Scribd
Upload
9 views7 pages

06.floating Point Numbers

Uploaded by

jayathilakakavin358fi
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
9 views7 pages

06.floating Point Numbers

Uploaded by

jayathilakakavin358fi
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/ 7

Digital technology

Numbering systems
Jari Hautamäki
Numbering systems
Floating point number
• If the number is presented by using the complement of two
• E.g.with an 8-bit processor only numbers -128...+127 can be presented
• Even with a 16-bit processor only numbers -32768...+32767 are reached
• Bigger numbers than this cannot be presented with the reserved numbers of bits
• Direct presentation is not possible with fractions.
• Hence, in computers Floating Point Representation is used:

L = S*KE

• L = number to be presented
• S = significand
• K = base number (in binary number presentation 2)
• E = exponent
12.1.2022 © Jari Hautamäki
Numbering systems
Floating point representation L = S*KE
• Enables a much wider number range representation with the same number of bits as in
straight?direct representation
Examples:
2000000 = 2*106
0.00101012 = 101012*2-7 (0.1640625 = 21*2-7)

• Can represent fractions but the precise value of zero cannot be represented

12.1.2022 © Jari Hautamäki


Numbering systems
Floating point representation
• In the floating point representation the number is always normalized i.e., most significant bit is
one
• E.g., with 8-bit significand the number 001010012*25 normalized is 101001002*23
• This type of prepresentation where the first bit is always 1, makes it possible that in
the representation not even most significant bit needs to be shown.
• Thus, one bit is saved
• E.g. a 32-bit floating number:
+/- <- EKSPONENTTI -> <- MANTISSA ->
BITTI 31 30 24 23 1 0

• Bit 31: Sign bit of significant: 1 = - and 0 = +


• Bits 30...24: 7-bit exponent
• bits 23...0: 24-bit significand

12.1.2022 © Jari Hautamäki


Numbering systems
Floating point representation
• Exponent is complement of two in representation when the number range is +63...-64

+63 = 0,111111
...
+1 = 0,000001
0 = 0,000000
- 1 = 1,111111
...
-64 = 1,000000

• The representation of the significand contains 25 bits although in the bit string?jono there are 24
• Most significant bit can be omitted from the bit string
• The number is normalized to the range 1...0,5, so that number 0.111111...˜1 and number 0.100000... ˜
12.1.2022
0,5. © Jari Hautamäki
Numbering systems
Floating point representation
• 32-bit floating number absolute value is in the range as represented previously
• 0.1111...*263 ---- 0.1*2-64 i.e. with decimal numbers 0,999...*263 ---- 0,5*2-64 i.e. 9,2*1018 - 2,7*10-20
• With a direct 32-bit number presentation we could only reach number range +231...-(231-1)

• Example: Represent number 23810 as a 32-bit floating number?


• Convert number 23810 to a binary number
-> 128+64+32+8+4+2=111011102
• Convert the representation of the number as for significant to range 1...0,5
> 0.1110111*28
• Now the floating point representation is in format:
• M=110111 (Note: the first one could be omitted)
• E=810=10002
• Sign bit =0=positive number
• I.e. the Floating point representation is: 0|0001000|110111000000000000000000
•
12.1.2022 © Jari Hautamäki
Exercises
6. Represent in 32-bit floating number
a) -123
b) 0,56640625
c) 0,01953125

7. What is the biggest and smalles positive number that can be represented with floating number representation
with 21 bits? In the character string the expontent=4 bits and significand =16 bits long.

12.1.2022 © Jari Hautamäki

You might also like