Lecture 1 :

Number and Data Representation

1
 1. Overview
1. Overview
1.1. Some Definitions
Representation of Numbers
Base/Radix:
• Decimal
• Binary
• Octal
• Hexadecimal
Unsigned and Signed numbers
Two’s Complement Subtraction
Non-integer numbers
2
1.1. Some Definitions
1.1 Some Definitions
1.1.1. Analogue
1.1.2. Digital
1.1.1 Analogue – Signals or information that are
represented by a continuously variable physical
quantity. For example, electricity, light, sound... In
fact, most things we encounter in our everyday world

Pic Ref: bbc.co.uk 3

1.1. Some Definitions 1.1 Some Definitions
1.1.1. Analogue
1.1.2. Digital 1.1.2 Digital – Discrete values typically represented
by values of a physical quantity. For example,
electricity

Pic Ref: answers.com 4

1.1. Some Definitions
1.1 Some Definitions
1.2. Representation Digital electronics takes different values (or ranges of
1.3. The decimal system values) of analogue signals to represent different, discrete,
values.
As an example, consider a light switch
A light switch only has two positions.
When it is ‘off’, we would measure 0 volts at the
supply to the light.
When it is ‘on’, we would measure 240 volts at the
supply to the light.
Thus, we can represent ‘on’ and ‘off’, or ‘0’ and ‘1’,
with two arbitrary analogue values; ‘0’ and ‘240’
5
1.1. Some Definitions
1.2 Representation
1.2. Representation
1.3. The decimal system All information when stored in a computer system,
must use some form of internal representation
▪ The program to be executed.
▪ The numerical data to be processed.
▪ Character strings ( A....Z etc. ).

6
1.3. The decimal system
1.3 The Decimal System
1.3.1. Decimal Example
What is a Decimal Value?
What does 331 really mean?
The position of each digit is assigned a weight.
Decimal Scheme right-most digit has a weight of
100 = 1.
Weighting increases by a factor of 10 for each
new symbol as the position moves to the left.

7
1.3. The decimal system 1.3.1 Decimal Example
1.3.1. Decimal Example

Again taking the value 331:

102 101 100

3 3 1
3 * 100 3 * 10 1*1
300 30 1

Value = 300 + 30 +1.

8
1.4. Binary 1.4 Binary
1.5. Radix or base
• Binary number system follows similar rules to
decimal system.
Except base is 2 not 10.
• By the BASE rules:
Symbols: 0, 1 (all symbols are unique)
Highest Symbol = BASE - 1.
BASE = 2 this is BINARY.
• Example: 112

9
1.4. Binary 1.5 RADIX or BASE
1.5. Radix or base
2. The Binary
Representation Scheme • In any number scheme the base value defines the
number of unique symbols. In the Decimal
scheme we use the symbols: 0,1,2,3,4,5,6,7,8,9.
• The highest symbol = BASE –1.
• All Symbol values are UNIQUE.
• Example: 1110

10
 2. The Binary Representation
2. The Binary
Representation Scheme
Scheme
2.1. Basic Conversion
2.2 Decimal to Binary • Taking a simple value: 01011010
Conversion
27 26 25 24 23 22 21 20

128 64 32 16 8 4 2 1

0 1 0 1 1 0 1 0

• Can you tell what this value is?

11
 2.1 Basic Conversion
2. The Binary
Representation Scheme • Converting this representation back to a more
2.1. Basic Conversion familiar DECIMAL form we can see that all
2.2 Decimal to Binary
the information necessary about the data is
Conversion
available.

27 26 25 24 23 22 21 20
0 1 0 1 1 0 1 0
0*128 1*64 0*32 1*16 1*8 0*4 1*2 0*1

0 + 64 + 0 + 16 + 8 + 0 + 2 + 0 = 90
12
 2.2 Decimal To Binary
2. The Binary
Representation Scheme
Conversion
2.1. Basic Conversion
• Rules:
2.2 Decimal to Binary
Conversion • Divide the number by 2.
• Look at the remainder ‘1’ or ‘0’.
• Use the remainder as Bit value.
• Loop till done.
• Convert 2510

13
13
 2.2.1 Example:
2.2 Decimal to Binary
Conversion
Conversion of 2510
2.2.1. Example Divis Quotie Remain
or nt der
2 25 1 LSB
2 12 0
2 6 0
2 3 1
2 1 1 MSB
0

• Binary Value = 110012

14
 2.3 Binary To Decimal
2.3 Binary to Decimal Conversion
Conversion
2.4 Most Significant • Calculate the value for each Digit Position as (Symbol *
Digit / Least Significant Base Weighting) for the position. Then sum all these
Digit
values.
3. 'Bits', 'Bytes' and
• Binary 11102
'Words'
• (1 * 23) + (1 * 22) + (1 * 21) + (0 * 20)
• (1 * 8) + (1 * 4) + (1 * 2) + (0 * 1)
• = 1410

This small number indicates the

base that we are working in –
Why? 15
 2.4 Most Significant Digit /
2.3 Binary to Decimal
Conversion Least Significant Digit
2.4 Most Significant
Digit / Least Significant • The concepts of the significance of the Symbol
Digit positions is not unique to binary.
3. 'Bits', 'Bytes' and • Units, Tens, Hundreds etc. In the decimal scheme.
'Words'
• Most Significant Digit is the Left - most symbol of the
Number.
• Least Significant Digit is the Right – most symbol of
the Number.
• Example: 110012

16
2.4 Most Significant
3. 'Bits', 'Bytes' and 'Words'
Digit / Least Significant
Digit
3. 'Bits', 'Bytes' and
• Within a computer system everything is stored or
'Words’ used in binary format.
4. Unsigned Numbers • Each Binary Digit is commonly known as a Bit.
• Within any computer system information is stored
or manipulated in groups of ‘n Bits’.

17
2.4 Most Significant
3. 'Bits', 'Bytes' and 'Words'
Digit / Least Significant
Digit • A Byte is the smallest grouping and it is defined to
3. 'Bits', 'Bytes' and contain ‘n’ Bits. By common usage, the Byte
'Words’
usually refers to a grouping of 8 Bits.
4. Unsigned Numbers
• A Word is a larger grouping of Bits.
• Size reflected by a PC’s architecture.
• Again by common usage, a Word is a grouping of
16 Bits and therefore is 2 Bytes.
• We also have Nibbles
• Which are?
18
3. 'Bits', 'Bytes' and
4.Unsigned Numbers
'Words’
4. Unsigned Numbers • Any number range is defined as:
• 0 Max Number.
• In the Decimal scheme, if we write a value as 27:
• The magnitude of the value is 27.
• By convention the value is Positive.
• On the other hand if we write the value as: -27
• It is not the same since the '-' SIGN identifies it as another
value.

19
4.Unsigned Numbers
4.1 Unsigned Binary Representation
4.1 Unsigned Binary
Representation • The Binary representation in an unsigned form
4.2 Kilo-Mega-Giga has only a MAGNITUDE.
4.3 Unsigned Number
Tables • All binary combinations of the "n" Bits will
uniquely identify a magnitude
• e.g. 0000 1111.
• Range 0 to 2n -1
• So program variable declarations can define the
size and type of storage used!

20
 4.2 Kilo - Mega - Giga
4.Unsigned Numbers
4.1 Unsigned Binary
Representation
• Within Binary these terms have slightly
4.2 Kilo-Mega-Giga
different meaning to those in Decimal. 20 1
21 2
4.3 Unsigned Number • Kilo = 103 = 1000 in Decimal. 22 4
Tables
• Kilo = 210 = 1024 in Binary. 23 8
.. ..
• Mega = Kilo * Kilo = (Kilo2). 29 512
• 1,048,576 Binary / 1,000,000 Decimal.
210 1K
• Decimal: Mega = 106 / Binary: Mega = 220.
211 2K
• Giga = Kilo * Mega. 220 1M
• Why? 230 1G
240 1T

21
 4.3 Unsigned Numbers Table
4.Unsigned Numbers
4.1 Unsigned Binary Look at this table of 16 values we need ‘4’
Representation Bits’ to represent all possible values.
4.2 Kilo-Mega-Giga
4.3 Unsigned Number Decimal Binary Decimal Binary
Tables 0 0000 8 1000
1 0001 9 1001
2 0010 10 1010
3 0011 11 1011
4 0100 12 1100
5 0101 13 1101
6 0110 14 1110
7 0111 15 1111 22
5. Signed Numbers
5. Signed Numbers
6. Radix Complement
• The Decimal number system uses ‘-‘ (minus)
• In Binary we only have two symbols 1 and 0.
• But we must be able to store signed numbers in
computer memory!
• One possible solution - signed magnitude
00010111 = +23
sign magnitude
10010111 = -23
sign magnitude
23
5. Signed Numbers
5. Signed Numbers
6. Radix Complement
• What effect does this have on the
size of integers a computer can use?
• 8 bit unsigned binary number
00000000 - 11111111
0 - 25510
• 8 bit signed binary number
01111111 - 11111111
+12710 - -12710
but we have 00000000 and 10000000! Is this
a problem?
Range -(2n-1 -1) to 2n-1 -1 24
5. Signed Numbers
6. Radix complement
6. Radix Complement
• 2's complement
• Variation of sign magnitude but has some very useful
characteristics.
• Subtraction becomes complementary addition.
• Convert -ve numbers to 2's complement and then add.
• Get subtraction through adding a negative.
30 + (-30) = 0

25
 6.1 Subtraction and Two’s
6. Radix Complement Complement
6.1. Subtraction and
Two’s Complement • To create the 2’s complement of a negative
6.1.1. Subtraction and number.
2’s Complement
1. Invert all '1's to '0' and '0's to '1’
2. Add 1.
• Each operand must have same number of bits.
• Example 25 - 7
25 = 00011001
7 = 00000111

26
6. Radix Complement
6.1.1 Subtraction and 2’s Comp
6.1. Subtraction and
First convert -7 to 2's complement
Two’s Complement
6.1.1. Subtraction and • Step 1 : form 1's complement
2’s Complement 00000111
11111000 1's complement

• Step 2 : add 1
11111000
+1
11111001 -7 in 2's complement representation

27
6. Radix Complement 6.1.1 Subtraction and 2’s Comp
6.1. Subtraction and
Two’s Complement
• Now add
6.1.1. Subtraction and
25 = 00011001
2’s Complement
-7 = 11111001
ans = 00010010 1810

• Check answer.

28
6.1. Subtraction and 6.1.2 2’s complement answers.
Two’s Complement
6.1.1. Subtraction and 2’s • ONLY CONVERT NEGATIVE NUMBERS!
Complement
6.1.2. 2’s complement
• When working out the real answer of a 2’s
answers complement sum.
• 1. Look at the most significant bit.
• 2. If it is a 1 then
• The answer is negative
• Change all ones to zero, all zeros to one and add one.
• Convert this to decimal and write the sign.

29
 Example
6.1. Subtraction and
Two’s Complement
6.1.1. Subtraction and 2’s
• Which of the following numbers are correctly
Complement
converted to 8 bit 2’s complement?
6.1.2. 2’s complement
answers
Example • -3310 = 000111112
• -308 = 10011000
• 3010 = 00011110
• I really don’t know

30
6.2. 2’s Comp and Sign
6.2 2’s Comp and Sign Magnitude
Magnitude

31
6.2. 2’s Comp and Sign
6.2 2’s Comp and Sign Magnitude
Magnitude
• Number range for 2's complement 8 and 16 bit
integers?
-128 to +127 8 bit
-32768 to +32767 16 bit
Range -(2n-1) to + 2n-1 -1
• Number range for the sign magnitude
representation for 8 and 16 bit integers?
-127 to +127 8 bit
-32767 to +32767 16 bit
Range -(2n-1 -1) to + 2n-1 -1
32
7. Hexadecimal(Hex)
7. Hexadecimal (Hex)
7.1. Conversion of
Hexadecimal to/from • Hexadecimal (Hex) number system is base 16.
Decimal
7.2. Binary to Hexadecimal
• 16 unique symbols.
Conversion • Symbols: 0,1,2,3,4,5,6,7,8,9,A,B,C,D,E,F.
7.3. Binary&Hex
Conversion
• A = 1010, B = 1110, C = 1210,
7.4. Making Binary easier D = 1310, E = 1410, F = 1510.
for Humans • Example: 3E816 = 100010.

33
 7. Hexadecimal (Hex)
7. Hexadecimal(Hex)
7.1. Conversion of • Weight is 16 raised to the power of the position.
Hexadecimal to/from
Decimal • Example: A1F16
7.2. Binary to Hexadecimal
Conversion
7.3. Binary&Hex
(A x 162) + (1 x 161) + (F x 160)
Conversion
7.4. Making Binary easier 162= 256 16 1
for Humans
10 x 256 + 1 x 16 15 x 1 =259110

34
 7.1 Conversion of Hexadecimal
7. Hexadecimal(Hex) to / from Decimal
7.1. Conversion of
Hexadecimal to/from • Hex 9AC16
Decimal (9 * 162)+ (A * 161) + (C * 160) = 2476
7.2. Binary to Hexadecimal 2304 + 160 + 12 = 2476
Conversion
• 137110
7.3. Binary&Hex Divisor Quotient Remainder
Conversion
7.4. Making Binary easier
for Humans 16 1371 11 =B
16 85 5
16 5 5
• Result = 55B16 0
35
7. Hexadecimal(Hex)
7.2 Binary to Hexadecimal Conversion
7.1. Conversion of
• Comparison of the
Hexadecimal to/from
Decimal
binary and
7.2. Binary to hexadecimal number
Hexadecimal Conversion systems.
7.3. Binary&Hex
Conversion
• What do you notice?
7.4. Making Binary easier
for Humans

36
7. Hexadecimal(Hex)
7.2 Binary to Hexadecimal Conversion
7.1. Conversion of
• A single Hex digit can
Hexadecimal to/from
represent a 4-bit binary
Decimal
number.
7.2. Binary to
Hexadecimal Conversion • Example:
7.3. Binary&Hex (110101111)2
Conversion
0001 1010 1111
7.4. Making Binary easier 1 A F
for Humans
(110101111)2 =
(1AF)16
1 A F
162 161 160
37
7. Hexadecimal(Hex)
7.3 Binary & Hex Conversion
7.1. Conversion of
Hexadecimal to/from • Consider the value:
Decimal
7.2. Binary to Hexadecimal
27 26 25 24 23 22 21 20
Conversion
7.3. Binary&Hex 0 1 0 1 1 0 1 1
Conversion
7.4. Making Binary easier • Split this into groupings of 4 Bits
for Humans
• This gives the Hex – 5B16
• Hex is simply groupings of 4 Binary Bits.

38
7. Hexadecimal(Hex)
7.4 Making Binary Easier for Humans
7.1. Conversion of
Hexadecimal to/from • Why was Hexadecimal chosen?
Decimal • Single hex digit = four bits.
7.2. Binary to Hexadecimal • 2 hex digits per byte.
Conversion • Allows simple conversion to and from binary.
7.3. Binary&Hex
• Simpler to convert between Base 2 and Base 10.
• The meaning of a value can be communicated simply and effectively,
Conversion
both at the computer (i.e. Base 2) level and also so that humans (i.e.
7.4. Making Binary easier Base 10) can visualise the meaning.
for Humans • Hex is easier for humans to understand and communicate.
• Example: A16 = 10102 = 1010.

39
8.OCTAL Radix 8
8. OCTAL Radix 8
9. Non Integer Numbers
• The only other commonly used number base is 8
or Octal.
• Why use octal?
• Direct conversion to binary
• How would you represent 23 decimal in octal?
• Hint 2310 is 000101112 in 8 bit binary.
• The answer is : 027

40
8.OCTAL Radix 8
9. Non Integer Numbers
9. Non Integer Numbers
• Can we always use WHOLE values?
• It would be nice if we could.
• Bank balances might be interesting.
• We need to look at what are called Floating Point numbers.

• Values such as 3.142, etc.

• Numbers with a decimal point!
• Data types such as double (in Java and C).

41
9. Non Integer Numbers
9.1 Scientific Notation
9.1. Scientific Notation
9.2. Scientific Notation: • This is a way of expressing very large and very
Floating Point Numbers small numbers.
• 537,000,000 is written 5.37 x 108
• 5.37 is called the Mantissa.
• 108 is called the Exponent.

• 0.000136 is written 1.36 x 10─4

• 1.36 is called the Mantissa.
• 10─4 is called the Exponent.
42
9. Non Integer Numbers
9.2 Scientific Notation:
9.1. Scientific Notation
• Note that a positive exponent tells you how many places to
9.2. Scientific Notation: move the decimal point to the right, making the number
Floating Point Numbers bigger.
9.3. Binary: Floating
• A negative Exponent is how many places to move the
Point Format
decimal point to the left, making the number smaller.

1.005 x 102 = 100.5 1.005 x 10-2 = 0.01005

1.005 x 100 = 1.005 1.005 x 104 = 10,050

• These are called Floating Point numbers.

43
9.2. Scientific Notation:
9.3 Binary: Floating Point Format
Floating Point Numbers
9.3. Binary: Floating • Expressing a number as a Mantissa and an
Point Format
Exponent allows us to store Floating Point
numbers in a format that only requires whole
9.4. Floating Point in
numbers.
Decimal
• Sign = 1 bit.
• Mantissa = 7 bits.
• Exponent = 8 bit signed magnitude.
• Total = 16 bits.
5.62510
+101.10102
1.011010 x 22
44
9.3. Binary: Floating 9.4 Floating Point in Decimal
Point Format
9.4. Floating Point in
oWhat does 3.14210 actually mean?
Decimal
9.5. Floating Point in 100 Point 10-1 10-2 10-3
Binary
1 1/10 1/100 1/1000
3 . 1 4 2

o3.142 = (3 * 1) + (1 * 1/10) + (1 * 4/100) + (1 *

2/1000)
= 3 + 0.1 + 0.04 + 0.002

45
9.4. Floating Point in
9.5 Floating Point in Binary
Decimal
9.5. Floating Point in
o What is 1.1012?
Binary
9.6. Limitation in Binary
20 Point 2-1 2-2 2-3
1 1/2 1/4 1/8
1 . 1 0 1

o 1.1012 = (1 * 1) + (1 * 1/2) + 0 + (1 * 1/8)

= 1 + 0.5 + 0 + 0.125
= 1.62510
46
9.5. Floating Point in
9.6 Limitation in Binary
Binary
9.6. Limitation in Binary • The Columns as we go RIGHT after the POINT
9.7. Converting Decimal to have the place values.
Binary
1/2 1/4 1/8 1/16 1/32
0.5 0.25 0.125 0.0625 0.03125

• When converting Floating Point decimal numbers

to binary using a fixed number of places.
• NOT ALL VALUES ARE POSSIBLE!
47
9.5. Floating Point in
9.7 Converting Decimal to Binary
Binary
9.6. Limitation in Binary • Use continuous multiplication by 2.
9.7. Converting Decimal
to Binary
• Each 1 or 0 to the left of the point of the result
9.7.1. Example: Decimal contributes to the binary number.
To Binary • Continue until the fractional part is 0.
• Or until you obtain the answer to the accuracy
you need.

48
9.6. Limitation in Binary
9.7.1 Example: Decimal to Binary
9.7. Converting Decimal
to Binary
Take the Decimal Fraction: 0.35
9.7.1. Example: Decimal
To Binary 0.35 x 2 0.70 0 MSB
0.70 x 2 1.4 1
0.4 x 2 0.8 0
0.8 x 2 1.6 1
0.6 x 2 1.2 1
LSB
0.2 x 2 0.4 0
and so on

Answer: 0.35 = 0.010110 …and so on

49
9.7.1. Example: Decimal to
9.7.2 How close is this answer to 0.35?
Binary
9.7.2. How close is this • 0.0101102
answer to 0.35?
20 2-1 2-2 2-3 2-4 2-5 2-6
9.8 Accuracy of Floating
Point Data 0 . 0 1 0 1 1 0

0*20 0
1* 2-2 = 1* 0.25 0.25
1* 2-4 = 1* 0.0625 0.0625
1* 2-5 = 1* 0.0625 0.03125
0.34375
50
9.7.1. Example: Decimal to
9.8 Accuracy of Floating Point Data
Binary
9.7.2. How close is this • Accuracy defined as closeness to true value.
answer to 0.35? • Data is inputted to a program in Decimal floating point format
9.8 Accuracy of Floating (e.g. via the keyboard).
Point Data • It is converted to a binary floating point format of fixed size.
• This may involve some loss of accuracy.
• Floating Point arithmetic is only approximate.
• Integer arithmetic is exact.

51
10. Storage Needs
10 Storage Needs
10.1. Precision
10.2. Range
10.3. Formats that have
been used
10.4. IEEE Format now
Used
10.5. Floating Point
Arithmetic

52
10. Storage Needs
10.1 Precision
10.1. Precision
10.2. Range • Determined by the number of bits used to store the
10.3. Formats that have Mantissa.
been used
10.4. IEEE Format now • A high precision format will result in more
Used accurate floating point arithmetic.
10.5. Floating Point
• Most languages provide a choice:
Arithmetic
• Single precision.
• Double precision.

53
10. Storage Needs
10.3 Formats That have been Used
10.1. Precision
10.2. Range
10.3. Formats that have
been used
10.4. IEEE Format now
Used
10.5. Floating Point
Arithmetic

55
10. Storage Needs
10.4 IEEE Format now Used
10.1. Precision
• IEEE 754 floating point standard.
10.2. Range
10.3. Formats that have
• Single Precision (32 bits):
been used
• Mantissa:
10.4. IEEE Format now • 1 Bit for Sign / 23 Bits for Value.
Used • Exponent:
10.5. Floating Point • 8 Bits.
Arithmetic • Double Precision (64 bits):
• Mantissa:
• 1 Bit for Sign / 52 Bits for Value.
• Exponent:
• 11 Bits.
56
10.4. IEEE Format now
11. Summary
Used
10.5. Floating Point • Representation of Numbers
Arithmetic
11. Summary
• Base/Radix: Hex and Octal

• Non-integer numbers and Floating-Point numbers

58
59