Just wondering if any of the CPU-intensive competitors are noticing higher frequency of timeouts during the finals on proven code?
I subbed an earlier version for the finals and gave it a bit more of a buffer to finish each turn gracefully, just in case, but I'm seeing a much higher time out rate than on that very version when it was live. About 1 in 10, as opposed to, well, 0 in 100 during the "season".
I'm a sample of 1 though.
Surely the finals are taxing the available hardware to the maximum, but if there's a difference in processing speed available to a bot, compared to what we've tuned for, that's not great
P.S. bot is in Python