| Active : 0 Machines / 0 Threads / 0 MNPS | ||||||
|---|---|---|---|---|---|---|
| Priority 0 | ||||||
| Rosent | Winter | graph_history_hack * | diff | 8.0+0.08 | LLR: 1.22 (-2.94, 2.94) [-5.00, 0.00] Games: 46972 W: 12429 L: 12586 D: 21957 Ptnml(0-2): 863, 5703, 10474, 5620, 826 | Cleaner version and testing without win adjudication. |
| Finished | ||||||
| Rosent | Winter | graph_history_hack * | diff | 8.0+0.08 | LLR: 3.05 (-2.94, 2.94) [-5.00, 0.00] Games: 56066 W: 14764 L: 14883 D: 26419 Ptnml(0-2): 1007, 6880, 12324, 6869, 953 | Hack to handle graph history interactions related to the 50 move rule. This exists in other engines as well, such as Stockfish. Plan is to have this until I add 50 move rule to the static evaluation, at which point I will instead add 50 move rule to position hashing in a way that matches it in the eval. |
| Rosent | Winter | deeper_net * | diff | 40.0+0.40 | LLR: 3.00 (-2.94, 2.94) [0.00, 5.00] Games: 8314 W: 2153 L: 2001 D: 4160 Ptnml(0-2): 83, 931, 1971, 1095, 77 | Added correct weights file. |
| Rosent | Winter | deeper_net * | diff | 8.0+0.08 | LLR: 2.99 (-2.94, 2.94) [-1.00, 4.00] Games: 4830 W: 1447 L: 1301 D: 2082 Ptnml(0-2): 86, 544, 1032, 644, 109 | Added correct weights file. |
| Rosent | Winter | deeper_net * | diff | 8.0+0.08 | LLR: 0.00 (-2.94, 2.94) [-1.00, 4.00] Games: 0 W: 0 L: 0 D: 0 Ptnml(0-2): 0, 0, 0, 0, 0 | Deeper net should have small N/s penalty, but should hopefully scale better. Testing against other network trained from scratch. |
| Rosent | Winter | qnet_422s_d32 * | diff | 8.0+0.08 | LLR: -2.98 (-2.94, 2.94) [0.00, 5.00] Games: 3584 W: 955 L: 1074 D: 1555 Ptnml(0-2): 107, 437, 791, 382, 75 | Testing net trained from scratch. |
| Rosent | Winter | qnet_422_d32 * | diff | 40.0+0.40 | LLR: 1.88 (-2.94, 2.94) [-1.00, 4.00] Games: 5662 W: 1094 L: 1012 D: 3556 Ptnml(0-2): 43, 569, 1533, 635, 51 | Updated network. Slightly more lax bounds to ensure we make it in time for the TCEC submission. |
| Rosent | Winter | qnet_422_d32 * | diff | 8.0+0.08 | LLR: 2.98 (-2.94, 2.94) [-1.00, 4.00] Games: 16562 W: 3975 L: 3799 D: 8788 Ptnml(0-2): 292, 1970, 3657, 1994, 368 | Updated network. Slightly more lax bounds to ensure we make it in time for the TCEC submission. |
| Rosent | Winter | disallow_consecutive_null_moves * | diff | 8.0+0.08 | LLR: -1.84 (-2.94, 2.94) [-4.00, 1.00] Games: 25332 W: 5810 L: 5958 D: 13564 Ptnml(0-2): 487, 3094, 5617, 3016, 452 | Consecutive null moves can happen as perspective can make both values fall above beta. This would erroneously register as a position repetition. For this reason we try disallowing repetitions. |
| Rosent | Winter | disallow_consecutive_null_moves * | diff | 40.0+0.40 | LLR: -0.30 (-2.94, 2.94) [-4.00, 1.00] Games: 1584 W: 366 L: 381 D: 837 Ptnml(0-2): 8, 200, 394, 179, 11 | Likely the change is exactly neutral at STC, as consecutive null moves require quite a bit of depth. Lets try LTC! |
| Rosent | Winter | move_order_bugfix * | diff | 40.0+0.40 | LLR: 2.95 (-2.94, 2.94) [-4.00, 1.00] Games: 6020 W: 1159 L: 1064 D: 3797 Ptnml(0-2): 40, 628, 1593, 695, 54 | Fixed move ordering bug where history heuristics were ignored if previous move was a null move. |
| Rosent | Winter | move_order_bugfix * | diff | 8.0+0.08 | LLR: 2.95 (-2.94, 2.94) [-4.00, 1.00] Games: 11914 W: 2756 L: 2661 D: 6497 Ptnml(0-2): 199, 1398, 2699, 1431, 230 | Restarted the test as a non-regression patch. It is unclear how much Elo the change should be worth, but it is the more reasonable form, so it should be added even if Elo neutral or slightly negative. |
| Rosent | Winter | move_order_bugfix * | diff | 8.0+0.08 | LLR: 0.04 (-2.94, 2.94) [0.00, 5.00] Games: 130 W: 30 L: 28 D: 72 Ptnml(0-2): 1, 15, 31, 17, 1 | Fixed move ordering bug where history heuristics were ignored if previous move was a null move. |
| Rosent | Winter | history_reductions * | diff | 8.0+0.08 | LLR: -2.96 (-2.94, 2.94) [0.00, 5.00] Games: 7890 W: 1770 L: 1863 D: 4257 Ptnml(0-2): 146, 997, 1749, 910, 143 | Support for history reductions |
| Rosent | Winter | qnet_402_d32 * | diff | 40.0+0.40 | Elo: 71.79 +- 6.17 (95%) [N=4000] Games: 4000 W: 1273 L: 458 D: 2269 Ptnml(0-2): 12, 185, 915, 752, 136 | Regression test vs v5.0 |
| Rosent | Winter | qnet_402_d32 * | diff | 40.0+0.40 | LLR: 2.95 (-2.94, 2.94) [0.00, 5.00] Games: 4092 W: 872 L: 744 D: 2476 Ptnml(0-2): 39, 381, 1085, 495, 46 | d32 net update. |
| Rosent | Winter | net382_d24f64b * | diff | 8.0+0.08 | LLR: 0.79 (-2.94, 2.94) [0.00, 5.00] Games: 9980 W: 2443 L: 2362 D: 5175 Ptnml(0-2): 212, 1170, 2133, 1275, 200 | Added net definition |
| Rosent | Winter | qnet_382_d32 * | diff | 40.0+0.40 | LLR: -0.55 (-2.94, 2.94) [0.00, 5.00] Games: 20744 W: 3953 L: 3894 D: 12897 Ptnml(0-2): 161, 2233, 5548, 2246, 184 | STC was passed against older master branch, but this is likely to scale even harder than the new master, so trying directly at LTC. |
| Rosent | Winter | qnet_382_d32 * | diff | 8.0+0.08 | LLR: 2.97 (-2.94, 2.94) [0.00, 5.00] Games: 17098 W: 4326 L: 4111 D: 8661 Ptnml(0-2): 375, 2021, 3560, 2200, 393 | Even larger quantized net. Might not yet be sufficiently mature. |
| Rosent | Winter | qnet_382_d24 * | diff | 40.0+0.40 | LLR: 2.95 (-2.94, 2.94) [0.00, 5.00] Games: 3404 W: 749 L: 622 D: 2033 Ptnml(0-2): 28, 331, 873, 426, 44 | Now that net is quantized, we try the larger size again. |
| Rosent | Winter | qnet_382_d24 * | diff | 8.0+0.08 | LLR: 2.96 (-2.94, 2.94) [0.00, 5.00] Games: 12870 W: 3262 L: 3070 D: 6538 Ptnml(0-2): 265, 1499, 2728, 1665, 278 | Now that net is quantized, we try the larger size again. |
| Rosent | Winter | quantize_net * | diff | 40.0+0.40 | LLR: 2.96 (-2.94, 2.94) [0.00, 5.00] Games: 1382 W: 324 L: 201 D: 857 Ptnml(0-2): 7, 121, 325, 218, 20 | Quantized network. Should be faster, at the cost of a little accuracy. |
| Rosent | Winter | quantize_net * | diff | 8.0+0.08 | LLR: 2.97 (-2.94, 2.94) [0.00, 5.00] Games: 1012 W: 301 L: 160 D: 551 Ptnml(0-2): 10, 80, 211, 169, 36 | Quantized network. Should be faster, at the cost of a little accuracy. |
| Rosent | Winter | net382_W50a * | diff | 8.0+0.08 | LLR: -2.96 (-2.94, 2.94) [0.00, 5.00] Games: 4510 W: 1066 L: 1178 D: 2266 Ptnml(0-2): 109, 580, 971, 504, 91 | Net trained with more uniform data. If this is not negative we are happy because it simplifies our life. If this is negative it is fine as it means we were doing something reasonably fine. |
| Rosent | Winter | net347_d24f64 * | diff | 40.0+0.40 | LLR: -2.95 (-2.94, 2.94) [0.00, 5.00] Games: 4604 W: 872 L: 969 D: 2763 Ptnml(0-2): 60, 555, 1154, 488, 45 | Another day another net. |
| Rosent | Winter | net347_d24f64 * | diff | N=40000 | Elo: 25.02 +- 7.56 (95%) [N=4000] Games: 4020 W: 1277 L: 988 D: 1755 Ptnml(0-2): 95, 414, 767, 575, 159 | Preliminary fixed node test with a potential larger net. |