1

Computer-History, Classification
and Basic Anatomy

1.1 Generations of Computer

The first electronic digital computer, called the Atanasoff–Berry Computer (ABC), was
built by Dr John V. Atanasoff and Clifford Berry in 1937. An electronic computer called the
Colossus was built in 1943 for the US army. Around the same time, many others were also
trying to develop computers. The first general-purpose digital computer, the Electronic
Numerical Integrator and Computer (ENIAC), was built in 1946.
Computers since 1946 are categorized in five generations:
∑ First Generation: Vacuum Tubes
∑ Second Generation: Transistors
∑ Third Generation: Integrated Circuits
∑ Fourth Generation: Microprocessors
∑ Fifth Generation: Artificial Intelligence
Follows a brief description of each generation:

1.1.1 First Generation (1946–1956) Vacuum Tubes

Vacuum tubes were used to make circuits of first generation computers. For building
memory, magnetic drums were used that were huge in size and weight. First generation
computers were so large in size that they often took an entire room. They were also very
prone to error. They were too expensive to operate and in addition consumed huge
electricity. It is worth mentioning the amount of heat they generated. Despite using liquid
based cooling system, they often got damaged due to heat.
Programs for first generation computers were written in machine language, the lowest-
level programming language understood by computers, to perform operations. They were
designed to solve only one problem at a time. Punched cards and paper tapes were used to
feed input, and output was displayed on printouts.
The ENIAC (Electronic Numerical Integrator and Computer) and UNIVAC (Electronic
Discrete Variable Automatic Computer) computers are examples of first-generation
computing devices.

BCPC_01.indd 1 2/9/2016 5:31:54 PM

 2 Basic Computation and Programming with C

Fig. 1.1 ENIAC.

ENIAC was the first electronic general-purpose computer. It was capable of being
reprogrammed to solve various numerical problems. ENIAC was primarily designed to
calculate artillery firing tables. It was mainly used in the United States Army’s Ballistic
Research Laboratory. ENIAC was introduced to the public at the University of Pennsylvania
in 1946 as “Giant Brain.” ENIAC’ was funded by the United States Army.
ENIAC had a modular design. It had individual panels to perform separate functions.
Twenty modules among them were accumulators that could add, subtract and hold a ten-
digit decimal number in memory. Numbers were passed between these modules through
several general-purpose buses. The modules were able to send and receive numbers,
compute, save the answer and trigger the next operation without any moving component.
That is why it could achieve high speed. Key to its versatility was the ability for branching.
It could switch to different operations, depending on the sign of a computed result.
ENIAC contained 17,468 vacuum tubes, 1500 relays, 70,000 resistors, 7200 crystal diodes,
10,000 capacitors. It had a whopping 5,000,000 hand-soldered joints. It weighed more than
27 tons, was roughly 8 × 3 × 100 feet in size, occupied 1800 ft2 and consumed 150 kW of
electricity.
After ENIAC, a much improved computer named EDVAC (Electronic Discrete Variable
Automatic Computer) was designed. EDVAC was a stored program computer. EDVAC
was the first computer to work in binary number system. This is a major difference with
ENIAC that used decimal number system.

1.1.2 Second Generation (1956–1963) Transistors

Transistors replaced vacuum tubes in the second generation of computers. The transistor was
far superior to the vacuum tube in terms of size, generated heat and energy consumption.
So computers made using transistors became smaller, faster, cheaper, more energy-efficient
and more reliable.

BCPC_01.indd 2 2/9/2016 5:31:55 PM

 Computer-History, Classification and Basic Anatomy 3

Fig. 1.2 Transistor.

The transistors also generated a lot of heat that subjected the computer to damage. But
it was much better than the vacuum tubes in terms of size and heat.
Second-generation computers used symbolic or assembly languages instead of binary
machine language that allowed programmers to specify instructions in words instead of
machine code or binary.
Second-generation computers were still using punched cards for input and print-outs
for output. At this time, high-level programming languages, like early versions of COBOL
and FORTRAN were being developed. The first computers that were developed around
this time were used in the atomic energy industry.

1.1.3 Third Generation (1964–1971) Integrated Circuits

Third generation computers were based on integrated circuits (IC circuits). Transistors were
much smaller. They were placed on silicon chips, called semiconductors. This invention
dramatically increased the speed and efficiency of computers.
In these computers, users interacted through keyboards and monitors and interfaced
with an operating system. Many different applications were possible to run at one time. A
central program usually resided in memory to monitor others. Computers for the first time
became accessible to a mass audience because they became cheaper and smaller.

1.1.4 Fourth Generation (1971–Present) Microprocessors

Brain of the fourth generation of computers is microprocessors. Thousands of integrated
circuits were built on a single silicon chip. Computers of the size of an entire room in first
generation could now fit in palm. The Intel 4004 chip, developed in 1971, contained all the
components of the computer like the central processing unit, memory and input/output
controls on a single chip.
In 1981 IBM introduced its first computer for the home user, and in 1984 Apple introduced
the Macintosh. Microprocessors also moved out of the realm of desktop computers and into
many areas of life as more and more everyday products began to use microprocessors.
As these small computers became more powerful, they could now be linked together to
form networks, which eventually led to the development of the Internet. Fourth generation

BCPC_01.indd 3 2/9/2016 5:31:55 PM

 4 Basic Computation and Programming with C

computers were also equipped with the mouse and other handheld devices. Graphical
user interface (GUI) was also designed for these computers.

1.1.5 Fifth generation (Present and Beyond) Artificial Intelligence

Fifth generation computing devices are based on artificial intelligence and are still being
developed, while some applications, like voice and handwriting recognisers, are in used
today. Fifth-generation computing aims to develop devices that will be responsive to
natural language input and will also be able to learn and self-organize.

1.1.6 Evolution of Intel processors

As Intel processors or compatible processors are most popular for desktop systems, it is
worth mentioning the various microprocessors introduced by Intel. The first processor
available to public was 8085 having 40 pin, 2 MHz clock frequency, 6500 transistors, 8
bit data bus and 16 bit address bus. But first personal computer made by IBM with Intel
microprocessor was based on 8086 having 1 MB addressable memory and 30000 transistors.
Then came 80286 followed by 386SX, 386DX, 486 SX and 486 DX.
The 486 processors contained up to 1.4 million transistors, reached 100 MHz frequency
and can address up to 4 Gigabyte of memory (32 bit address bus so 232 = 4 GB). After 486,
came the decade of Pentium, Pentium with MMX (multimedia extended) and Pentium
II processors. Then the Pentium III processor touched the 1 GHz clock frequency mark.
From mid of year 2000 the market leader was various versions of Pentium IV for steady
six years, achieved clock frequency up to 3.8 GHz and packed up to whopping 180 million
transistors in it. Presently core-i3, core-i5 and core-i7 are ruling the market though Intel
dual core processors are also available. (In this span, Intel marketed few other processors
but they did not get good response.)

1.2 Classification of Computers

Computers can be classified in many ways depending on various features and criteria.
Many of them will be mentioned in this section. Though readers of this are being introduced
to computer and do not have an in-depth knowledge of computer architecture, some
classifications based on various architectural difference are also being listed. This will grow
interest of the readers in the subjects such as ‘MICROPROCESSOR’, ‘ARCHITECTURE’,
‘COMPUTER ORGANIZATION’, ‘PARALLEL PROCESSING’, ‘ASSEMBLY LANGUAGE
PROGRAMMING’, and ‘PIPELINING’.
Note A computer's architecture is its abstract model and is the programmer's
view in terms of instructions, addressing modes and registers. A computer's
organization expresses the realization of the architecture. Architecture
describes what the computer does and organization describes how it does it.

Here in this topic, following types of classifications will be discussed:

i. Based of computation power.
ii. Based on number of operands.
iii. Scalar VS Vector processor.

BCPC_01.indd 4 2/9/2016 5:31:55 PM

 Computer-History, Classification and Basic Anatomy 5

iv. Flynn’s taxonomy

v. von Neumann vs. Harvard Architecture
vi. Big-endian vs. Little-endian

(i) Based on Size, Computing Power and Price

Based on their size, computing power and price the computers are broadly classified into
four categories:
1. Microcomputers
2. Minicomputers
3. Mainframe computers
4. Supercomputer.
As the list descends, things will become large, expensive, complex and fast. Now follows
a small introduction to all of them:
(1) Microcomputers are the most common kind of computers used by people today,
whether in a workplace, at school or on the desk at home. These are microprocessor based
systems. Microcomputers are small, low-cost and single-user digital computer. They
consist of CPU, input unit, output unit, storage unit and the software. Microcomputers are
stand-alone machines but they can be connected together to create a network of computers
that can serve more than one user. Microcomputers include desktop computers, notebook
computers or laptop, tablet computer, handheld computer, smart phones, etc. The most
popular microprocessors in world are made by Intel and Apple for their Pentium based
PCs and Apple Macintosh.
Here the evolution of Intel microprocessors will be worth mentioning. Intel first
introduced the 8085 microprocessor having 8 bit data bus and 16 bit address bus but
no personal computer was made based on it. The journey started with 8086 when IBM
made IBM-AT based on it. Gradually came 80286, 386, 486 SX and 486 DX. Next processor
was named Pentium with clock frequency 60 MHz to 300 MHz and introduced in the
period 1993–1997. It had a 32 bit data bus and 32 bit address bus. The speed and internal
cache memory size increased with time. Pentium II, Pentium III and Pentium IV were the
successors. Then comes Pentium DUAL CORE, Core2-DUO, and Quad-core processors
having frequency in GHz level. Now the latest processors from Intel’s stable are core-I3,
core-I5 and core-I7.
(2) Minicomputer is a class of multi-user computers that lies in the middle range of
the computing spectrum, in between the top end single-user systems (microcomputers
or personal computers) and low end mainframe computers. They have high processing
speed and high storage capacity than the microcomputers. Minicomputers can support
up to 200 users simultaneously. The users can access the minicomputer through their PCs
or terminal. They are used for various processor hungry applications in industries and
research centers.
(3) Mainframe computers are multi-user, multi-programming and high performance
computers. They operate at a very high speed, have huge storage capacity and can handle
the workload of many users. The user accesses the mainframe computer via a terminal that
may be a dumb terminal, an intelligent terminal or a PC. The terminals and PCs utilize
the processing power and the storage facility of the mainframe computer. Mainframe

BCPC_01.indd 5 2/9/2016 5:31:55 PM

 6 Basic Computation and Programming with C

computers are used primarily by corporate and governmental organizations for critical
applications, bulk data processing such as census, industry and consumer statistics,
enterprise resource planning and transaction processing.
(4) Supercomputers are the fastest and the most expensive machines. They have
high processing speed compared to other computers. The speed of a supercomputer is
generally measured in FLOPS (Floating point Operations Per Second). Some of the faster
supercomputers can perform trillions of calculations per second. Supercomputers are built
by interconnecting thousands of processors that can work in parallel. Supercomputers are
used for highly calculation-intensive tasks, such as, weather forecasting, climate research,
molecular research, biological research, nuclear research and aircraft design. Some examples
of supercomputers are IBM Roadrunner, IBM Blue gene. The supercomputer assembled in
India by C-DAC (Center for Development of Advanced Computing) is PARAM. PARAM
Padma is the latest machine in this series. The peak computing power of PARAM Padma
is One Tera FLOP.

(ii) Based on Number of Operands

The instruction-set of a computer is designed first, and then accordingly the architecture
is designed. Just like a KEY is designed first then corresponding LOCK is assembled. A
computer instruction should contain one operation code and zero or more operands on
which the operation will be performed or where the result will be stored. Depending on
maximum number of operands allowed in an instruction, computers can be classified:
1. Zero address machines (stack machine)
2. One address machine (accumulator machine)
3. Two address machine (Intel processors are of this type)
4. Three address machine

(iii) Scalar Processor vs Vector Processor

Each instruction executed by a scalar processor generally manipulates one or two data
items at a time. On the contrary, instructions of a vector processor operate simultaneously
on many data items.

(iv) By Flynn’s Taxonomy

Flynn’s taxonomy is a classification of computer architectures, proposed by Michael
J. Flynn in 1966. There are four categories defined by Flynn. His classification is based
upon the number of concurrent instruction (or control) and data streams available in the
architecture:
1. Single Instruction, Single Data stream (SISD) A sequential computer which
exploits no parallelism in either the instruction or data streams. Single control unit (CU)
fetches single Instruction Stream (IS) from memory. The CU then generates appropriate
control signals to direct single processing element to operate on single Data Stream.
2. Single Instruction, Multiple Data streams (SIMD) A computer which exploits
multiple data streams against a single instruction stream to perform operations which may
be naturally parallelized. For example, an array processor or GPU.
3. Multiple Instruction, Single Data stream (MISD) Multiple instructions operate on
a single data stream. Uncommon architecture which is generally used for fault tolerance.

BCPC_01.indd 6 2/9/2016 5:31:55 PM

 Computer-History, Classification and Basic Anatomy 7

4. Multiple Instruction, Multiple Data streams (MIMD) Multiple autonomous

processors simultaneously executing different instructions on different data. Distributed
systems are generally recognized to be MIMD architectures; either exploiting a single
shared memory space or a distributed memory space.

(v) von Neumann vs. Harvard Architecture

While talking about computers, there should be a topic on John von Neumann. This
Hungary-born American mathematician gave the concept of stored-program architecture
first but unfortunately in his life span no fully functional computer was made. His ideas
will be discussed later on this chapter.

von Neumann architecture Harvard architecture

1. Same memory for data & program 1. Separate data and code memory
2. Instructions are executed one by one. Execution 2. Pipelining is possible.
requires at least two clock cycles. One for fetch
and others for execution and pipelining is not
possible.

(vi) Big-endian vs Little-endian

This is based on “How a large binary number is stored in 8 bit (byte) wide memory”
(remember – each memory address can store 1 byte of data). Suppose the following 32 bit
number have to be stored:
11111111 00000000 10101010 11110000
In case of big-endian, the most significant byte is stored in the smallest address. Here’s
how it would look:

Address Data value

400 11111111
401 00000000
402 10101010
403 11110000

In case of little-endian, the least significant byte is stored in the smallest address. Here’s
how it would look:

Address Data value

400 11110000
401 10101010
402 00000000
403 11111111

It can be easily understood by a little thought that “performing arithmetic operations

are easier in little-endian computers where as comparing strings chronologically is easier
in big-endian computers”.

BCPC_01.indd 7 2/9/2016 5:31:55 PM

 8 Basic Computation and Programming with C

1.3 Basic Anatomy of a Computer

A digital computer is an electronic device that receives data, performs arithmetic and
logical operations and produces results according to a predetermined program. It receives
data from input devices and gives results to output devices. The central processing unit,
also known as processor processes the data. Memory (primary and secondary) is used to
store data and instructions. Follows block diagram of a digital computer identifying the
key components and their interconnection.

Fig. 1.3 Components of a computer

The Central Processing Unit (CPU) is like the brain of the computer. It is responsible for
executing instructions. It controls the sequence of execution of instructions. It comprises
a Control Unit (CU), an Arithmetic & Logic Unit (ALU) and huge number of registers.
The CU controls the execution of instructions. First it decodes the instruction and then
generates micro-operations in a particular order with the help of control memory. The
ALU is responsible for performing arithmetic and logic operations.
The interconnections are referred as BUS. Buses are nothing but bunch of wires used to
carry digital signals. There are three kinds of bus:
1. Address bus
2. Data bus
3. Control bus
Address bus carries address of memory from where to read/to where to write data.
Size of address bus of a processor defines the amount of memory addressable by it.
For example a processor with 16 bit address bus can access 216 = 64 KB memory (26 = 64
and 210 = 1024 = 1 K) and a processor with 32 bit address bus can access 232=4 GB memory
(22 = 4 and 230 = 1024 × 1024 × 1024 = 1024 × 1024 × 1 K = 1024 × 1 Mega = 1 Giga.). Address
bus is unidirectional, i.e., it carries signal from CPU to other components (only CPU is
intelligent enough to generate address).
Data bus is bidirectional. It carries data read from/to be written to a device. Its size and
width of registers signify the size of data that can be crunched by the processor in one go.
Control bus carries control signals that activate/deactivate various circuits.

BCPC_01.indd 8 2/9/2016 5:31:55 PM

 Computer-History, Classification and Basic Anatomy 9

Block diagram of the common bus architecture follows:

Fig. 1.4 Common bus architecture.

1.4 Von Neumann Architecture

While studying computer architecture, name of Von Neumann comes first. Though he
could not see any working model based on his proposal in his life span, design of all
modern computers is based on the stored program model proposed by him. In 1945,
the mathematician and physicist John von Neumann, along with others, had described
computer architecture in the First Draft Report on the EDVAC. The Von Neumann
architecture, which is also known as the Von Neumann model and Princeton architecture,
is based on that. In this report a digital computer has been proposed that will contain the
following parts:
∑ A processing unit containing an arithmetic logic unit and processor registers.
∑ A control unit containing an instruction register and program counter.
∑ Memory to store both data and instructions.
∑ External mass storage.
∑ Input and output mechanisms.
Stored-program computers were much advanced as compared to the program-controlled
computers of the 1940s, like the Colossus and the ENIAC. These were programmed by
switches and patches, which led to route data and could control the signals between the
various functional units.
Some key points to remember about this architecture follows:
∑ A program should totally reside in the main memory prior execution.*
∑ Data and code will reside in the same memory and will be indistinguishable.†

*
As the programs become larger and larger, it was not possible to put the total program in the
relatively smaller main memory. So virtual memory was introduced in operating systems. It is a
technique that shows the free part of secondary storage as main memory. It keeps the total program
in the secondary storage in blocks and fetches the required blocks to primary when necessary.
†
Here Von Neumann design differs from HAVARD ARCHITECTURE that uses separate code and
data memory on separate bus.

BCPC_01.indd 9 2/9/2016 5:31:55 PM

 10 Basic Computation and Programming with C

∑ Instructions will be fetched from memory and executed one at a time in a linear
fashion.‡ $$

1.5 Memory Classification and Hierarchy

Broadly classified, a computer system has two types of memory – Primary and
Secondary.
RAM (Random Access Memory), ROM (Read Only Memory) and Cache memory (one
kind of very fast random access memory) falls in the primary category where as Magnetic
tape, CD-ROM, DVD, Hard disk, pen drive (flash memory) falls in the secondary mass
storage category.
Computer system uses memory hierarchy to optimize hardware cost. For example, in
a system where 4 megabyte of cache memory is being used, the size of RAM can be 2 GB
or 4 GB. If a system is build with total 2 GB cache instead, the system cost will go beyond
imagination but speed boost will be only 10% of the system mentioned prior.
If memory is classified in terms of access strategy then there are three categories:
1. Sequential access.
2. Random access.
3. Direct access.
In sequential access memory, access time is directly proportionate to address. Magnetic
tape falls in this category.
In random access memory, access time is constant, i.e., to access content at 1st location
or at millionth location, same time will be required.
All sorts of circular and rotating memories like hard disk, CD, DVD fall in direct access
category where access time can be expressed as a function T = Ax + By where A, B are
constants and x, y are variables. These sorts of memory are divided in tracks and sectors.
Time required by the read/write head to reach a track is known as seek time. Time required
by a sector to reach under read/write head is known as latency time.

1.6 Input and Output Devices

In this section various input and output devices commonly used in computer systems will
be discussed.

1.6.1 Input devices
1.6.1.1 Keyboard
Text information is entered in the computer by typing on the computer keyboard. Most
keyboards, for example – the 101-key US traditional keyboard or the 104-key Windows
keyboard, have alphabetic and numeric characters, punctuation marks and function keys.
Keyboards are generally connected to the computer by a PS/2 connector or USB port.
‡
To speed up execution, multiple instructions are fetched in a queue in a pipelined computer.

BCPC_01.indd 10 2/9/2016 5:31:55 PM

 Computer-History, Classification and Basic Anatomy 11

Fig. 1.5 Keyboard.

Fig. 1.6 USB and PS/2 ports.

1.6.1.2 Mouse
Mouse is a pointing device. The cursor on the screen is moved by moving the mouse.
A mouse have mainly two buttons – left and right. Newer mouses may contain other
buttons and a roller for scrolling. In older models, a ball in the bottom of the mouse rolls
on the surface as the mouse moves, and internal rollers sense the movement of the ball and
transmit the information to the computer via the mouse cable. The newer optical mouse
does not use a rolling ball, instead it uses a light and a small optical sensor to detect the
motion of the mouse by tracking a tiny image of the desk surface. Even cordless mouse are
very much affordable now a days, but they need regular battery change that reduces their
popularity. Mouse sends two information to the computer – one is the X and Y coordinate
of the pointer merged in a single number, another is the code of key pressed. Any activity
on the mouse generates an interrupt that ultimately causes an event such as mouse move,
left click, double click, right click, and drag to the operating system. What incidents will
happen with the muse events are completely programmable.

BCPC_01.indd 11 2/9/2016 5:31:56 PM

 12 Basic Computation and Programming with C

1.6.1.3 Scanners
A scanner is a device that converts a printed page or graphic to an image file that can be
stored in computer by digitizing it. It produces an image made of tiny pixels of different
brightness and color values which are represented numerically and sent to the computer.

Fig. 1.7 Scanner.

Pages containg text is also scanned and saved as images. To convert such images
containing text to editable text files OCR (Optical Character Recognition) software is used.
Scanners are available in various sizes with various scan resolution capability. A4 size
scanners (largest page size it can scan is A4-8.27≤ × 11.69≤) are most popular and affordable.
Scanners with automatic document feeder is also available.

1.6.1.4 Microphone
A microphone can be attached to a computer through a sound card input or circuitry built
into the motherboard to record sound. The sound is digitized and stored in the computer.

Fig. 1.8 Microphone.

Our ear can hear analog sound of frequency range 20 to 20 MHz. But for storing sound in
computer, the analog signal is converted to digital signal. Again when the sound is played

BCPC_01.indd 12 2/9/2016 5:31:56 PM

 Computer-History, Classification and Basic Anatomy 13

back through speakers, the stored digital signal is converted to analog. (The process of
digitization is beyond the scope of this book.)

1.6.2 Output devices
1.6.2.1 Monitor
Monitor is the maximum used output device of a personal computers. At the begining,
CRT-(Cathode Ray Tube) display in monochrome with low resolution was available.
Gradually high resolution colour monitors was introduced. Presently bulky CRT monitors
has been mostly replaced by LCD or LED monitors as LCD or LED display consumes less
electricity, takes less space and capable of displaying more clear and flicker free images.

1.6.2.2 Printer
When some output on paper is needed, a printer is must. The three most common types
of printers are Dot matrix, Inkjet and Laser. Dot matrix printers provide poorest quality
output in lowest cost. It is mainly used for text based outputs on pre-printed stationary
such as cash-memo and ticket.

Fig. 1.9 Printers.

Laser printers produce best quality printouts using powder formed inks filled in toner
cartiges. Cost of laser printers are high but printing cost is less than that of a deskjet printer
that uses liquid inks with magnetic particles filled in ink cartiges.

Exercises
1. Write short notes on Von Neumann architecture.
2. Differentiate between Von Neumann and Havard architecture.
3. With a neat block diagram, describe various components on a computer.
4. Classify memories in terms of access strategy.
5. “All ROMs are RAM but all RAMs are not ROMs” – Justify.
6. For a medicine or grocery store what kind of printer you will suggest and why?
7. Write down various ports availble in a computer system.
8. What is a BUS? How many types of BUS is available?
9. Write full form of ENIAC.
10. What do you mean by Virtual Memory. What is the largest possible size of it? Why it is called
virtual?

BCPC_01.indd 13 2/9/2016 5:31:56 PM