2 digit figures (such as two dice) are unstable enough as it is, but a 6 digit number is a whole other story. Nevertheless, I do understand that using a large number could cut down on the number of trials, but I don't want to make the experiment so astronomical that it has the possibility of missing less astronimical yet relevent data, you know what I mean?

If you try to go too big too fast, you may not get a positive results even if there is something to the phenomenon.