I see what you mean but that only works in the assumption of crystals being the same shape and point size probability distribution. Which is a reasonable assump

I see what you mean but that only works in the assumption of crystals being the same shape and point size probability distribution. Which is a reasonable assumption for crystals but real materials are derived by grinding, so you can easily have a wider distribution of shapes (rounder) and sizes (small particles filling the gaps between larger ones).