Greg Graham received his Ph.D. in particle physics from the University of Chicago in 1999. During his graduate career, he got hooked on computing, woodworking and bicycling. Computing because you can’t really do particle physics research without computers these days; woodworking because you have to have crates in which to ship your particle physics equipment; and bicycling because its a lot more fun than driving everywhere! After horsing around with computers for several years waiting around for the “next big experiment,” he bailed, got a M.S. in Computer Science from DePaul University, and now works as a software developer for a medium sized company in the Midwest. After a long and productive career, he hopes someday to bail on that too, and spend the rest of his time biking around and building furniture.
:)
I see that you’ve been keeping busy these past….what, 13 years? :)
Amazing stuff, my friend.
Greetings Gregory,
Putting any customary introduction aside, I can only assume you are using random number generation in the rolling routine given the multimillion number of rolls. I approached the problem from a more organized approach if I may be so bold. “Base Six Esque” (nested loops 6 deep). Every possible throw of the dice analyzed one at a time, evaluated to achieve the highest scoring configuration and incrementing that configurations variable, then cycling through the nested loop to eval. the next possible roll in the sequence. I.E.
if (111111){ six of a kind++;)//other configs ! incremented
111111
111112
111113
.
.
.
666664
666665
666666
All 46656 one at a time. I then used the same technique for five dice rolls, four dice rolls, three dice rolls etc.
I’ve generated data tables and was wondering if you would take a look at them.
email me at greggroessger@yahoo.com and I’ll reply with a text file attachment. Or a way of your choosing.