.47 +- 5.13 (95%) LLR -1.94 (-2.25, 2.89) [0.00, 5.00] GAMES N: 3888 W: 813 L: 863 D: 2212 PENTA [13, 404, 1157, 360, 10]

Finished
Caissa Witek	pv-pruning-margins Search: PV-node margins for history, futility and SEE pruning	diff	10.0+0.10	ELO 0.17 +- 0.67 (95%) LLR -2.96 (-2.94, 2.94) [0.00, 2.00] GAMES N: 270620 W: 63798 L: 63662 D: 143160 PENTA [832, 32385, 68748, 32505, 840]
Integral Fury	spsa please ascend me!	smp diff	20.0+0.20	TUNE 106 parameters ITER 6,009 / 20,000 GAMES 24,038 / 80,000
Integral Fury	hmc_inputs_and_scaling	diff	10.0+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	diff	60.0+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	diff	10.0+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]
Integral Fury	hmc_inputs_and_scaling	diff	40.0+0.40	ELO 3.70 +- 2.40 (95%) LLR 2.91 (-2.25, 2.89) [0.00, 3.00] GAMES N: 19354 W: 4858 L: 4652 D: 9844 PENTA [20, 2150, 5125, 2368, 14]
Quanticade DarkNeutrino	pp Implement Pawn Pawn input set	diff	60.0+0.60	ELO -1.25 +- 8.85 (95%) LLR -0.14 (-2.25, 2.89) [0.00, 3.00] GAMES N: 1394 W: 326 L: 331 D: 737 PENTA [2, 159, 379, 156, 1]
Quanticade Anematode	rapture ojpowifjpawoifjepowijfpoi	diff	4.0+0.04	ELO 0.56 +- 0.68 (95%) LLR -2.35 (-2.25, 2.89) [0.00, 3.00] GAMES N: 251810 W: 62154 L: 61745 D: 127911 PENTA [1103, 27112, 69080, 27493, 1117]
Quanticade DarkNeutrino	pp Implement Pawn Pawn input set	diff	10.0+0.10	ELO -9.03 +- 5.07 (95%) LLR -2.32 (-2.25, 2.89) [0.00, 3.00] GAMES N: 5002 W: 1140 L: 1270 D: 2592 PENTA [30, 654, 1248, 554, 15]
Quanticade DarkNeutrino	pp Pawn pawn	diff	N=5000	ELO 13.17 +- 5.83 (95%) LLR 3.07 (-2.25, 2.89) [0.00, 3.00] GAMES N: 7680 W: 2708 L: 2417 D: 2555 PENTA [283, 795, 1479, 914, 369]
Integral Fury	hmc_inputs_and_scaling	diff	10.0+0.10	ELO 2.07 +- 1.63 (95%) LLR 2.90 (-2.25, 2.89) [0.00, 3.00] GAMES N: 46272 W: 11652 L: 11376 D: 23244 PENTA [162, 5354, 11823, 5640, 157]
Caissa Witek	stats-per-thread-counters	smp diff	2.0+0.02	ELO 1.82 +- 2.87 (95%) LLR 2.97 (-2.94, 2.94) [-4.00, 0.00] GAMES N: 16772 W: 4307 L: 4219 D: 8246 PENTA [114, 1996, 4082, 2076, 118]
Integral Fury	hmc_inputs_and_scaling	diff	N=20000	ELO 1.54 +- 1.17 (95%) LLR 3.08 (-2.25, 2.89) [0.00, 3.00] GAMES N: 142280 W: 42474 L: 41845 D: 57961 PENTA [3210, 16701, 30707, 17294, 3228]
Quanticade Anematode	move-lover move lover	diff	4.0+0.04	ELO 2.30 +- 1.76 (95%) LLR 2.94 (-2.25, 2.89) [0.00, 3.00] GAMES N: 39656 W: 9923 L: 9660 D: 20073 PENTA [193, 4321, 10543, 4572, 199]
Caissa Witek	tm-cross-thread-instability TimeManager: Cross-thread best move instability Each search thread counts how often its own root move changes between iterations. The main thread sums those counters, normalizes by thread count and folds the result into the soft time limit as one more multiplicative factor, so time allocation sees the helper threads disagreeing instead of only the main thread's own stability streak.	smp diff	5.0+0.05	ELO -5.48 +- 3.48 (95%) LLR -2.95 (-2.94, 2.94) [0.00, 3.00] GAMES N: 9516 W: 2180 L: 2330 D: 5006 PENTA [12, 1212, 2460, 1062, 12]

1 2 3 370 371 372

FuryBench Testing Framework