I would want to keep this stored as a float. Because chances are, I am going to be multiplying other floats by this number. And the performance penalty (not to

I would want to keep this stored as a float. Because chances are, I am going to be multiplying other floats by this number. And the performance penalty (not to mention the maintenance penalty) of converting back-and-forth between whatever you choose and floats will probably out-weigh the gains of storing it efficiently. That said something like a uint would probably be most efficient if you did care. At that point, allignment issues also start cropping up performance wise.