While we're here - could we benchmark the performance with just 5 machines per group vs 10 per group vs 20? If there isn't a significant drop in performance maybe we could be a bit more generous with how many machines do we show outright?
While we're here - could we benchmark the performance with just 5 machines per group vs 10 per group vs 20? If there isn't a significant drop in performance maybe we could be a bit more generous with how many machines do we show outright?