Part 1: Computer System

Chapter 1: Basic Theories of Information

Overview
In order to make a computer work, information needs to be converted into a format that can be understood by the computer.

Part 1: Computer System

Basic Theories of Information (Text No. 1 Chapter 1)

Objectives
Understanding a computers basic data units such as binary numbers, bits, bytes, words, etc. and their conversions from and to octal, decimal, and hexadecimal digits Understanding basic concepts of computer internal data representation, focusing on numeric data, character codes etc Understanding proposition calculus and logical operations

Some Terminology
Data representation unit and processing unit
1. Binary Digits (Bits)
Two levels of status in computers electronic circuits
Whether the electric current passes through it or not Whether the voltage is high or low

1 digit of the binary system represented by 1 or 0 Smallest unit that represents data inside the computer 1 bit can represent 2 values of data, 0 or 1 2 bits can represent 4 different values 00, 01, 10, 11

(or Table)

(or Row)

(or Column)

Bit representation
Switches
Current Lights

Open (0) or closed (1)

Not flowing (0) or flowing (1) Off (0) or on (1)

Numeric Conversion
3. Bytes
A byte is a unit that represents with 8 bits 1 character or number, 1 byte = 8 bits E.g. 00000000, 00000010, etc. 1 bit can be represented in 2 ways, i.e. combination of 8 bit patterns into 1 byte enables the representation of 28 = 256 types of information Using a 1-byte word, 256 different characters can be represented sufficient for most Western character sets However, the number of kanji (Chinese characters) amounts to thousands of different characters, hence a 1-byte word system is insufficient Two bytes are connected to obtain 16 bits, 216 = 65,536 A 2-byte word

Numeric Conversion
4. Word
The smallest unit that represents data inside a computer Increase operation speed

5. Number systems
Binary system is used to simplify the structure of electronic circuits that make up a computer Hexadecimal number is a numeric value represented by 16 numerals from 0 to 15 to ease the representation of binary numbers for humans computers are capable of only using binary numbers

Numeric Systems
Also known as Base Systems or Radix Systems Available digits:
Decimal system (base 10)
0, 1, 2, 3, 4, 5, 6, 7, 8, 9

Binary system (base 2)

0, 1

Octal system (base 8)

0, 1, 2, 3, 4, 5, 6, 7

Hexadecimal (base 16)

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

Numeric Data Representation

The true value of numbers are the same The representation of numbers vary
Decimal Binary Octal Hexadecimal

Numeric data representation

DECIMAL number (Radix/Base = 10)

Weight
Value Final (true) value

104
2*104

103
2*103

102
2*102

101
9*101

100
8*100

20000 + 1000 + 900 + 90 + 8 = 2199810 1 1 0 0 1

BINARY number (Radix/Base = 2)

Weight
Value Final (true) value

24
1*24

23
1*23

22
0*22

21
0*21

20
0*20

16 + 8 + 0 + 0 + 1 = 252

Binary Arithmetic
Addition and subtraction of binary numbers
Addition
0 + 0 = 0 (or 010) 0 + 1 = 1 (or 110) 1 + 0 = 1 (or 110) 1 + 1 = 10 (or 210) 00=0 0 1 = -1 10=1 11=0

Subtraction

Binary Addition

Result = 1001102

Binary Subtraction

Result = 10102

Hexadecimal arithmetic
4. Addition and subtraction of hexadecimal numbers
Addition
Performed starting at the lowest (first from the right) digit A carry to the upper digit is performed when the result is higher than 16

Subtraction
Performed starting at the lowest (first from the right) digit A borrow from the upper digit is performed when the result is negative

Hexadecimal Addition

First column from right

D + 7 = (In the decimal system: 13 + 7 = 20) = 16 (carried 1) + 4 The sum of the first column is 4 and 1 is carried to the second column.

Second column from right

1 + 8 + 1 = (In the decimal system: 10) = A Carried from the first column

Third column from right

A + B = (In the decimal system: 10 + 11 = 21) = 16 (carried 1) + 5 The sum of the third column is 5 and 1 is carried to the fourth column.

The result is (15A4)16.

Hexadecimal Subtraction

First column from right

Since 3 4 = 1, a borrow is performed from D in the second digit (D becomes C). 16 (borrowed 1) + 3 4 = F (In the decimal system: 19 4 = 15)

Second column from right

C 7 = 5 (In the decimal system: 12 7 = 5)

Third column
61=5

The result is (55F)16.

Exercises
Compute the following
a) b) c) d) 2710 + 1510 110112 + 11112 338 + 178 1B16 + F16

Compute the following

a) b) c) d) 5010 2210 1100102 - 101102 628 268 3216 - 1616

Numeric data representation

Radix/Base

300010 = 3 * 10
1. Radix and weight

Exponent

Representation of numeric data

Decimal numbers weight and its meaning
10 is called Radix upper right of 10 (in this example, 4) is called exponent

Binary digits weight and its meaning

2. Auxiliary units and power representation

Used to represent big, small amounts, and exponent to which the radix is raised

Radix/Base Conversion
In order to process numeric values in a computer, decimal numbers are converted into binary or hexadecimal numbers However, since we ordinarily use decimal numbers, it would be difficult to understand the meaning of the result of a process if it were represented by binary or hexadecimal numbers. This operation is called radix conversion The following radix/base conversion techniques will be discussed:
1. 2. 3. 4. 5. 6. Decimal to Binary Binary to Decimal Binary to Hexadecimal Hexadecimal to Binary Octal to Binary Binary to Octal

1. Decimal to Binary (Integer)

1. Decimal integer is divided into 2 2. The quotient and remainder are obtained 3. The quotient is divided into 2 again until the quotient becomes 0 4. The binary value is obtained by placing the remainder(s) in reverse order

1. Decimal to Binary (Integer)

1. Decimal to Binary (Fraction)

Decimal fraction is multiplied by 2
Resulting integer portion is extracted (always be 0 or 1) Resulting fraction portion is multiplied by 2 Operation is repeated until the fraction portion becomes 0

When decimal fractions are converted into binary fractions, most of the times, the conversion is not finished, since no matter how many times the fraction portion is multiplied by 2, it will not become 0. Most decimal fractions become infinite binary fractions.

1. Decimal to Binary (Fraction)

2. Binary to Decimal (Integer)

Performed by adding up the weights of each of the digits of the binary bit string

2. Binary to Decimal (Fraction)

Same technique as for binary integers.

3. Binary to Hexadecimal
4-bit binary strings are equivalent to 1 hexadecimal digit The binary number is divided into groups of 4 digits starting from the decimal point In the event that there is a bit string with less than 4 digits, the necessary number of 0s is added and the string is considered as a 4-bit string

3. Binary to Hexadecimal (Integer)

3. Binary to Hexadecimal (Fraction)

4. Hexadecimal to Binary (Integer)

1 digit of the hexadecimal number is represented with a 4-digit binary number

4. Hexadecimal to Binary (Fraction)

Same technique as per integer

Octal-Binary Conversions
Binary to/from Octal conversion
Conversion of binary to/from octal (whole numbers) Conversion of octal fractions
In decimal, 26.9210 = (2 * 101) + (6 * 100) + (9 * 10-1) + (2 * 10-2) 0.48 means 4 * 8-1 = (4/8) 10 = 10 = 0.510 0.2118 means (2 * 8-1) + (1 * 8-2) + (1 * 8-3) Binary fractions can be converted in a similar manner to octal as that of octal fractions The number can then be converted to decimal by adding up the whole numbers and convert the fractions to decimals

Conversion of binary fractions

Exercises
Convert into binary, octal and hexa
a) 2710 b) 1510 c) 50.2210

Convert into decimal

a) 110112 b) 338 c) 1B.F16

Quiz
Try this
A. What number does the next digit position represent in the hexadecimal system?

256

16

B. Use the answer to evaluate the decimal equivalent of 2A9D16 C. What is the highest decimal number which may be represented by four hexadecimal digits? D. What is the highest decimal number which may be represented by four octal digits?

Numeric Presentation
Fixed Point (Integers) Numeric Data Data Character Data Decimal Numbers Unpacked Decimal Represented using decimal arithmetic Binary Numbers Floating Point (Real Numbers)

Packed Decimal

Decimal digit representation

Binary coded decimal Unpacked decimal format Packed decimal format

Decimal digit representation

o Binary-coded decimal (BCD) code
Uses 4-bit binary digits (correspond to numbers 0 to 9 of decimal system)

Decimal digit representation

BCD code Example:

Decimal digit representation

Unpacked decimal format
Uses 1 byte for each digit of decimal number Represents values from 0 to 9 in least significant 4 bits of 1 byte and in most significant 4 bits (zone bits) Half of a byte is used (excepting the least significant byte) where the least significant half-byte is used to store the sign
1100 = +ve 1101 = -ve

Waste of resources (eliminated by packed decimal format)

Decimal digit representation

Unpacked decimal format

+78910 = F7F8C916

-78910 = F7F8D916

Decimal digit representation

o Packed decimal format
1 byte represents a numeric value of 2 digits the least significant 4 bits represent the sign bit pattern for the sign is the same as per unpacked decimal format

+78910 = 789C16

-78910 = 789D16

Questions
A) Represent 7089310
B) Represent 789310
in Unpacked Decimal Format in Packed Decimal Format in Unpacked Decimal Format in Packed Decimal Format

C) F3F9C116 is represented in standard Unpacked Decimal Format

What is its equivalent in decimal? Possible solution?

D) 3F9C16 is represented in standard Packed Decimal Format

What is its equivalent in decimal? Possible solution?

Decimal digit representation

o Packed decimal format versus Unpacked decimal format
A numeric value can be represented by fewer bytes The conversion into the binary system is easy

Binary Representation
Representation of negative integers
Absolute value representation
0 for positive, 1 for negative

Complement representation
Decimal complement 9s complement 10s complement Binary complement 1s complement 2s complement

Binary Representation
Absolute value representation
Examples
(00001100)2 = (+12)10 (10001100)2 = (-12)10

Issues
(00000000)2 = +0 (10000000)2 = -0

Range of values (assumption: 7-bit absolute value representation used)

-63 to +63 equivalent to (26-1) to +(26-1)

Binary Representation
Complement representation of negative numbers
Decimal complement The subtraction of each of the digits of a numeric value from the complement

Binary Representation
Binary complement
1s complement of a given numeric value is the result of the subtraction of each of the digits of this numeric value from 1, as a result, all the 0 and 1 bits of the original bit string are switched.

Binary Representation
Binary complement
2s complement is 1s complement + 1

Binary Representation
1s complement and 2s complement representation of negative numbers

Binary Representation
Advantages of 2s complement
Less complicated (only one zero value) Range of values to be represented is wider Subtractions can be performed with addition circuits, simplifying hardware structure

Binary Representation
1s complement and 2s complement representation of negative integers
range of represented numeric values when n-bit binary number is represented by adopting the 1s complement method:

-(2n-1 1) to (2n-1 1)
range of represented numeric values when n-bit binary number is represented by adopting the 2s complement method:

-(2n-1) to (2n-1 1)

Binary Representation
Addition circuits only

Binary Representation (Fixed Point)

Fixed point
Integer representation
Fixed point is a data representation format used mainly when integer type data is processed One word is represented in a fixed length (e.g. 16 bits and 32 bits) Overflow problem when attempt is made to represent a numeric value that exceeds the fixed length allocated

Fraction representation
Decimal point is considered to be immediately preceded by the sign bit

Binary Representation (Fixed Point)

Fixed point
Integer representation Range of values
-(2n-1) to (2n-1 1)

Binary Presentation (Fixed Point)

Fixed point
Fraction representation

Binary Representation (Floating Point)

Floating point
Used to represent real number type data Used to represent extremely large or small size of data

Floating point representation format in mainframe computers

This format was adopted in the first general-purpose computer in the world the "IBM System/360" and it was called Excess 64.

Floating point representation format in mainframe computers

Exponent portion
7 bits Range: (0000000)2 to (1111111)2 , which in the decimal system is 0 to 127. However, a numeric value 64 times larger than the real exponent is represented. For that reason, the real exponent is equivalent to 64 to +63. Likewise, since the radix is considered to be 16, the numeric values that can be represented with the exponent portion range between 16-64 to 1663 Then, including the sign bit, the range of numeric values that can be represented with the exponent portion is further increased (see next slide)

Floating point representation format in mainframe computers

Floating point representation format in mainframe computers

Mantissa portion
When the decimal fraction 0.05 is converted into a binary fraction, it becomes a repeating binary fraction. (0.0000110011001100110011001100...)2

Floating point representation format in mainframe computers

Representing 0.0510

Floating point representation format in mainframe computers

Normalisation
Since the mantissa portion has 24 bits, in this case, the decimal fraction 0.05 will not be represented correctly. (The error that occurs in this case is called a rounding error) If we look at the bit pattern of the mantissa portion, we can see that the 4 top bits are 0, if we then extract these 4 bits and shift the remaining bits to the left, 4 rounded bits can be represented. As a result of shifting the mantissa portion 4 bits to the left, the original value of the mantissa portion was increased by 24 = 16. In order to cancel this increase it is necessary to divide it into 16 (16-1). Since the radix is 16, the value of the exponent portion can be set to -1 (63 64 = -1). Used to reduce the rounding error to its minimum as well as to maximize precision. Also known as normalization. Furthermore, as a result of this normalization technique, the bit strings that represent a value are standardized. This operation is performed automatically by the hardware.

Floating point representation format in mainframe computers

http://www.cis.usouthal.edu/faculty/feinstein/50 2/chap2.htm

Binary Representation (Floating Point)

IEEE Floating point representation format
S Exponent portion (8 bits) E
Radix: 2

Mantissa portion (23 bits) F

Only binary fraction lower than 1 can be represented

Mantissa sign (1 bit) 0: Positive 1: Negative

The position of the decimal point is considered to be here

Value represented using the floating point format: (-1)S x 2E-127 x (1 + F) A value resulting from the addition of 127 to the value of the original exponent portion is represented (this addition is called bias)

Bit Shift Operations

Using bit shifts, the multiplication and division of numeric values can be easily performed

Shifting a binary digit 1 bit to the left, its value is doubled.

When a binary number is shifted n bits to the left, its former value is increased 2n times When a binary number is shifted n bits to the right, its former value decreases 2-n times (divided by 2n)

Arithmetic Shift
To calculate numeric values in the fixed point format using 2s complement representation Rules
Sign bit is not shifted Bit shifted out is lost Bit to be filled into the bit position is vacated as a result of the shift is
For left shifts, insert 0 For right shifts, insert the same bit as the sign bit

Logical Shift
To change the bit position Rules
Sign bit is also shifted (moved) Bit shifted out is lost Bit to be filled into the bit position vacated as a result of the shift is 0.

Bit Shifts
(-16)2 to be shifted 2 bits to the right Arithmetic Shift

Bit Shifts
Logical Shift

Operation and Precision

Precision of the numeric value presentation
o The precision of a number is the range of its error o High precision = small error o Single precision
Range of numeric values presentable with 16 bits (in the case of an integer without a sign)
Minimum value = (0000 0000 0000 0000)2 = 0 Maximum value = (1111 1111 1111 1111)2 = 65,535 (values higher than 65,535 cannot be represented)

Operation and Precision

Range of numeric values presentable with 16 bits (in the case of a fraction without a sign)
Minimum value = (0000 0000 0000 0001)2 = 2-16 = 0.0000152587890625000 Maximum value = (1111 1111 1111 1111)2 = 1 2 16 = 0.9999847412109370000 (values lower than 0.00001525878, and values higher than 0.99984741210937 cannot be represented)

Operation and Precision

o Double precision
Number of digits is increased to widen the range of represented numeric values Represent 1 numeric value with 2 words 1 numeric value presentable with 32 bits (in the case of an integer without a sign)
Minimum value = (0000 0000 0000 0000 0000 0000 0000 0000)2 = 0 Maximum value = (1111 1111 1111 1111 1111 1111 1111 1111)2 = 4,294,967,295 (values up to 4,294,967,295 can be represented)

Operation and Precision

Range of numeric values presentable with 16 bits (in the case of a fraction without a sign)
Minimum value = (0000 0000 0000 0000 0000 0000 0000 0001)2 = 2-32 = 0.00000000023283064365387 Maximum value = (1111 1111 1111 1111 1111 1111 1111 1111)2 = 1 2 32 = 0.99999999976716900000000

Operation and Precision

Operation precision
o Precision of fixed point representation
Range of presentable numeric values depends on the computer hardware (number of bits in one word) Range of represented numeric values differs depending on the number of bits in one word Step size of the integer part is always 1 (regardless of number of bits), and only the maximum value changes In the fraction part, the smaller the step size becomes, the error is also reduced

o Precision and underflow

Overflow and underflow
Overflow occurs when product is higher than the maximum value that can be represented with the exponent portion (Maximum absolute value < Overflow) Underflow occurs when product is lower than the minimum absolute value (0 < Underflow < Minimum absolute value)

Operation and Precision

Cancellation
When subtraction of 2 floating point numbers of almost equal values is performed Result becomes extremely small, it is left out of the range of numeric values which can be represented

Loss of information
Addition of extremely small value and extremely large value is performed Exponents adjusted to the exponent of the largest value (mantissa portion of the small value is shifted largely to the right), leading to the loss of information that should have been presented

Non-numeric Value Representation

In order to represent characters using binary digits, codes are used
ASCII, ANSI, UNICode

Character Representation
0 to 9 10 types Uppercase: A to Z Lower case: a to z 52 types Symbolic keys: 40 types Control character keys: 34 types (Space key etc) Numeric keys: Character keys:

To assign a unique bit pattern corresponding to these 136 types of characters and symbols, 256 types of bit patterns i.e. 8 bits are used.

Non-numeric Value Representation

Character codes
o ASCII (American Standard Code for Information Interchange) code
Character code of 8 bits (alphabet, numeric characters, etc.) Used in PCs and data transmission

o ISO (International Organization for Standardization) code

7-bit character code Base of the character codes used in all countries of the world

o JIS (Japanese Industrial Standard) code

Represents 1 character with 2 bytes (16 bits)

Non-numeric Value Representation

o EBCDIC (Extended Binary Coded Decimal Interchange Code)
Established to be used as standards

o Shift JIS (Japanese Industrial Standards) code

Represents 1 character with 2 bytes

o Unicode
2-byte code system unified to all countries To smooth the exchange of data amongst PCs

Non-numeric Value Representation

Audio representation
o Multimedia audio
o Audio analysis is performed using a numeric formula and once it is converted into digital codes it is processed in the computer. o Word processors that accept audio input and speaker recognition are examples of its recent applications.

Image representation
o Image data must be processed to support current multimedia o Image data is processed as a set of dots o Example
o 1 bit is used to register the information of each dot (black, white) o The representation method that combines the basic colors in each dot is used. Systems that combine the three primary colors (Red, Green and Blue) in 256 levels respectively and represent approximately 16,000,000 colors. In this case, since 8 bits are needed for 1 color, in order to register the information of 1 dot, 24 bits are used.

Information and Logic

Proposition Logic Logical operation

Proposition Logic
A proposition is an assertion that something is the case. We use sentences to express propositions. Examples:
(i) The following sentences express the same proposition: - John loves Mary - Mary is loved by John

Proposition Logic
Proposition will always be either true or false Philosophers argue a lot about what constitutes truth. For now, we'll keep it simple:
o "P" is true if and only if P. o "P" is false if and only if not P.

Examples: (i) The proposition "Snow is white" is true if and only if snow is white. (ii) The proposition "Snow is white" is false if and only if snow is not white.

Proposition Logic
Truth table
Proposition 1 Proposition 2 If the wind blows The wind is blowing It is raining it rains True True True

True
False False

False
True False

False
False False

Proposition Logic
Examples
p T T F F q T F T F

p T T F F

q T F T F

p and q T F F F

Logical Operation
A logical operator joins two propositions to form a new, complex, proposition. The truth value of the new proposition is determined by the truth values of the two propositions being joined and by the operator that joins them.

Logical Operation
Negation
o Any proposition p can be converted into its negation with a negation operator, producing the new, complex, proposition:
p means Not p The proposition Not p is true if and only if p is false It is false only if p is true

Logical Operation
o Truth tables for Negation
p T F p F T

p T T F F

q T F T F

Not p F F T T

Logical Operation
Logical Product
o Any two propositions p and q can be connected with the conjunction AND, producing the new, complex, proposition:
p and q (p q) The proposition p and q is true if and only if both p and q are true It is false otherwise

Logical Operation
o Truth tables for Logical product p T T F F q T F T F pq T F F F

Logical Operation
Logical Sum
o Any two propositions p and q can be connected with the conjunction OR, producing the new, complex, proposition:
p or q (p q) The proposition p and q is true if and only if either p or q are true It is false only if both p and q are false

Logical Operation
o Truth tables for Logical sum

p T T F F

q T F T F

pq T T T F

Logical Operation
Exclusive OR
o Any two propositions p and q can be connected with the conjunction EOR, producing the new, complex, proposition:
p eor q (p q) The proposition p eor q is true only if when p or q is true It is false when both p and q are true or false

Logical Operation
o Truth tables for Exclusive OR

p T T F F

q T F T F

p
F T T F

Logical Operation
Negative AND (NAND)
o Any proposition p can be converted into its negation with a negation operator, producing the new, complex, proposition:
Not p The proposition Not p is true if and only if p is false It is false only if p is true

Logical Operation
o Truth tables for Negative AND (NAND)

P T T F F

Q T F T F

P.Q T F F F

Not (P.Q) F T T T

Logical Operation
Negative logical sum (NOR)
o Negation of the logical sum o (p q) Summary of the truth table for the logical operations
P Q NOT p P AND q P OR q P EOR q P NAND q P NOR q

T T F F

T F T F

F F T T

T F F F

T T T F

F T T F

F T T T

F F F T

Logical Operation
Logical expression laws
o Logical symbols
Meaning Negation Logical product Logical sum Exclusive OR Symbols Notation example XY + X+Y X Y

NOT AND

OR EOR

Logical Operation - Laws of logical expressions

Logical product law
Logical sum law Exclusive OR law Commutative law Associative law Distributive law Absorption law Restoring law De Morgans law

X X = X, X
X + X = X, X + X X = 0, X

= 0, X 0 = 0, X 1 = X
= 1, X + 0 = X, X + 1 = 1 = 1, X 0 = X, X 1=

X + Y = Y + X, X Y = Y X X + (Y + Z) = (X + Y) + Z, X (Y Z) = (X Y) Z X + (Y Z) = (X + Y) (X + Z) X (Y + Z) = (X Y) + (X Z) X + (X Y) = X, X (X + Y) = X =X = , = +

Exercises
Use the Laws of Logical Propositions to simplify each of the propositions below to one of the propositions F, T, p, q, p.q, p+q
a) p + q + -p b) p + (q + p) + -q