None of the benchmarks are useful since they aren't run in a full district with customized player characters. Before anyone says anything about minimizing run to run variance by not having players in district actually having heavily customized character playing will shows more real world numbers.