Well, you could use the fact that the average of a large number of dice throws (discrete uniform distribution) will approach a normal distribution, and that you'd need an average below 2.5 to not exceed 320.
But that'd be too easy. Monte Carlo it, no one will ever know.