Not a real number.

Float and double are approximate values in real number storage and cannot be accurate.

In fact, many real numbers have no exact values.

Such as π or natural logarithm e, etc.