Machine Level
Representation
of Data
CSC159 – Computer Organization
Lesson Outcomes
Numeric conversion between number bases
Decimal ↔ Binary ↔ Octal ↔ Hexadecimal
Fractional Conversions
Signed Numbers Representation
Sign-magnitude, 1s complement, 2s complement
Arithmetic operations
Addition and Subtraction
Floating point format
IEEE 754 standard – single precision
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 2
Number System
Number system is based on positional
notation (place value)
Decimal system: system of positional notation based on
powers of 10.
Binary system: system of positional notation based
powers of 2
Octal system: system of positional notation based on
powers of 8
Hexadecimal system: system of positional notation
based powers of 16
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 3
Number System (cont.)
Decimal or base 10 number system
Origin: counting on the fingers
“Digit” from the Latin word digitus meaning “finger”
Base: the number of different digits including
zero in the number system
Example: Base 10 has 10 digits, 0 through 9
Binary or base 2 1 byte = 8-bit
1 word = ?
Bit (binary digit): 2 digits, 0 and 1
Octal or base 8: 8 digits, 0 through 7
Hexadecimal or base 16: 16 digits, 0 through F
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 4
Number System (cont.)
Base or Radix
The number of different symbols required to
represent any given number
The larger the base, the more numerals are
required but the fewer digits needed
Base 10: 0, 1, 2, 3, 4, 5, 6, 7, 8, 9
Base 2: 0, 1
Base 8: 0, 1, 2, 3, 4, 5, 6, 7
Base 16: 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, A, B, C, D, E, F
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 5
Converting from Base 10
Powers Table
8 7 6 5 4 3 2 1 0
2 256 128 64 32 16 8 4 2 1
8 32,768 4,096 512 64 8 1
16 65,536 4,096 256 16 1
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 6
From Base 10 to Base 2
Base 10 42 Remainder
Quotient 2 ) 42 ( 0 Least significant bit
2 ) 21 ( 1
2 ) 10 ( 0
2) 5 (1
2) 2 (0
2) 1 Most significant bit
Base 2 101010
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 7
From Base 10 to Base 16
5,73510 = 166716
Power
4 3 2 1 0
Base
16 65,536 4,096 256 16 1
1 6 6 7
Integer 5,735 /4,096 1,639 / 256 103 /16 7
=1 =6 =6
Remainder 5,735 - 4,096 1,639 –1,536 103 – 96
= 1,639 = 103 =7
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 8
From Base 10 to Base 16
Base 10 5,735 Remainder
Quotient 16 ) 5,735 ( 7 Least significant bit
16 ) 358 ( 6
16 ) 22 ( 6
16 ) 1 ( 1 Most significant bit
16 ) 0
Base 16 1667
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 9
From Base 10 to Base 16
Base 10 8,039 Remainder
Quotient 16 ) 8,039 ( 7 Least significant bit
16 ) 502 ( 6
16 ) 31 ( 15
16 ) 1 ( 1 Most significant bit
16 ) 0
Base 16 1F67
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 10
From Base 8 to Base 10
72638 = 3,76310
Power 83 82 81 80
512 64 8 1
x7 x2 x6 x3
Sum for
Base 10 3,584 128 48 3
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 11
From Base 8 to Base 10
72638 = 3,76310
7
x8
56 + 2 = 58
x8
464 + 6 = 470
x8
3760 + 3 = 3,763
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 12
From Base 16 to Base 2
The nibble approach
Hex easier to read and write than binary
Base 16 1 F 6 7
Base 2 0001 1111 0110 0111
Why hexadecimal?
Modern computer operating systems and networks
present variety of troubleshooting data in hex
format
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 13
Fractions
Number point or radix point
Decimal point in base 10
Binary point in base 2
No exact relationship between fractional
numbers in different number bases
Exact conversion may be impossible
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 14
Decimal Fractions
Move the number point one place to the right
Effect:
multiplies the number by the base number
Example: 139.010 139010
Move the number point one place to the left
Effect:
divides the number by the base number
Example: 139.010 13.910
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 15
Fractions: Base 10 and Base 2
.258910
Place 10-1 10-2 10-3 10-4
Value 1/10 1/100 1/1000 1/10000
Evaluate 2 x 1/10 5 x 1/100 8 x 1/1000 9 x1/1000
Sum .2 .05 .008 .0009
.1010112 = 0.67187510
Place 2-1 2-2 2-3 2-4 2-5 2-6
Value 1/2 1/4 1/8 1/16 1/32 1/64
Evaluate 1 x 1/2 0 x 1/4 1x 1/8 0 x 1/16 1 x 1/32 1 x 1/64
Sum .5 0.125 0.03125 0.015625
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 16
Fractions: Base 10 and Base 2
No general relationship between fractions
of types 1/10k and 1/2k
Therefore a number representable in base 10
may not be representable in base 2
But: the converse is true: all fractions of the
form 1/2k can be represented in base 10
Fractional conversions from one base to
another are stopped
If
there is a rational solution or
When the desired accuracy is attained
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 17
Mixed Number Conversion
Integer and fraction parts must be converted
separately
Radix point: fixed reference for the
conversion
Digit to the left is a unit digit in every base
B0 is always 1 regardless of the base
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 18
Complementary Representations
We normally represent signed integers by a
plus or minus sign and a value
In computers, we cannot use a sign symbol
Must restrict to 0’s and 1’s
Representation of signed numbers:
Sign-and-magnitude
1s complement
2s complement
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 19
Sign-and-Magnitude Representation
Represent signed integers by a plus or minus
sign and a value
In computer, we cannot use a sign
Must restrict to 0’s and 1’s
Example: +ve value -ve value
+1 00000001 -1 10000001
+2 00000010 -2 10000010
+3 00000011 -3 10000011
+4 00000100 -4 10000100
+5 00000101 -5 10000101
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 20
One’s Complement
Numbers that begin with 0 are positive
Numbers that begin with 1 are negative
Performed by changing every 0 to 1 and every
1 to 0 (inversion) FLIP!
Example: +ve value -ve value
+1 00000001 -1 11111110
+2 00000010 -2 11111101
+3 00000011 -3 11111100
+4 00000100 -4 11111011
+5 00000101 -5 11111010
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 21
Two’s Complement
Find the 1’s complement and add 1 to the result
Example, find the 2’s complement of +2 in
binary:
+2 = 000000102 , -2 = ?
2’s complement:
1) Find 1’s complement
00000010 (flip)
11111101
2) Add 1
11111101
+ 1
11111110 = -2
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 22
Floating Point (Scientific Notation)
In many calculations the range of numbers used
is very large
A number can be represented by using scientific
representation commonly used in physics,
chemistry, and engineering
The computer version of the scientific notation is
called floating point
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 23
Floating Point (Scientific Notation)
Example (Decimal):
1. 523.467 = 52.3467 x 101
2. 3417000000000000 = 3.417 x 1015
3. - 0.000000001234 = - 1.234 x 10-9
Example (binary):
1. -10000.1 = -100.001 x 22
= -1.00001 x 24
2. 0.000101 = 1.01 x 2-4
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 24
Floating Point (Scientific Notation)
number = m x a e
m – mantissa
a – base
e – exponent (positive or negative integer)
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 25
Floating Point (Scientific Notation)
Single precision format uses 32-bit
SIGN EXPONENT SIGNIFICAND
< 1-bit > <8-bit > < 23-bit >
Double precision format uses 64-bit
< 1-bit > < 11-bit > < 52-bit>
UiTM Terengganu
Topic 2: Machine Level Representation of Data
26
- naza
Floating Point (Scientific Notation)
Sign 0 = positive , 1 = negative
Exponent excess-127 (single precision)
Significand similar to mantissa but with
an implied 1 bit, an implied binary point,
followed by the rest of the fraction
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 27
IEEE 754 format
(single-precision floating point)
Step 1 - Convert to binary
Step 2 - Put into 1.xxxx X 2y
Step 3 - Get the sign
Step 4 - Get the biased exponent
Step 5 - Get the mantissa & the
significand
Step 6 - Put into IEEE format Sign Biased Mantissa /
Exponent significand
1-bit 8-bit 23-bit
Step 7 - Convert to hexadecimal
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 28
Exercise
1. Represent the decimal value 15.25 using
32-bit IEEE single precision format in
hexadecimal form.
2. Convert BBE0000016 IEEE single precision
floating point number to its decimal value.
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 29
Convert the following numbers to IEEE single
precision format in hexadecimal form:
a) 1011110.100012
b) 0.0000000000110112
c) -712.510
UiTM Terengganu - naza Topic 2: Machine Level Representation of Data 30
