Converting real numbers to floating point representations based on the IEEE 754 standard

The design of a float of the IEEE 754 standard consists of a sequence of bits arranged in fields to represent the sign of the number, the biased binary exponent of the number, and the significand or modified mantissa.

bit
Single precision
Double precision function comment

31
63 sign 1=negative

30-16
62-46 Biased exponent Bias = 2^(n-1)-1

15-0 Significand Most significant bit not stored

bit Single precision Double precision	function	comment
31 63	sign	1=negative
30-16 62-46	Biased exponent	Bias = 2^(n-1)-1
15-0	Significand	Most significant bit not stored