C＃中的十六进制到浮点转换

Question

I have a hex value 0x492655FE that want to convert to float. My code is

uint num = uint.Parse(hexString, System.Globalization.NumberStyles.AllowHexSpecifier);
byte[] floatVals = BitConverter.GetBytes(num);
BitConverter.ToSingle(floatVals, 0).Dump();

Result: 681311,9

But in ModbusPoll program : display float inverse result: 681311,8750

I tried another code. But result is same. What can I do for this problem?

Answer 1

According to this 0x492655FE in single-precision represents exactly 6.81311875E5 or 681311.875 in decimal. If you print more digits after the radix point you'll get the same output

However float has only 23 bits of mantissa which corresponds to ~6-7 decimal digits of precision. The rest are generally just noise because powers of 2 cannot represent exactly numbers in fractional numbers in decimal, thus it's safe to round to only the first 6-7 digits