Yes, you are right, but it's the same.
The problem is that the parameters of Power function is double type, so the system convert p3 from Int to Double type, and we know that double type has a precision problem because it isn't bit-bit (like an Int) but is representive by mantissa and exponent, so you can have a precision problem (like, for example numer 3).
I saw this originally in the unit testing for my Pre-Algebra Tester app. Since the largest exponent in that program is 4, I didn't use Power(). I took the more pedestrian but safer route of multiplying.