Scribd
Upload
125 views1 page

Ieee 754 Notes

The document provides a step-by-step explanation of how to convert a decimal number to its 32-bit IEEE 754 floating point representation using -176.375 as an example. It describes: 1) observing the sign; 2) converting the decimal to binary; 3) moving to scientific notation format with the significant and exponent; 4) calculating the exponent in binary; and 5) combining the sign, exponent, and significant into the final 32-bit representation.

Uploaded by

Mohsin Bhat
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
125 views1 page

Ieee 754 Notes

The document provides a step-by-step explanation of how to convert a decimal number to its 32-bit IEEE 754 floating point representation using -176.375 as an example. It describes: 1) observing the sign; 2) converting the decimal to binary; 3) moving to scientific notation format with the significant and exponent; 4) calculating the exponent in binary; and 5) combining the sign, exponent, and significant into the final 32-bit representation.

Uploaded by

Mohsin Bhat
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/ 1

IEEE 754 FLOATING POINT NOTES AND EXAMPLE

POWER OF 2 The following is a step by step roadmap to go from a decimal number to its IEEE 754
TABLE 32 bit floating point representation. The example will use -176.375 as an example
210 = 1024
29 = 512 STEP 1: OBSERVE THE SIGN
28 = 256 For -176.375 the sign is negative. This means the first bit will be 1, if positive
27 = 128 the first bit will be 0. Going forward the sign will be ignored, but then used again in the
26 = 64 last step.
25 = 32
24 = 16 STEP 2: FROM DECIMAL TO BINARY
23 = 8 Transform the decimal to binary (ignoring the sign). To do this subtract the
22 = 4 largest power of 2 relative to the decimal until you reach 0. Note that floating point is
21 = 2 an approximation and cannot be perfectly represented – but the potential rounding error
20 = 1 is very small.
. Calculations
2-1 = 0.5 Use 27 Use 25 Use 24 Use 2-2 Use 2-3
2-2 = 0.25 176.375 48.375 16.375 0.375 0.125
2-3 = 0.125 - 128.000 - 32.000 - 16.000 - 0.250 - 0.125
2-4 = 0.0625 48.375 16.375 0.375 0.125 0.000
2-5 = 0.03125
2-6 = 0.015625 As a sum : 27 + 25 + 24 + 2-2 + 2-3 = 176.375
As binary : 10110000.011 (Place a 1 in each position used)

STEP 3: MOVE TO SCIENTFIC NOTATION AND GET SIGNIFICANT


IEEE Floating points need to be in the format of 1.xxxxx * 2y. The significant is the xxxxx component
(ignore the 1. ) and has 23 bits. If your significant is shorter then 23 bits add trailing zeros.
10110000.011 = 1.0110000011 * 27

Significant = 01100000110000000000000 (had to add 13 trailing zeros)

STEP 4: CALCULATE EXPONENT IN BINARY


The exponent is represented by 8 bits (256 states) and is shifted by 127. In our example
( 1.0110000011 * 27 ) the exponent is 7. So we need to express 134 (from 7+127) in binary. Using the
same technique as step 2:
As a sum : 128 + 4 + 2 = 134
As a sum : 27 + 22 + 21 = 134
As binary : 10000110 (Place a 1 in each position used)

STEP 5: COMBINE SIGN, EXPONENT, AND SIGNIFICANT


The format is:

Sign Exponent Significant


(1 bit) (8 bits) (23 bits)
1 10000110 01100000110000000000000

Or: 11000011001100000110000000000000 (Hex : 0xc3306000)

You might also like