Filter
Top Products
Instruction Descriptions 4–65
true zero
The value +0, represented as exactly 64 zeros in a floating-point register.
4.7.4 Encodings
Floating-point numbers are represented with three fields: sign, exponent, and fraction. The sign
is 1 bit; the exponent is 8, 11, or 15 bits; and the fraction is 23, 52, 55, or 112 bits. Some
encodings represent special values:
The values of MIN and MAX for each of the five floating-point data formats are:
Sign Exponent Fraction
Vax
Meaning
VA X
Finite
IEEE
Meaning
IEEE
Finite
x All-1’s Non-zero Finite Yes +/–NaN No
x All-1’s 0 Finite Yes +/–Infinity No
0 0 Non-zero Dirty zero No +Denormal No
1 0 Non-zero Resv. operand No –Denormal No
0 0 0 True zero Yes +0 Yes
1 0 0 Resv. operand No –0 Yes
x Other x Finite Yes finite Yes
Data
Format
MIN MAX
F_floating 2**–127 * 0.5 2**127 *(1.0 – 2**–24)
(0.293873588e–38) (1.7014117e38)
G_floating 2**–1023 * 0.5 2**1023 * (1.0 – 2**–53)
(0.5562684646268004e–308) (0.89884656743115785407e308)
S_floating 2**–126 * 1.0 2**127 * (2.0 – 2**–23)
(1.17549435e–38) (3.40282347e38)
T_floating 2**–1022 * 1.0 2**1023 * (2.0 – 2**–52)
(2.2250738585072013e–308) (1.7976931348623158e308)
X_floating 2**–16382*1.0 2**16383*(2.0–2**–112)
(See below
†
)
†
(1.18973149535723176508575932662800702e4932)
(See below
‡
)
‡
(3.36210314311209350626267781732175260e–4932)