Closed PGG106 closed 10 months ago
I have no idea why we were using that awful tm code for regular time controls STC:
Elo | 12.26 +- 6.54 (95%) SPRT | 8.0+0.08s Threads=1 Hash=16MB LLR | 2.93 (-2.25, 2.89) [0.00, 3.00] Games | N: 5244 W: 1363 L: 1178 D: 2703 Penta | [20, 506, 1386, 689, 21]
Note: at STC this patch only exacerbate the existing time out issue, with the entire test having 13 time outs (10 from the dev version).
LTC:
Elo | 4.52 +- 3.83 (95%) SPRT | 40.0+0.40s Threads=1 Hash=64MB LLR | 2.89 (-2.25, 2.89) [0.00, 3.00] Games | N: 14612 W: 3473 L: 3283 D: 7856 Penta | [16, 1499, 4094, 1673, 24]
Note: no time out noticed at LTC.
I have no idea why we were using that awful tm code for regular time controls STC:
Note: at STC this patch only exacerbate the existing time out issue, with the entire test having 13 time outs (10 from the dev version).
LTC:
Note: no time out noticed at LTC.