Hi, when the program is running without parameters, i.e. awaiting input via stdin, processing and awaiting the next request, is it necessary to, for example, repeat the set_pot, set_effective_stack, set_bet_sizes, set_thread_num, set_accuracy, range etc. commands each time if they will be the same values ?
If those values will always be the same, would it make any difference if having provided them on the first solve; in subsequent solves (without a restart), just the ones that change, such as the pot size and board, followed by the build_tree, start_solve and dump_result commands were provided as the input ?
99%+ of the run time is the solve itself of course, so very little performance gained, but do the set_max_iteration, set_use_isomorphism etc. need to be provided each time if they don't change ?
From some experiments it seems there's no need to supply them if they stay the same, but thought I'd check.
Hi, when the program is running without parameters, i.e. awaiting input via stdin, processing and awaiting the next request, is it necessary to, for example, repeat the set_pot, set_effective_stack, set_bet_sizes, set_thread_num, set_accuracy, range etc. commands each time if they will be the same values ?
If those values will always be the same, would it make any difference if having provided them on the first solve; in subsequent solves (without a restart), just the ones that change, such as the pot size and board, followed by the build_tree, start_solve and dump_result commands were provided as the input ?
99%+ of the run time is the solve itself of course, so very little performance gained, but do the set_max_iteration, set_use_isomorphism etc. need to be provided each time if they don't change ?
From some experiments it seems there's no need to supply them if they stay the same, but thought I'd check.