Tests

Finished 25 shown

Author	Engine	Identifier	Code	Time control	Results
MasterStruct	Eunoia	feat/rfp-lerp #9068 Finished	diff ↗	8.0+0.08	LLR: 0.76 (-2.25, 2.89) [0.00, 5.00] Games: 8854 W: 2641 L: 2559 D: 3654 Ptnml(0-2): 279, 1034, 1733, 1088, 293
Lerp RFP score
Ellie	WakWak	fused-accumulator #9070 Passed Duck960	diff ↗	4.0+0.04	LLR: 2.94 (-2.25, 2.89) [0.00, 4.20] Games: 2744 W: 1008 L: 842 D: 894 Ptnml(0-2): 56, 214, 705, 302, 95
fused accumulator updates from icarus non functional speedup
Ellie	WakWak	score-before-split #9069 Passed Duck960	diff ↗	4.0+0.04	LLR: 2.91 (-2.25, 2.89) [0.00, 4.20] Games: 2854 W: 1044 L: 876 D: 934 Ptnml(0-2): 66, 218, 727, 314, 102
move scoring before splitting ducks nonfunctional speedup change
Ellie	WakWak	depth-aware-tt #9067 Finished Duck960	diff ↗	40.0+0.40	LLR: -0.29 (-2.25, 2.89) [0.00, 4.20] Games: 74 W: 10 L: 25 D: 39 Ptnml(0-2): 3, 14, 15, 5, 0
prefer deeper TT entries
dan	WakWak	feature/post-lmr-cont #9059 Failed Duck960	diff ↗	8.0+0.08	LLR: -2.28 (-2.25, 2.89) [0.00, 4.20] Games: 15584 W: 5070 L: 5127 D: 5387 Ptnml(0-2): 336, 1534, 4102, 1491, 329
dan	WakWak	feature/lmr-loses-material #9060 Failed Duck960	diff ↗	8.0+0.08	LLR: -2.31 (-2.25, 2.89) [0.00, 4.20] Games: 6502 W: 2047 L: 2138 D: 2317 Ptnml(0-2): 135, 685, 1693, 612, 126
Ellie	WakWak	punish-failed-ducks #9062 Failed Duck960	diff ↗	8.0+0.08	LLR: -2.25 (-2.25, 2.89) [0.00, 4.20] Games: 23378 W: 7656 L: 7684 D: 8038 Ptnml(0-2): 537, 2339, 5980, 2281, 552
punish failed ducks
Ellie	WakWak	noisy-tt-lmr #9061 Failed Duck960	diff ↗	8.0+0.08	LLR: -2.25 (-2.25, 2.89) [0.00, 4.20] Games: 14454 W: 4695 L: 4755 D: 5004 Ptnml(0-2): 306, 1457, 3768, 1383, 313
increase reduction on quiet moves when the TT is noisy
MasterStruct	Eunoia	feat/nmp-margin #9057 Failed	diff ↗	8.0+0.08	LLR: -2.26 (-2.25, 2.89) [0.00, 5.00] Games: 3164 W: 870 L: 966 D: 1328 Ptnml(0-2): 114, 410, 617, 340, 101
Increase NMP margin
MasterStruct	Eunoia	feat/insufficient-material #9058 Passed	diff ↗	8.0+0.08	LLR: 2.93 (-2.25, 2.89) [-5.00, 0.00] Games: 4600 W: 1402 L: 1298 D: 1900 Ptnml(0-2): 88, 440, 1151, 522, 99
Add insufficient material detection
dan	Hobbes	feature/see-quiet-check #9052 Finished	diff ↗	8.0+0.08	LLR: -0.76 (-2.25, 2.89) [0.00, 3.00] Games: 18114 W: 4411 L: 4416 D: 9287 Ptnml(0-2): 77, 2161, 4586, 2156, 77
Ellie	WakWak	kesra #9054 Failed Duck960	diff ↗	40.0+0.40	LLR: -2.26 (-2.25, 2.89) [0.00, 4.20] Games: 2360 W: 771 L: 884 D: 705 Ptnml(0-2): 93, 228, 614, 189, 56
king buckets and 1536 l1
Ellie	WakWak	kesra #9055 Failed Duck960	diff ↗	N=25000	LLR: -2.27 (-2.25, 2.89) [0.00, 4.20] Games: 8634 W: 3260 L: 3362 D: 2012 Ptnml(0-2): 413, 823, 1949, 717, 415
king buckets and 1536 l1
Ellie	WakWak	kesra #9053 Failed Duck960	diff ↗	8.0+0.08	LLR: -2.29 (-2.25, 2.89) [0.00, 4.20] Games: 1956 W: 663 L: 786 D: 507 Ptnml(0-2): 99, 181, 495, 150, 53
king buckets and 1536 l1
MasterStruct	Eunoia	feat/inc-fp-depth #9022 Finished	diff ↗	8.0+0.08	LLR: 0.58 (-2.25, 2.89) [0.00, 5.00] Games: 18842 W: 5660 L: 5536 D: 7646 Ptnml(0-2): 621, 2150, 3727, 2330, 593
Increase Futility Pruning max depth
MasterStruct	Eunoia	feat/qs-pv #9021 Finished	diff ↗	8.0+0.08	LLR: -1.21 (-2.25, 2.89) [-5.00, 0.00] Games: 13528 W: 3957 L: 4086 D: 5485 Ptnml(0-2): 471, 1629, 2668, 1550, 446
Fix short PVs and add seldepth Fix threefold in qsearch
Drex	Blini	q-search #9049 Passed	diff ↗	8.0+0.08	LLR: 3.03 (-2.25, 2.89) [0.00, 5.00] Games: 550 W: 225 L: 46 D: 279 Ptnml(0-2): 3, 18, 103, 99, 52
q-search
dan	Hobbes	feature/lmr-noisy-hist #9047 Failed	diff ↗	8.0+0.08	LLR: -2.27 (-2.25, 2.89) [0.00, 3.00] Games: 14166 W: 3372 L: 3470 D: 7324 Ptnml(0-2): 39, 1557, 3988, 1461, 38
MasterStruct	Eunoia	feat/aw-centering #9023 Finished	diff ↗	8.0+0.08	LLR: 1.07 (-2.25, 2.89) [0.00, 5.00] Games: 17726 W: 5223 L: 5091 D: 7412 Ptnml(0-2): 418, 1996, 3953, 2028, 468
Widen aspiration windows based on score
MasterStruct	Eunoia	feat/aw-centering #9039 Finished	diff ↗	40.0+0.40	LLR: 1.64 (-2.25, 2.89) [0.00, 5.00] Games: 11132 W: 3181 L: 3060 D: 4891 Ptnml(0-2): 214, 1222, 2602, 1285, 243
Widen aspiration windows based on score
dan	Hobbes	feature/lmr-important-capture #9044 Finished	diff ↗	8.0+0.08	LLR: 0.17 (-2.25, 2.89) [0.00, 3.00] Games: 22696 W: 5450 L: 5390 D: 11856 Ptnml(0-2): 52, 2694, 5819, 2708, 75
Ellie	WakWak	anadama_4 #9042 Failed Duck960	diff ↗	40.0+0.40	LLR: -2.30 (-2.25, 2.89) [0.00, 4.20] Games: 5558 W: 1778 L: 1871 D: 1909 Ptnml(0-2): 116, 575, 1476, 510, 102
Ellie	WakWak	anadama_4 #9041 Failed Duck960	diff ↗	8.0+0.08	LLR: -2.26 (-2.25, 2.89) [0.00, 4.20] Games: 6766 W: 2280 L: 2375 D: 2111 Ptnml(0-2): 197, 701, 1671, 628, 186
nocturn9x	Heimdall	tyrfing-12ksn #9043 Passed	diff ↗	8.0+0.08	LLR: 2.93 (-2.25, 2.89) [0.00, 3.00] Games: 8286 W: 2265 L: 2053 D: 3968 Ptnml(0-2): 65, 935, 1952, 1105, 86
Merge master and select A8607157 (bench 5611448) Bring tyrfing-12ksn up to date with master and select the new A8607157 network with the unchanged multilayer TI architecture. Keep the weights local in networks/files/tyrfing-12ksn-v2.bin; network publication remains deferred until the net is merged. Calibrate on all 7,153,653 positions in lichess-big3-resolved.book. At scale 305, getStats reports absolute means of 927.90 for master's 510EEBF8 network and 1004.55 for A8607157. getScale recommends 281.727639, rounded to 282. At scale 282, the new absolute mean is 928.76. Update the documented default scale. Validation: make dev and the runtime-check build pass. Scalar and AVX2 NNUE tests each pass 17,731 incremental/full-refresh comparisons with identical checksums. Python regressions pass (99 tests, 5 skipped) with Git signing disabled for temporary test commits. The repository commit hook also passes make test. The optimized depth-13 bench is 5611448 nodes. Network SHA-256: a8607157fbb8ac900ac7fe25d45d772509db9ecf3d9b5cb5000e5a581e1d0890 Co-authored-by: GPT 6 <codex@openai.com>
dan	Hobbes	bugfix/se-early-tt-overwrite #9040 Passed	diff ↗	8.0+0.08	LLR: 2.90 (-2.25, 2.89) [-4.00, 0.00] Games: 20200 W: 4829 L: 4760 D: 10611 Ptnml(0-2): 68, 2319, 5232, 2438, 43