Without knowing the particulars of a Implementation, it's hard to say. Some can refine results by running the model a few more times, so yeah, better processing

Without knowing the particulars of a Implementation, it's hard to say. Some can refine results by running the model a few more times, so yeah, better processing and/or more time would help, though probably not by much. Most models, however, don't, so no special benefit from better processing other than speed