double.parse 返回精度差异

Question

Why does double.parse ("1833.35") return value 1833.349999999999 ?

Answer 1

It's because floating-point values are represented in the machine with binary exponents, and 10 is not an even power of 2. Thus, some base-10 fractions turn into repeating fractions when converted.

To illustrate, just work out how you'd represent 35/100 in binary.

Answer 2

You could switch to using a decimal instead of a double if you need it to be absolutely correct. Decimal stores the number with a base 10 exponent.

There is no way to represent.35 exactly using a double or float.