Computer Architecture

Unit - 1

 Introduction

 In Digital Computer, data and instructions are stored in computer memory using binary code (or
machine code) represented by Binary digIT’s 1 and 0 called BIT’s.
 The data may contain digits, alphabets or special character, which are converted to bits,
understandable by the computer.
 The number system uses well defined symbols called digits.
 Number systems are basically classified into two types. They are:
o Non-positional number system
o Positional number system

 Non-Positional Number System

 In olden days people use of this type of number system for simple calculations like additions and
subtractions.
 The non-positional number system consists of different symbols that are used to represent
numbers.
 Roman number system is an example of the non-positional number system i.e. I=1, V=5, X=10,
L=50.
 This number system cannot be used effectively to perform arithmetic operations.

 Positional Number System

 This type of number system are:

o Decimal number system
o Binary number system
o Octal number system
o Hexadecimal number system
 The total number of digits present in any number system is called its Base or Radix.
 Every number is represented by a base (or radix) x, which represents x digits.
 The base is written after the number as subscript such as 512(10).It is a Decimal number as its base
is 10.

1|Page
 To determine the quantity that the number represents, the number is multiplied by an integer
power of x depending on the position it is located and then finds the sum of the weighted digits.
 Example: Consider a decimal number 512.45(10) which can be represented in equivalent value
as: 5x102 + 1x101 + 2x100 + 4x10-1 + 5x10-2

 Decimal Number System

 It is the most widely used number system.

 The decimal number system consists of 10 digits from 0 to 9.
 It has 10 digits and hence its base or radix is 10.
 These digits can be used to represent any numeric value.
 Example: 123(10), 456(10), 7890(10).
 Consider a decimal number 542.76(10) which can be represented in equivalent value as:
5x102 + 4x101 + 2x100 + 7x10-1 + 6x10-2
Hundreds Tens Units One-tenth One-hundredth
Weights 102 101 100 10-1
10-2
Digits 5 4 2 7 6
Values 500 40 2 0.7 0.06

 Binary Number System

 Digital computer represents all kinds of data and information in the binary system.
 Binary number system consists of two digits 0 (low voltage) and 1 (high voltage).
 Its base or radix is 2.
 Each digit or bit in binary number system can be 0 or 1.
 The positional values are expressed in power of 2.
 Example: 1011(2), 111(2), 100001(2)
 Consider a binary number 11011.10(2) which can be represented in equivalent value
as: 1x24 + 1x23 + 0x22 + 1x21 + 1x20 + 0x2-1 + 0x2-2
Weights 24 23 22 21 20 2-1 2-2
Digits 1 1 0 1 1 1 0
Values 16 8 4 2 1 0.5 0.25

 Note: In the binary number 11010(2)

 The left most bit 1 is the highest order bit. It is called as Most Significant Bit (MSB).
 The right most bit 0 is the lower bit. It is called as Least Significant Bit (LSB).

2|Page
 Octal Number System

 The octal number system has digits starting from 0 to 7.

 The base or radix of this system is 8.
 The positional values are expressed in power of 8.
 Any digit in this system is always less than 8.
 Example: 123(8), 236(8), 564(8)
 The number 6418 is not a valid octal number because 8 is not a valid digit.
 Consider a Octal number 234.56(8) which can be represented in equivalent value as:
2x82 + 3x81 + 4x80 + 5x8-1 + 6x8-2
Weights 82 81 80 8-1 8-2
Digits 2 3 4 5 6
Values 64 8 1 0.125 0.015625

 Hexadecimal Number System

 The hexadecimal number system consists of 16 digits from 0 to 9 and A to F.

 The letters A to F represent decimal numbers from 10 to 15.
 That is, ‘A’ represents 10, ‘B’ represents 11, ‘C’ represents 12, ‘D’ represents 13, ‘E’ represents
14 and ‘F’ represents 15.
 The base or radix of this number system is 16.
 Example: A4 (16), 1AB (16), 934(16), C (16)
 Consider a Hexadecimal number 5AF.D(16) which can be represented in equivalent value as:
5x162 + Ax161 + Fx160 + Dx16-1
Weights 162 161 160 16-1
Digits 5 A F D
Values 256 16 1 0.0625

Number System Base Symbol used

Binary 2 0, 1
Octal 8 0,1,2,3,4,5,6,7
Decimal 10 0,1,2,3,4,5,6,7,8,9
0,1,2,3,4,5,6,7,8,9, A,B,C,D,E,F
Hexadecimal 16
where A=10; B=11; C=12; D=13; E=14; F=15

3|Page
 Number System Conversions

Conversion from Decimal to

Binary:

1. Steps to convert decimal number to binary number:

 Step 1: Divide the given decimal number by 2.

 Step 2: Take the remainder and record it on the right side.
 Step 3: Repeat the Step 1 and Step 2 until the decimal number cannot be divided further.
 Step 4: The first remainder will be the LSB and the last remainder is the MSB. The equivalent
binary number is then written from left to right i.e. from MSB to LSB.

Example: To convert the decimal number 87(10) to binary.

 So 87 decimal is written as 1010111 in binary.

 It can be written as 87(10)= 1010111(2)

4|Page
2. Steps to convert decimal fraction number to binary number:

 Step 1: Multiply the given decimal fraction number by 2.

 Step 2: Note the carry and the product.

 Step 3: Repeat the Step 1 and Step 2 until the decimal number cannot be divided further.

 Step 4: The first carry will be the MSB and the last carry is the LSB. The equivalent binary

fraction number is written from MSB to LSB.

Example 1: To convert the decimal number 0.3125(10) to binary.

Multiply by 2 Carry Product

0.3125 x 2 0 (MSB) 0.625
0.625 x 2 1 0.25
0.25 x 2 0 0.50
0.50 x 2 1 (LSB) 0.00
0.00

 Therefore, 0.3125(10) = 0.0101(2)

OR

5|Page
 Example 2: To convert the decimal number 152.671875(10) to binary.

3. Steps to convert binary number to decimal number

 Step 1: Start at the rightmost bit.

 Step 2: Take that bit and multiply by 2n, when n is the current position beginning at 0 and
increasing by 1 each time. This represents a power of two.
 Step 3: Then, add all the products.
 Step 4: After addition, the resultant is equal to the decimal value of the binary number.

Example 1: To convert the binary number 1010111(2) to decimal.

 Therefore, 1010111(2) = 87(10)

6|Page
 Example 2: To convert the binary number 11011.101(2) to decimal.

= 1x24 + 1x23 + 0x22 + 1x21 + 1x20 + 1x2-1 + 0x2-2 + 1x2-3

= 1x16 + 1x8 + 0x4+ 1x2 + 1x1 + 1x0.5+ 0x0.25+ 1x0.125
= 16 + 8 + 2 + 1 + 0.5 + 0.125
= 27.625(10)
OR
Weights 24 23 22 21 20 2-1 2-2 2-3
Digits 1 1 0 1 1 1 0 1
Values 16 8 4 2 1 0.5 0.25 0.125

 Therefore, 11011.101(2) = 27.625(10)

Conversion from Decimal to Octal

1. Steps to convert decimal number to octal number

 Step 1: Divide the given decimal number by 8.

 Step 2: Take the remainder and record it on the side.
 Step 3: Repeat the Step 1 and Step 2 until the decimal number cannot be divided further.
 Step 4: The first remainder will be the LSB and the last remainder is the MSB. The equivalent
octal number is then written from left to right i.e. from MSB to LSB.

Example 1: To convert the decimal number 3034(10) to octal number.

 So 3034 decimal is written as 5732 in octal.

 It can be written as 3034(10) = 5732(8)

 Note: If the number is less than 8 the octal number is same as decimal number.

Example 2: To convert the decimal number 0.3125(10) to octal number.

0.3125 x 8 = 2.5000 2

0.5000 x 8 = 4.0000 4

 Therefore, 0.3125(10) = 0.24(8)

7|Page
2. Steps to convert octal number to decimal number

 Step 1: Start at the rightmost bit.

 Step 2: Take that bit and multiply by 8n, when n is the current position beginning at 0 and
increasing by 1 each time. This represents the power of 8.
 Step 3: Then, add all the products.
 Step 4: After addition, the resultant is equal to the decimal value of the octal number.

Example 1: To convert the octal or base-8 number 5732(8) to decimal

 Therefore, 5732(8) = 3034(10)

Example 2: To convert the octal number 234.56(8) to decimal number.

= 2x82 + 3x81 + 4x80 + 5x8-1 + 6x8-2

= 2x64+ 3x8 + 4x1 + 5x0.125+ 6x0.015625
= 128 + 24 + 4 + 0.625 + 0.09375
= 156.71875(10)
OR
Weights 82 81 80 8-1 8-2
Digits 2 3 4 5 6
Values 64 8 1 0.125 0.015625

 Therefore, 234.56(8) = 156.71875(10)

Conversion from Decimal to Hexadecimal

1. Steps to convert decimal number to hexadecimal number

 Step 1: Divide the decimal number by 16.

 Step 2: Take the remainder and record it on the side.
 Step 3: Repeat the Step 1 and Step 2 until the decimal number cannot be divided further.

8|Page
  Step 4: The first remainder will be the LSB and the last remainder is the MSB. The equivalent
hexadecimal number is then written from left to right i.e. from MSB to LSB.

Example To convert the decimal number 16242(10) to hexadecimal

 So 16242 decimal is written as 3F72 in hexadecimal.

 It can be written as 16242(10) = 3F72 (16)
 Note: If the number is less than 16 the hexadecimal number is
same as decimal number.
2. Steps to convert hexadecimal number to decimal number

 Step 1: Start at the rightmost bit.

 Step 2: Take that bit and multiply by 16n, where n is the current position beginning at 0 and
increasing by 1 each time. This represents a power of 16.
 Step 3: Then, add all the products.
 Step 4: After addition, the resultant is equal to the decimal value of the hexadecimal number.

Example 1: To convert the Hexadecimal or base-16 number 3F72 to a decimal number.

Therefore, 3F72(16)= 16242(10)

Example 2: To convert the hexadecimal number 5AF.D(16) to decimal number.

= 5x162 + 10x161 + 15x160 + 13x16-1

= 5x256+ 10x16 + 15x1 + 13x0.0625
= 1280 + 160 + 15 + 0.8125
= 1455.8125(10)
OR
Weights 162 161 160 16-1
Digits 5 A F D
Values 256 16 1 0.0625

 Therefore, 5AF.D(16) = 1455.8125(10)

9|Page
 Conversion from Binary to Octal

Steps to convert Binary to octal

 Take a binary number in groups of 3 and use the appropriate octal digit in its place.

 Begin at the rightmost 3 bits. If we are not able to form a group of three, insert 0s to the

left until we get all groups of 3 bits each.

 Write the octal equivalent of each group. Repeat the steps until all groups have been converted.

Example 1: Consider the binary number 1010111(2)

1 010 111

1 2 7
Therefore, 1010111(2) = 127 (8)

Example 2: Consider the binary number 0.110111(2)

0 110 111

0 6 7
Therefore, 0.110111 (2) = 0.67 (8)

Example 3: Consider the binary number 1101.10111(2)

001 101 101 110

1 5 5 6
Therefore, 1101.10111(2) = 15.56 (8)

Note: To make group of 3 bits, for whole numbers, it may be necessary to add a 0’s to the left of MSB
and when representing fractions, it may be necessary to add a 0’s to right of LSB.

Conversion from Octal to Binary

Steps to convert octal to binary

 Step 1: Take the each digit from octal number

 Step 2: Convert each digit to 3-bit binary number. (Each octal digit is represented by a three-

bit binary number as shown in Numbering System Table)

10 | P a g e
 Octal digit 0 1 2 3 4 5 6 7

Binary Equivalent 000 001 010 011 100 101 110 111

Example 1: Consider the octal number 456(8) into binary

4  100
5  101
6  110
Therefore, 456(8) = 100101110 (2)

Example 2: Consider the octal number 73.16(8) into binary

7  100
3  101
1  001
6  110
Therefore, 73.16(8) = 100101.001110 (2)

Conversion from Binary to Hexadecimal

Steps to convert Binary to Hexadecimal

 Take a binary number in groups of 4 and use the appropriate hexadecimal digit in its place.

 Begin at the rightmost 4 bits. If we are not able to form a group of four, insert 0s to the

left until we get all groups of 4 bits each.

 Write the hexadecimal equivalent of each group. Repeat the steps until all groups have

been converted.

Example 1: Consider the binary number 1011001(2)

0101 1001

5 9 Therefore, 1011001 (2) = 59 (16)

Example 2: Consider the binary number 0.11010111(2)

0 1101 0111

0 D 7
Therefore, 0.110111 (2) = 0.D7 (16)

11 | P a g e
 Conversion from Hexadecimal to Binary

Steps to convert hexadecimal to binary

 Step 1: Take the each digit from hexadecimal number

 Step 2: Convert each digit to 4-bit binary number. (Each hexadecimal digit is represented by a
four-bit binary number as shown in Numbering System Table)

Example: Consider the hexadecimal number CEBA (16)

Therefore, CEBA (16) = 1100 1110 1011 1010 (2)

Conversion from Octal to Hexadecimal

Steps to convert Octal to Hexadecimal

Using Binary system, we can easily convert octal numbers to hexadecimal numbers and vice-versa

 Step 1: write the binary equivalent of each octal digit.

 Step 2: Regroup them into 4 bits from the right side with zeros added, if necessary.
 Step 3: Convert each group into its equivalent hexadecimal digit.

Example: Consider the octal number 274 (8)

2  010
7  111
4  100
Therefore, 274 (8) = 010 111 100 (2)
Group the bits into group of 4 bits as 0 1011 1100

0 1011 1100

0 B C
Therefore, 274 (8) = BC (8)

12 | P a g e
Chapter 3- Data Representation I PUC, MDRPUC, Hassan

Conversion from Hexadecimal to Octal

Steps to convert Hexadecimal to Octal

 Step 1: write the binary equivalent of each hexadecimal digit.

 Step 2: Regroup them into 3 bits from the right side with zeros added, if necessary.
 Step 3: Convert each group into its equivalent octal digit.

Example: Consider the hexadecimal number FADE (16)

F  1111
A  1010
D  1101
E  1110

Therefore, FADE (16) = 1111 1010 1101 1110 (2)

Group the bits into group of 3 bits from LSB as 001 111 101 011 011 110

001 111 101 011 011 110

1 7 5 3 3 6

Therefore, FADE (16)= 175336 (8)

Shortcut method to convert hexadecimal to decimal number

Note

 Binary Arithmetic

 It involves addition, subtraction, multiplication and division operations.

 Binary arithmetic is much simpler to learn because system deals with only two digit 0’s and 1’s.
 When binary arithmetic operations are performed on binary numbers, the results are also 0’s and
1’s.

13 | P a g e
Chapter 3- Data Representation I PUC, MDRPUC, Hassan

 Binary Addition

 The addition of two binary numbers is performed in same manner as the addition of decimal
number.
 The basic rules of binary addition are:
Addend1 Addend2 Sum Carry
0 0 0 0
0 1 1 0
1 0 1 0
1 1 0 1
 It adds only two bits and gives sum and carry. If a carry is generated, it should be carried over to
the addition of next two bits.

 Example 1: Add the following number

a) 4 and 3 b) 9 and 5

Decimal Binary Decimal Binary

4 101 9 1001
+3 + 011 +5 + 101
7 111 14 1110

 Example 2: Add 75 and 18 in binary number.

Convert this decimal number into binary
i.e.

26 25 24 23 22 21 20
64 32 16 8 4 2 1
1 0 0 1 0 1 1
75 = 64 + 8 + 2 + 1 = 1001011
18 = 16 + 2 = 10010
Carry  1
Append 1  1001011
Append 2  10010
Sum  1011101
 Example 3: Add binary number 1011.011 and 1101.111
Carry  1111 11
Append 1  1011.011
Append 2  1101.111 Exercise:
Sum  1) Add 10101 and 11011
11001.010 2) Add 1011101 and 1100111
14 | P a g e
Chapter 3- Data Representation I PUC, MDRPUC, Hassan

 Binary Subtraction

 This operation is same as the one performed in the decimal system.

 This operation is consists of two steps:
o Determine whether it is necessary for us to borrow. If the subtrahend (the lower digit) is
larger than the minuend (the upper digit), it is necessary to borrow from the column to the
left. In binary two is borrowed.
o Subtract the lower value from the upper value
 The basic rules of binary subtraction are:
Minuend Subtrahend Difference Barrow
0 0 0 0
0 1 1 1
1 0 1 0
1 1 0 0
 When we subtract 1 from 0, it is necessary to borrow 1 from the next left column i.e. from the
next higher order position.
 Example 1: Subtract the following number
a) 10 from 14 b) 9 from 29 c) 3 from 5

Decimal Binary Decimal Binary Decimal Binary

14 1110 29 11101 5 101

- 10 - 1010 - 09 - 01001 -3 - 011
4 0100 20 10100 2 010

 Example 2: Add 75 and -18 in binary

number. 75 = 64 + 8 + 2 + 1 =
1001011
18 = 16 + 2 = 10010
Borrow 
Minuend  1001011 75
Subtrahend  10010 - 18
Difference  0111001 57
 Example 3: Subtract binary number 0110100 from 1011000

Exercise:
1) Subtract 01110 from 10101
2) Subtract 25 from 35

15 | P a g e
 Chapter 3- Data Representation I PUC, MDRPUC, Hassan

 Representation of signed Integers

 The digital computer handle both positive and negative integer.

 It means, is required for representing the sign of the number (- or +), but cannot use the sign (-) to
denote the negative number or sign (+) to denote the positive number.
 So, this is done by adding the leftmost bit to the number called sign bit.
 A positive number has a sign bit 0, while the negative has a sign bit 1.
 It is also called as fixed point representation.
 A negative signed integer can be represented in one of the following:
1) Sign and magnitude method
2) One’s complement method
3) Two’s complement method

 Sign and magnitude method

 An integer containing a sign bit followed by magnitude bits are called sign-magnitude integer.
 In this method, first bit (MSB) is considered as a sign bit and the remaining bits stand for
magnitude.
 Here positive number starts with 0 and negative number starts with 1.
 Example: Consider a decimal number 25(10)
2 25
2 12 1 24 23 22 21 20
16 8 4 2 1
2 6 0 1 1 0 0 1
2 3 0
1 1

 So the binary number is 11001(2). If we take the size of the word is 1 byte( 8 bits), then the number 25
will be represented as

0 0 0 1 1 0 0 1

Sign Bit Magnitude = 25

 Suppose, if the number is -25, and then it will be represented as:
1 0 0 1 1 0 0 1

Sign Bit Magnitude = 25

16 | P a g e
Chapter 3- Data Representation I PUC, MDRPUC, Hassan

 1’s Complement representation

 This is the simplest method of representing negative binary number.

 The 1’s complement of a binary number is obtained by changing each 0 to 1 and each 1 to 0.
 In other words, change each bit in the number to its complement.
 Example 1: Find the 1’s complement of 101000.
Original binary number  1 0 1 0 0 0

Find 1’s Complement  0 1 0 1 1 1

Thus 1’s complement of 101000 is 010111.

 Example 2: Find the complement of 1010111.

1’s Original binary  1 0 1 0 1 1 1
number

Find 1’s Complement  0 1 0 1 0 0 0

Thus 1’s complement of 101000 is 010111.

 2’s Complement representation

 The 2’s complement of a binary number is obtained by taking 1’s complement of the number and
adding 1 to the Least Significant Bit (LSB) position.
 The general procedure to find 2’s complement is given by:
2’s Complement = 1’s Complement + 1

 Example 1: Find the 2’s complement of 101000.

Original binary number  10 1 0 0 0

Find 1’s Complement  01 0 1 1 1

Add 1 to LSB + 1
Hence 2’s Complement of 101000 is  01 1 0 0 0

 Example 2: Find the 1’s and 2’s complement of 1011101.

Original binary number  10 1 1 1 0 1
1’s Complement  01 0 0 0 1 0
+ 1
Hence 2’s Complement is  0 1 0 0 0 1 1

17 | P a g e
Chapter 3- Data Representation I PUC, MDRPUC, Hassan

Note:
Negation: It is the operation of converting a positive number to its negative equivalent or a negative
number to its positive equivalent. Negation is performed by performing 2’s complement system.

 Example 1: Consider the number +12. Its binary representation is 01100(2). Find the 2’s
complement of +12.

Original binary number  01 1 0 0

Find 1’s Complement  10 0 1 1

Add 1 to LSB + 1
2’s Complement  10 1 0 0

Clearly, this is a negative number since the sign bit is 1. Actually, 10100 represent -12(10)
which is the negative equivalent of the number 12(10)

 Example 2: Consider the number -12. Its binary representation is 10100(2). Find the 2’s
complement of -12.

Original binary number  10 1 0 0

Find 1’s Complement  01 0 1 1

Add 1 to LSB + 1
2’s Complement  01 1 0 0

Clearly, this is a positive number since the sign bit is 0. Actually, 01100 represent 12(10) which
is the negative equivalent of the number -12(10)

 Subtraction of Binary Number using Complement

 Most of the computers perform subtraction using complemented number. This is less expensive
because the same addition circuit is used for subtraction with slight changes in the circuit.
 In the binary number system we can perform subtraction operation using two methods of
complements:
o Subtraction using 1’s Complement
o Subtraction using 2’s Complement

18 | P a g e
Chapter 3- Data Representation I PUC, MDRPUC, Hassan

 Subtraction using 1’s Complement

 Case 1: Subtracting a smaller number from a larger number (Minuend is greater than Subtrahend)
o Step 1: Find the 1’s complement of the subtrahend.
o Step 2: Add this to the minuend.
o Step 3: Carry is generated, this carry is called as the end around carry
o Step 4: Add the end around carry back to the LSB to get the final difference.

 Example 1: Subtract 15 from 23 using 1’s complement.

Decimal Binary
Minuend 23 10111
Subtrahend -15 01111

1’s complement of subtrahend is 10000

Minuend  10 1 1 1
1’s Complement of subtrahend  +10 0 0 0
End around carry  110 1 1 1
Add end around carry  + 1
Difference is  10 0 0

 Case 2: Subtracting a larger number from a smaller number (Minuend is less than Subtrahend)
o Step 1: Find the 1’s complement of the subtrahend.
o Step 2: Add this to the minuend.
o Step 3: There will be no carry, Re complement the answer to get the difference

 Example 1: Subtract 52 from 25 using 1’s complement.

Decimal Binary
Minuend 25 011001
Subtrahend -52 110100

1’s complement of subtrahend is 001011

Minuend  01 1 0 0 1
1’s Complement of subtrahend  +00 1 0 1 1
 10 0 1 0 0
Since there is no carry
take 1’s complement and  01 1 0 1 1
attach a negative sign Hence, the result = - 011011 i.e. - 27

19 | P a g e
 Chapter 3- Data Representation I PUC, MDRPUC, Hassan

 Subtraction using 2’s Complement

 Case 1: Subtracting a smaller number from a larger number (Minuend is greater than Subtrahend)
o Step 1: Find the 2’s complement of the subtrahend.
o Step 2: Add this to the minuend.
o Step 3: Carry is generated, Discard the carry and the remaining bits give the difference.

 Example 1: Subtract 09 from 17 using 2’s complement.

Decimal Binary
Minuend 17 10001
Subtrahend -09 01001

1’s complement of subtrahend (9) is 1 0 1 1 0

Add 1 to LSB + 1 STEP 1
2’s complement of 9 is 10111

Minuend  10 0 0 1
2’s Complement of subtrahend  +1 0 1 1 1 STEP 2
End around carry 
10 1 0 0 0
Discard the carry STEP 3
Difference is 
10 0 0

Hence, the result = 1000 i.e. 8

 Case 2: Subtracting a larger number from a smaller number (Minuend is less than Subtrahend)
o Step 1: Find the 2’s complement of the subtrahend.
o Step 2: Add this to the minuend.
o Step 3: There will be no carry, hence take the 2’s complement of the answer and place a
negative sign in front.

 Example 1: Subtract 47 from 26 using 2’s complement.

Decimal Binary
Minuend 26 011010
Subtrahend -47 101111

1’s complement of subtrahend (47) is 0 1 0 0 0 0

Add 1 to LSB + 1 STEP 1
2’s complement of 9 is 010001

20 | P a g e
Chapter 3- Data Representation I PUC, MDRPUC, Hassan

Minuend  01 1 0 1 0
2’s Complement of subtrahend  +0 1 0 0 0 1 STEP 2
1 0 1 0 1 1

1’s Complement of answer  0 1 0 1 0 0

2’s Complement of answer  0 1 0 1 0 0 STEP 3

Hence, the result = 10101 i.e. 21

21 | P a g e