$0xC00000000 = (11000000000000000000000000000000)_2$
Sign(1bit) |
Exponent(8bits) |
Mantissa(23bits) |
1 |
10000000 |
00000000000000000000000 |
the numerical value of the number is evaluated as $(-1)^{s}*(1.m)*2^{e}$
where, sign (s), a base (b), a mantissa (m), and an exponent (e)
A bias of $(2^{n-1} - 1)$, where n is # of bits used in the exponent, is added to the exponent $(e)$ to get biased exponent $(E)$. So, the biased exponent (E) of a single precision number can be obtained as:
$E = e + 127$
$128 = e + 127$
$e = 1$
$(-1)^{1}*(1.00000000000000000000000 )*2^{1}$
$=-2.0$