Scribd
Upload
56 views

Floating Point Representation Examples

The document discusses three examples of floating point representation: 8-bit floating point with and without a hidden bit, and the IEEE 754 standard for single precision floating point numbers. The 8-bit examples show the breakdown of the bit patterns into sign, exponent, and fraction fields. The exponent uses excess-3 notation with a range of -3 to 4. The fraction is either normalized with an implied leading 1 or uses a hidden leading 1 bit. The IEEE 754 standard uses a 32-bit single precision format with a sign bit, 8-bit exponent with excess-127 notation having a range of -126 to 127, and a 23-bit fraction with a hidden leading 1 bit. It describes

Uploaded by

mohamed
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
56 views

Floating Point Representation Examples

The document discusses three examples of floating point representation: 8-bit floating point with and without a hidden bit, and the IEEE 754 standard for single precision floating point numbers. The 8-bit examples show the breakdown of the bit patterns into sign, exponent, and fraction fields. The exponent uses excess-3 notation with a range of -3 to 4. The fraction is either normalized with an implied leading 1 or uses a hidden leading 1 bit. The IEEE 754 standard uses a 32-bit single precision format with a sign bit, 8-bit exponent with excess-127 notation having a range of -126 to 127, and a 23-bit fraction with a hidden leading 1 bit. It describes

Uploaded by

mohamed
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Comp 255: Floating Point Representation Examples

8-bit floating point example



| | | |
+---+---+---+---+---+---+---+---+
sign exponent fraction

sign: 0 = plus / 1 = minus
exponent: excess 3 (011) notation (range is -3 to 4)
fraction: normalized so leading 1 is right of implied binary point (or binary point to left of
leading 1)
precision depends on exponent

Example: 5.25 -> 101.01 -> 0.101012
11
-> 0 | 110 | 1010

Largest value is 0 | 111|1111 -> 0.1111 2
100
-> 1111.0 = 15 (decimal)
Smallest non-zero value is 0|000|1000 -> 0.12
-11
= 0.0001 = 1/16 = 0.0625
Note: 0 | 000 | 0000 is zero

Not all bit patterns can be used

8-bit floating point example with hidden bit

| | | |
+---+---+---+---+---+---+---+---+
sign exponent fraction

sign: 0 = plus / 1 = minus
exponent: excess 3 (011) notation (range is -3 to 4)
fraction: hidden 1 bit to left of implied binary point (or binary point to right of leading 1)
precision depends on exponent

Example: 5.25 -> 101.01 -> 1.01012
10
-> 0 | 101 | 0101

Largest value is 0 | 111 | 1111 -> 1.11112
100
= 11111 = 31 (decimal)
Smallest non-zero value is 0 | 000 | 0001 = 1.00012
-11
= 0.0010001 = 1/8 + 1/128 = 0.1328125
Note: 0 | 000 | 0000 is zero









IEEE 754 Standard for Single Precision Floating Point Numbers

|s| exponent | fraction |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
3 3 2 2 2 2 2 2 2 2 2 2 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0
1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0

Sign: 0 = plus / 1 = minus
Exponent: excess 127 (01111111) 00000000 and 11111111 are reserved patterns
Therefore exponent range is -126 .. +127
Fraction: hidden 1 bit to left of implied binary point (or binary point to right of leading 1)
precision depends on exponent

5 basic types

Non-zero normalized numbers (exponents between -126 and +127)
Clean zero: exponent and mantissa all 0s (note +0 and -0)
Infinity: exponent = 11111111, fraction all 0s (+ and infinity)
NaN (not a number): exponent = 11111111 and fraction is not all 0s
Denormalized: exponent = 00000000 (-126) and fraction is actual value w/o/ hidden bit

You might also like