That’s not real math though, that’s a quirk of floating point math.

But NaN is also a quirk of IEEE754 floating pount math.