| Active : 0 Machines / 0 Threads / 0 MNPS | ||||||
|---|---|---|---|---|---|---|
| Priority 0 | ||||||
| Vshche | Igel | ttfix | diff | 10.0+0.10 | LLR: -1.00 (-2.94, 2.94) [0.00, 3.00] Games: 7706 W: 1968 L: 2014 D: 3724 Ptnml(0-2): 63, 956, 1864, 904, 66 | 3.6.14: improve tt store algo |
| Kieren | Halogen | r141 | diff | 40.0+0.40 | LLR: -1.04 (-2.94, 2.94) [0.00, 3.00] Games: 5692 W: 1397 L: 1445 D: 2850 Ptnml(0-2): 19, 674, 1503, 636, 14 | |
| Mhoupp | Stash | lmr_less_for_checks | diff | 40.0+0.40 | LLR: -1.75 (-2.94, 2.94) [0.00, 5.00] Games: 12642 W: 2974 L: 2990 D: 6678 Ptnml(0-2): 69, 1499, 3181, 1523, 49 | Decrease LMR if the move gives check |
| Priority -1 | ||||||
| Kieren | Halogen | r140 | diff | 40.0+0.40 | LLR: -0.07 (-2.94, 2.94) [0.00, 3.00] Games: 5630 W: 1391 L: 1383 D: 2856 Ptnml(0-2): 18, 651, 1459, 679, 8 | |
| Kieren | Halogen | r140 | diff | 8.0+0.08 | LLR: -2.12 (-2.94, 2.94) [0.00, 3.00] Games: 16384 W: 4203 L: 4299 D: 7882 Ptnml(0-2): 138, 1979, 4041, 1909, 125 | |
| Finished | ||||||
| Kieren | Halogen | r141 | diff | 8.0+0.08 | LLR: -2.96 (-2.94, 2.94) [0.00, 3.00] Games: 17808 W: 4497 L: 4643 D: 8668 Ptnml(0-2): 153, 2180, 4360, 2082, 129 | |
| Kieren | Halogen | r141 | diff | N=40000 | Elo: 1.53 +- 2.08 (95%) [N=40000] Games: 40038 W: 11527 L: 11351 D: 17160 Ptnml(0-2): 706, 4651, 9046, 4993, 623 | |
| Vshche | Igel | ttfix | diff | 60.0+0.60 | LLR: 0.00 (-2.94, 2.94) [0.00, 3.00] Games: 0 W: 0 L: 0 D: 0 Ptnml(0-2): 0, 0, 0, 0, 0 | 3.6.14: improve tt store algo |
| Kieren | Halogen | r140 | diff | N=40000 | Elo: 4.78 +- 2.11 (95%) [N=40000] Games: 40232 W: 11932 L: 11379 D: 16921 Ptnml(0-2): 715, 4570, 9052, 5005, 774 | |
| Rosent | Winter | net_502_24x64 * | diff | N=40000 | Elo: 2.86 +- 7.28 (95%) [N=4000] Games: 4006 W: 1111 L: 1078 D: 1817 Ptnml(0-2): 102, 478, 824, 483, 116 | Get an understanding of how far apart the based increased size net is compared to the current version. Fixed nodes in this case as system load would otherwise distort results. |
| Rosent | Winter | net_502_16x96 * | diff | 10.0+0.10 | LLR: -2.95 (-2.94, 2.94) [-5.00, 0.00] Games: 8948 W: 1999 L: 2170 D: 4779 Ptnml(0-2): 186, 1157, 1942, 1020, 169 | The larger 24x64 net architecture is failing STC so we instead try 16x96. This has far fewer parameters but still has a 5% slowdown we need to overcome. Based on training metrics this lies squarely between the baseline and 24x64. |
| Kieren | Halogen | extmove-int16-sse4 | diff | 8.0+0.08 | LLR: -1.47 (-2.94, 2.94) [0.00, 3.00] Games: 3118 W: 755 L: 839 D: 1524 Ptnml(0-2): 19, 409, 787, 325, 19 | SIMD movepicker + clamp scores to int16 to compress |
| Rosent | Winter | net_502_24x64 * | diff | 10.0+0.10 | LLR: -2.95 (-2.94, 2.94) [-5.00, 0.00] Games: 1566 W: 309 L: 453 D: 804 Ptnml(0-2): 65, 210, 333, 154, 21 | Having passed the fixed node test as expected, we try an STC test with [-5, 0] bounds, as the slowdown is expected to hurt more than in actual play. |
| Rosent | Winter | net_502_24x64 * | diff | N=40000 | LLR: 2.99 (-2.94, 2.94) [0.00, 5.00] Games: 2282 W: 745 L: 582 D: 955 Ptnml(0-2): 69, 209, 455, 306, 102 | Larger net, hopefully a clear improvement at fixed nodes. Around 16% less Nps. |
| Vshche | Igel | nn-epoch1109 | diff | 10.0+0.10 | LLR: -3.06 (-2.94, 2.94) [0.00, 3.00] Games: 9958 W: 2520 L: 2697 D: 4741 Ptnml(0-2): 94, 1311, 2350, 1126, 98 | 3.6.12: fix tb tt bug (#324) |
| Mhoupp | Stash | lmr_less_for_checks | diff | 8.0+0.08 | LLR: 2.98 (-2.94, 2.94) [0.00, 5.00] Games: 23534 W: 5929 L: 5710 D: 11895 Ptnml(0-2): 251, 2814, 5469, 2931, 302 | Decrease LMR if the move gives check |
| Kieren | Halogen | tt-prefetch-key-after | diff | 4.0+0.04 | LLR: 3.05 (-2.94, 2.94) [0.00, 3.00] Games: 11928 W: 3316 L: 3096 D: 5516 Ptnml(0-2): 100, 1244, 3063, 1450, 107 | Prefetch TT entry before apply_move using precomputed key AVX2: +2.10% +/- 0.19% |
| Kieren | Halogen | [Berserk] main | diff | 8.0+0.08 | Elo: -31.54 +- 11.66 (95%) [N=10000] Games: 1204 W: 273 L: 382 D: 549 Ptnml(0-2): 38, 154, 292, 115, 3 | Update default network to ed164cf2.nn (#890) |
| Kieren | Halogen | r139 | diff | 40.0+0.40 | LLR: 2.96 (-2.94, 2.94) [0.00, 3.00] Games: 8544 W: 2249 L: 2056 D: 4239 Ptnml(0-2): 21, 925, 2183, 1126, 17 | 768 l1 |
| Kieren | Halogen | r139 | diff | N=40000 | Elo: 13.84 +- 2.11 (95%) [N=40000] Games: 40006 W: 12341 L: 10748 D: 16917 Ptnml(0-2): 628, 4205, 8968, 5350, 852 | 768 l1 |
| Kieren | Halogen | r139 | diff | 8.0+0.08 | LLR: 2.95 (-2.94, 2.94) [0.00, 3.00] Games: 18422 W: 4918 L: 4682 D: 8822 Ptnml(0-2): 169, 2107, 4417, 2355, 163 | 768 l1 |
| Kieren | Halogen | r138 | diff | 40.0+0.40 | LLR: 2.94 (-2.94, 2.94) [0.00, 3.00] Games: 5116 W: 1401 L: 1215 D: 2500 Ptnml(0-2): 14, 517, 1313, 697, 17 | |
| Kieren | Halogen | r138 | diff | 8.0+0.08 | LLR: 2.99 (-2.94, 2.94) [0.00, 3.00] Games: 6316 W: 1749 L: 1542 D: 3025 Ptnml(0-2): 38, 693, 1510, 858, 59 | |
| Rosent | Winter | net_311m * | diff | 8.0+0.08 | LLR: -3.02 (-2.94, 2.94) [-2.50, 2.50] Games: 17246 W: 4156 L: 4292 D: 8798 Ptnml(0-2): 402, 2127, 3657, 2079, 358 | Trying to better understand the hybrid loss. Both nets are not fully trained, but are at a similar point in training. |
| Rosent | Winter | net_311l * | diff | 8.0+0.08 | LLR: -3.10 (-2.94, 2.94) [-3.00, 2.00] Games: 17742 W: 4218 L: 4374 D: 9150 Ptnml(0-2): 424, 2152, 3826, 2094, 375 | Trying to better understand the hybrid loss. Both nets are not fully trained, but are at a similar point in training. 311n removes the CE loss component completely. Even if worth Elo, we may not want this. |
| Rosent | Winter | net_311l * | diff | 8.0+0.08 | LLR: -2.97 (-2.94, 2.94) [0.00, 5.00] Games: 23074 W: 5479 L: 5505 D: 12090 Ptnml(0-2): 487, 2821, 4936, 2817, 476 | Final net with less CE loss component. |
| Rosent | Winter | net_311d * | diff | 8.0+0.08 | LLR: -2.95 (-2.94, 2.94) [0.00, 5.00] Games: 20226 W: 5005 L: 5044 D: 10177 Ptnml(0-2): 475, 2478, 4222, 2487, 451 | Reduced the weighting of the draw prediction loss component. |
| Mhoupp | Stash | se_negext_pv | diff | 8.0+0.08 | LLR: -2.98 (-2.94, 2.94) [0.00, 5.00] Games: 8150 W: 1970 L: 2057 D: 4123 Ptnml(0-2): 114, 1000, 1922, 937, 102 | Increase Negative Singular Extensions by one ply for PV nodes Bench: 4,382,866 |
| Kieren | Halogen | r137 | diff | 8.0+0.08 | LLR: -2.99 (-2.94, 2.94) [0.00, 3.00] Games: 4072 W: 992 L: 1179 D: 1901 Ptnml(0-2): 63, 541, 978, 428, 26 | check if non-Syzygy data does better with access to Syzygy than without |
| Kieren | Halogen | r137 | diff | 8.0+0.08 | LLR: -2.95 (-2.94, 2.94) [0.00, 3.00] Games: 4496 W: 1140 L: 1323 D: 2033 Ptnml(0-2): 65, 598, 1070, 485, 30 | 9.3B new data |