Extended-precision floating-point format – Texas Instruments TMS320C3x User Manual
Page 113
Floating-Point Formats
5-8
You must use the following reserved values to represent 0 in the single-precision
floating-point format:
e = – 128
s = 0
f = 0
The following examples illustrate the range and precision of the single-precision
floating-point format:
Most positive:
x = (2 – 2
– 23
)
Ч
2
127
= 3.4028234
Ч
10
38
Least positive:
x = 1
Ч
2
–127
= 5.8774717
Ч
10
–39
Least negative:
x = (–1–2
– 23
)
Ч
2
–127
= – 5.8774724
Ч
10
–39
Most negative:
x = – 2
Ч
2
127
= – 3.4028236
Ч
10
38
5.3.4
Extended-Precision Floating-Point Format
In the extended-precision format, the floating-point number is represented by
an 8-bit exponent field (
e ) and a 32-bit mantissa field (man) with an implied
most significant nonsign bit (see Figure 5–9).
Figure 5–9. Extended-Precision Floating-Point Format
Exponent
Sign
Fraction
39
32 31 30
0
Mantissa
Operations are performed with an implied binary point between bits 31 and 30.
When the implied most significant nonsign bit is made explicit, it is located to
the immediate left of the binary point. The floating-point number
x is given by
the following:
x = 01.f
×
2
e
if
s = 0
x = 10.f
×
2
e
if
s = 1
x = 0
if
e = –128
You must use the following reserved values to represent 0 in the extended-
precision floating-point format:
e = –128
s = 0
f = 0