Computer System

A computer is not one machine but a system — a set of parts, each doing one job, wired together so that data flows from input, through processing, to output, while instructions and results sit in memory.

This unit takes the system apart: first the history and the classification, then the architecture, then every physical component you will be asked to name, compare and describe.

At 20 teaching hours it is the largest unit in Grade 11 Computer Science and it supplies more exam marks than any other.

1.1 Definition, characteristics and application of computer

The processing cycle is usually written as the IPO cycle — Input → Process → Output — with Storage added as a fourth stage, giving IPO-S.

Characteristics.

Characteristic	What it means
Speed	Operations are measured in millions or billions per second; internal events take nanoseconds
Accuracy	Errors come from wrong data or a wrong program, not from the machine — GIGO: Garbage In, Garbage Out
Diligence	It does not tire, get bored or lose concentration; the millionth calculation is as good as the first
Versatility	The same hardware runs a spreadsheet, a game and a weather model — only the program changes
Storage capacity	Terabytes of data can be held and retrieved in a fraction of a second
Automation	Once started, a program runs to completion without human help
Reliability	Consistent results over long periods with very low failure rates
No IQ / No feeling	It has no intelligence of its own; it obeys instructions literally

Characteristics explained. Each characteristic below is written the way a 10-mark answer expects it: name, meaning, and a Nepali example.

1. Speed. A modern processor performs billions of simple operations every second, and one operation takes only a few nanoseconds. Work that would keep a clerk busy for a week can finish before you blink.

For example, the examination board can sort and grade the marks of lakhs of students in minutes, where manual tabulation once took many days. Speed is the reason computers are used wherever a huge amount of data must be handled quickly.

2. Accuracy. A computer performs arithmetic in exactly the same way every time, so the machine itself does not make random mistakes.

Wrong output comes from wrong input data or a wrong program, which is summed up by the rule GIGO (Garbage In, Garbage Out).

When a bank adds interest to thousands of accounts, every rupee is calculated to the same number of decimal places. If a teacher types a wrong mark, however, the computer will faithfully print the wrong result.

3. Diligence. A human becomes tired, bored and careless after hours of repetitive work, but a computer does not. It carries out the millionth calculation with the same care as the first one.

An electricity billing system, for instance, can process the meter readings of every household month after month without a break. This makes the computer ideal for dull, repetitive jobs.

4. Versatility. A computer is a general-purpose machine: the same hardware does very different jobs when the program is changed.

In one hour a student may type a letter, watch a lecture, run a spreadsheet and play a game on a single laptop. A bank terminal, a school server and a smartphone differ mainly in the software they run.

This ability to change tasks is what separates a computer from a calculator.

5. Storage capacity. A computer can keep a very large amount of data in a small space and bring it back in a fraction of a second. A single 1 TB disk holds about 250,000 photographs or millions of pages of text.

National identity records, land records and examination records are kept this way instead of in room-sized cupboards. Stored data can also be searched, copied and backed up easily.

6. Automation. Once a program and its data are loaded, the computer carries out the instructions one after another without help from a person.

A payroll program can calculate salary, tax and deductions for every employee and print the payslips while the operator is away. Automation saves labour and removes the waiting time between steps.

The ATM that counts and releases notes after a few key presses is a small, familiar example.

7. Reliability. A well-built computer gives the same result for the same input over a long period and fails very rarely. Banks, airlines and hospitals depend on this, and they add backup power and duplicate systems to keep running.

A Fonepay or ConnectIPS payment, for example, must complete correctly at any hour of the day. High reliability is what lets people trust a computer with money and health records.

8. No intelligence and no feelings. A computer cannot think for itself; it obeys the instructions of the programmer literally, even when they are foolish. It cannot judge, feel tired, or become angry or happy.

Only a program written by a human tells it what to do, so the human remains responsible for the result. This limitation is also why artificial intelligence is such an active field of research.

Applications explained. Eight areas, each with the Nepali example an examiner likes to see.

1. Education. Schools and colleges use computers for e-learning, online classes, digital libraries, result processing and student records.

The National Examinations Board publishes results on its website and by SMS, and university portals handle forms and results for thousands of students.

Teachers prepare lessons with presentations and simulations, and students practise with educational software. Computers make learning possible even when a student cannot reach a classroom.

2. Banking and finance. Banks use computers for core banking, ATMs, cheque clearing with MICR, mobile banking and card payments. ConnectIPS moves money between bank accounts, and Fonepay lets you pay in a shop by scanning a QR code.

Because every branch connects to one central database, you can deposit in Kathmandu and withdraw in Jumla. Computers also keep the records safe and make audits faster.

3. Government and public services. Government offices use computers for the national ID card, passport and driving-licence systems, land records, tax collection and customs.

The Nagarik App brings digital documents and several services onto one mobile screen. This cuts paperwork, queues and the chance of files being lost or hidden. Citizens can apply, pay and track many services from home.

4. Health and medicine. Hospitals keep patient records, appointments and bills on computers, and CT, MRI and ultrasound machines need computer processing to form their images.

Telemedicine lets a health worker in a remote hill village consult a specialist in Kathmandu over a video call. Computers also help to manage vaccination records and medicine stocks. Faster and clearer diagnosis can save lives.

5. Science, engineering and weather forecasting. Scientists use computers to run models that would take years by hand.

The Department of Hydrology and Meteorology (DHM) processes weather and river-flow data to issue forecasts and flood warnings. Engineers use CAD software to design bridges, roads and buildings and to test them on screen before building.

Earthquake research in Nepal also depends on the computer analysis of seismic data.

6. Business and industry. Shops and companies use computers for billing, inventory, accounting and payroll. A supermarket's barcode scanner finds the price and reduces the stock count at the same moment.

Factories use computer-controlled machines and robots for accurate, repeatable work. Even a small business can keep exact records and make better decisions with a spreadsheet.

7. Communication and media. Email, video calls, messaging and social media all run on computers and networks.

Newspapers, FM stations and TV channels in Nepal use computers for writing, editing and broadcasting, and many Nepalis abroad talk to their families every day over video calls. Online news reaches readers the moment it is published.

Communication that once took days by post now takes seconds.

8. Transport, tourism and entertainment. Computers run online bus and air ticketing, air-traffic control and GPS navigation. Trekkers book permits, guides and hotels online, and hotels manage bookings with software.

Films use computer animation, and games are a large form of entertainment. Traffic lights and vehicle registration in the cities are also computer controlled.

1.2 Evolution of computer technology (generations)

Before electronics, calculation was mechanical: the abacus (about 3000 BC), Napier's Bones (1617), Pascal's Pascaline (1642), Leibniz's stepped reckoner (1671), Jacquard's punched-card loom (1801) and Charles Babbage's Difference Engine (1822) and Analytical Engine (1833).

Because the Analytical Engine already had a mill, a store, input and output, Babbage is called the Father of Computer, and Ada Lovelace, who wrote its algorithms, the first programmer.

Herman Hollerith's punched-card tabulator (1890) led to the company that became IBM.

Electronic computers are grouped into five generations, each defined by the switching technology it uses.

Generation	Switching device	Memory	Language	Typical speed
First	Vacuum tube	Magnetic drum	Machine (binary)	milliseconds
Second	Transistor	Magnetic core	Assembly, early FORTRAN/COBOL	microseconds
Third	Integrated circuit (SSI/MSI)	Magnetic core, early semiconductor	High-level, OS with multiprogramming	nanoseconds
Fourth	Microprocessor (LSI/VLSI)	Semiconductor RAM	3GL, 4GL, GUI	nanoseconds
Fifth	ULSI, parallel and AI processors	Very large semiconductor, optical	Natural language, AI	picoseconds

The fourth generation begins with the Intel 4004 (1971) — the first microprocessor, a 4-bit chip of about 2,300 transistors clocked at roughly 740 kHz. The IBM PC of 1981 used the 16-bit Intel 8088.

(a) 1st gen: vacuum tube, 1940s-50s

(b) 2nd gen: transistor, late 1950s

(c) 3rd gen: integrated circuits (SSI/MSI)

(d) 4th gen: VLSI microprocessor

(e) 5th gen: AI accelerator board

(a) ENIAC: 17,468 tubes, so a tube failed every day or two

(b) IBM 1401 (1959), used for the 2028 BS census

(c) Punched card: 80 columns, one hole pattern per character

The five generations in detail. A ten-mark question asks for the technology, the machines, the advantages and the drawbacks of every generation, so learn each paragraph below as a unit.

First generation (1946-1958): vacuum tubes. The first electronic computers, ENIAC (1946) and UNIVAC I, used thousands of vacuum tubes as switches. A tube is a glass bulb with a heated filament, and it gets very hot.

ENIAC held about 17,000 tubes, weighed roughly 27 tonnes and filled a large hall. Programs were written in machine language (binary) and entered by plugging wires or by punched cards, and the memory was a magnetic drum.

Advantages the first fully electronic machines, thousands of times faster than mechanical calculators.

Limitations huge size, very high power use and heat, frequent tube failure, high cost, a single program at a time and no operating system.

Second generation (1959-1964): transistors. The transistor, invented at Bell Laboratories in 1947, replaced the tube. It is a tiny solid-state switch that is smaller, cooler, cheaper and far more reliable.

Magnetic core memory appeared, and assembly language and the first high-level languages, FORTRAN and COBOL, replaced pure binary. Machines such as the IBM 1401 handled business accounts and census data.

Advantages smaller, faster, more reliable and cheaper than first generation machines.

Limitations still costly, still needed air-conditioning, and programs were written for one particular model.

Third generation (1965-1970): integrated circuits. An integrated circuit (IC), invented by Jack Kilby in 1958, puts many transistors, resistors and capacitors on one small silicon chip.

The IBM System/360 (1964) and the PDP-8 are typical machines.

Small-scale and medium-scale integration (SSI and MSI) brought operating systems with multiprogramming, keyboards and monitors in place of punched cards, and high-level languages such as BASIC and Pascal.

Advantages smaller size, lower power, higher speed and reliability, and lower cost.

Limitations ICs were still hard to make, and computers were not yet cheap enough for homes.

Fourth generation (1971-present): microprocessors. The microprocessor puts the whole CPU on one chip. The Intel 4004 (1971) held about 2,300 transistors; today a processor holds billions (LSI, then VLSI).

The result was the personal computer: the IBM PC (1981), Apple Macintosh, laptops and smartphones. This generation brought the GUI, mouse, networks, the Internet, semiconductor RAM and fourth-generation languages.

Advantages small, cheap, very fast, reliable and available to ordinary people.

Limitations the machine still has no intelligence of its own, and e-waste and cyber-crime became new problems.

Fifth generation (present and beyond): ULSI, AI and parallel processing. The aim is a computer that can learn, reason and understand speech.

Features include very large chips with many cores, artificial intelligence, machine learning, voice assistants, robotics, parallel and cloud computing, and early quantum computers. Natural-language interfaces are replacing menus.

Advantages it can handle huge data, recognise images and speech, and make predictions.

Limitations the systems are expensive, need huge data and power, and raise questions about privacy, jobs and fairness.

Comparison of the five generations.

Feature	1st	2nd	3rd	4th	5th
Period	1946-58	1959-64	1965-70	1971-now	emerging
Switching device	Vacuum tube	Transistor	IC (SSI/MSI)	Microprocessor (LSI/VLSI)	ULSI, AI chips
Size	A room	A large cabinet	A cabinet	Desktop, laptop, pocket	Pocket to cloud
Heat and power	Very high	High	Moderate	Low	Low per operation
Reliability	Very low	Moderate	High	Very high	Very high
Input and output	Punched cards, plugboards	Cards, printers	Keyboard, monitor	Mouse, GUI, touch	Voice, gesture, sensors
Software	Machine language	Assembly, FORTRAN, COBOL	OS, BASIC, Pascal	GUI, 4GL, Internet	AI, natural language
Examples	ENIAC, UNIVAC-I	IBM 1401	IBM 360, PDP-8	IBM PC, Macintosh	AI systems, quantum

Eight improvements from generation to generation.

1. Size. A first-generation computer filled a hall, but a smartphone today is far more powerful and fits in a pocket. Smaller switches, from tube to transistor to chip, made this possible.

2. Speed. Switching time fell from milliseconds to nanoseconds and then to picoseconds. Billions of operations per second are now normal.

3. Power and heat. A tube needed a heater, while a chip uses a tiny current. Less heat means no huge cooling system and lower electricity bills.

4. Cost. Mass production of ICs and microprocessors made each generation cheaper per operation. This is why a school lab can now afford dozens of computers.

5. Reliability. A tube burned out like a bulb, but a chip has no filament and can run for years. Fewer failures mean less repair time and more trust.

6. Memory. Memory moved from the magnetic drum to core to semiconductor RAM and SSDs. Capacity rose from a few kilobytes to many gigabytes.

7. Software. Programmers moved from binary to assembly, high-level languages, GUIs and AI tools. Programs became easier to write and to use.

8. Access. Mainframes belonged only to governments and large companies. Personal computers, the Internet and mobile phones put computing within reach of students and villagers.

1.3 Measurement unit of processing speed and storage unit

Speed. Two different things are measured, and students mix them up.

Quantity	Unit	Meaning
Clock frequency	hertz (Hz), MHz, GHz	clock pulses per second; 1\ \text{GHz} = 10^{9} cycles s⁻¹
Cycle / access time	ms, μs, ns, ps	time for one cycle or one memory access
Instruction rate	MIPS	millions of instructions per second
Arithmetic rate	FLOPS, GFLOPS, TFLOPS, PFLOPS, EFLOPS	floating-point operations per second

Clock period and frequency are reciprocals: T = 1/f. A 3.2 GHz processor has a clock period of 1/(3.2\times10^{9}) = 0.3125 ns.

Storage. The smallest unit is the bit (binary digit, 0 or 1). Four bits make a nibble, eight bits a byte — one byte stores one character in ASCII.

Unit	Symbol	Binary value	Approximate size
Kilobyte	KB	2^{10} = 1,024 B	half a page of text
Megabyte	MB	2^{20} B	a 3-minute MP3 song ≈ 3 MB
Gigabyte	GB	2^{30} B	a 2-hour HD film ≈ 4 GB
Terabyte	TB	2^{40} B	a typical 2026 laptop hard disk
Petabyte	PB	2^{50} B	a large national data centre
Exabyte	EB	2^{60} B	global internet traffic in a few hours

1.4 Super, Mainframe, Mini and Microcomputers

Digital computers are classified by size, processing power and cost.

Type	Processing power	Users at once	Typical use	Example
Supercomputer	Fastest; measured in PFLOPS/EFLOPS	few, batch jobs	weather modelling, nuclear simulation, genome research	El Capitan, Frontier, Fugaku
Mainframe	Very high throughput, huge I/O	thousands of terminals	bank core systems, census, airline reservation	IBM z-series
Minicomputer (midrange)	Medium	tens to hundreds	departmental servers, process control	PDP-11, IBM AS/400
Microcomputer	One microprocessor	usually one	desktop, laptop, tablet, smartphone	PC, MacBook, smartphone

A supercomputer gets its speed from massive parallelism, not from one very fast processor.

As of the November 2025 TOP500 list, the fastest machine in the world was El Capitan at Lawrence Livermore National Laboratory, USA, rated at 1.809 exaFLOPS (about 1.8\times10^{18} floating-point operations per second) on the LINPACK benchmark, using more than 11 million processor cores.

A mainframe is optimised for throughput and reliability rather than raw arithmetic: it moves enormous volumes of transactions with near-continuous uptime, which is why banks and airlines still use them.

Microcomputers are further divided into desktop, laptop, notebook, netbook, tablet, palmtop/PDA, smartphone and workstation (a high-end single-user machine for CAD or video editing).

1.5 Mobile Computing and its Application

It rests on three legs: mobile hardware (smartphone, tablet, laptop, wearable), mobile software (mobile OS and apps) and mobile communication (Wi-Fi, Bluetooth, 4G LTE, 5G, NFC, GPS, satellite).

Applications. Mobile banking and digital wallets (eSewa, Khalti, ConnectIPS and fonepay QR payments in Nepal); m-commerce; m-learning; telemedicine and tele-consultation in remote districts; GPS navigation and ride hailing (Pathao, InDrive); field data collection for the census and for disaster response; agriculture advisories by SMS; and mobile ticketing.

Advantages — mobility, immediate access, location awareness, lower infrastructure cost. Limitations — limited battery, small screen, weaker processing and storage, dependence on network coverage, and greater security and privacy risk.

1.6 Concept of computer architecture and organization

These two terms are examined together almost every year because they are easy to confuse.

Architecture	Organization
Logical / functional view	Physical / implementation view
Visible to the programmer	Transparent to the programmer
Decided first, changes rarely	Decided later, changes with each model
Instruction set, word size, addressing modes	Control signals, bus widths, memory chips used
e.g. "the x86-64 instruction set"	e.g. "how L2 cache is wired in this particular CPU"

Two classical architectures appear in the syllabus:

• Von Neumann (stored-program) architecture — one memory holds both instructions and data, and one bus carries both. Simple and cheap, but the single path between CPU and memory limits speed; this is the von Neumann bottleneck. Almost all general-purpose computers use it.
• Harvard architecture — separate memories and separate buses for instructions and data, so both can be fetched at once. Used in microcontrollers and in DSP chips, and inside modern CPUs at the cache level (separate instruction and data L1 caches — a "modified Harvard" design).

1.7 Components of computer system: input, output, processing, memory and storage

A computer system = hardware + software + data + people (liveware) + procedures. The hardware is organised into five functional units.

1. Input unit — converts data from a human-readable form into binary and feeds it to the CPU.
2. Central Processing Unit (CPU) — the "brain". It contains: - the Control Unit (CU), which fetches instructions, decodes them and generates the control signals that direct every other unit (it does not do any arithmetic itself); - the Arithmetic and Logic Unit (ALU), which performs arithmetic (+, −, ×, ÷) and logic (AND, OR, NOT, comparison) operations; - registers, small very fast stores inside the CPU that hold the data being worked on right now.
3. Memory unit (primary memory) — holds the program and data currently in use; directly addressable by the CPU.
4. Storage (secondary memory) — keeps data permanently; not directly accessible by the CPU, data must be copied to RAM first.
5. Output unit — converts binary results back into human-readable form.

The CU and ALU together with registers form the CPU; the CPU and primary memory together are sometimes called the main unit or system unit.

1.8 Microprocessor: basic concepts, clock speed, word length, components and functions

Components and their functions.

Component	Function
ALU	arithmetic and logical operations
Control unit	decodes instructions, generates timing and control signals
Register array	temporary high-speed storage (accumulator, PC, MAR, MDR/MBR, IR, general registers)
Clock / timing circuit	synchronises every operation
Cache (L1, L2, L3)	keeps recently used instructions and data close to the cores
Bus interface unit	connects the chip to the external address, data and control buses

Key registers. Program Counter (PC) holds the address of the next instruction; Memory Address Register (MAR) holds the address being accessed; Memory Data Register (MDR) holds the word in transit; Instruction Register (IR) holds the instruction being decoded; the Accumulator (A) holds ALU results.

Clock speed is the number of clock pulses per second, in MHz or GHz. Each instruction takes a whole number of clock cycles, so — for one given design — a higher clock means more instructions per second.

Typical desktop processors in 2026 run at a base clock of 3–4 GHz with boost clocks above 5 GHz.

Word length is the number of bits the CPU can fetch, store and process as a single unit — 8, 16, 32 or 64 bits on real machines. A longer word means more precision, larger directly-addressable memory and greater speed.

A 32-bit processor can address 2^{32} = 4\ \text{GB} of memory; a 64-bit processor can in principle address 2^{64} bytes, which is 16 exabytes.

1.9 Bus System: data bus, address bus and control bus

Bus	Direction	Carries	Width decides
Address bus	unidirectional (CPU → memory/IO)	the address of the location to be read or written	how much memory can be addressed: 2^{n} locations
Data bus	bidirectional	the actual data or instruction word	how many bits move per transfer (usually = word length)
Control bus	bidirectional (individual lines)	READ, WRITE, CLOCK, RESET, INTERRUPT, memory/IO select, bus request/grant	timing and coordination

1.10 Primary memory: RAM, ROM, Cache, Buffer

Primary (main) memory is semiconductor memory that the CPU can address directly. It is fast, expensive per byte, and small compared with storage.

RAM (Random Access Memory) is read/write, volatile (contents lost when power goes off) and used as working memory. "Random access" means any location takes the same time to reach, unlike tape.

	SRAM	DRAM
Storage cell	flip-flop (4–6 transistors)	one transistor + one capacitor
Refresh needed	no	yes, thousands of times per second
Speed	very fast	slower
Cost and density	expensive, low density	cheap, high density
Power	higher (static)	lower per bit
Used as	cache memory	main memory

DRAM has evolved SDRAM → DDR → DDR2 → DDR3 → DDR4 → DDR5, which is the standard for new PCs in 2026 with module speeds commonly from 4800 to about 8000 MT/s.

ROM (Read Only Memory) is non-volatile and holds firmware — the BIOS/UEFI that starts the machine (the bootstrap program), and the control programs of embedded devices.

ROM type	Written by	Erasable?
MROM (Masked ROM)	the manufacturer, during fabrication	no
PROM (Programmable ROM)	the user, once, with a PROM burner	no
EPROM (Erasable PROM)	the user	yes — ultraviolet light through a quartz window, whole chip
EEPROM (Electrically Erasable PROM)	the user	yes — electrically, byte by byte
Flash ROM	the user	yes — electrically, in blocks; fast; used for BIOS and pen drives

Cache memory is a small, very fast SRAM between the CPU and main memory that holds recently and frequently used instructions and data. When the CPU finds what it wants in cache that is a hit, otherwise a miss.

Levels: L1 (on each core, smallest, fastest), L2 (per core, larger), L3 (shared by all cores). Cache works because of locality of reference.

Buffer is a temporary holding area in RAM (or inside a device) that compensates for the difference in speed between two units — for example a print buffer lets the CPU dump a document and carry on while the slow printer prints, and a video buffer stores the next few seconds of a stream.

1.11 Secondary Memory: Magnetic Disk, Flash Memory, Optical Disk, External Storage

Secondary memory is non-volatile, large, cheap per byte and not directly addressable by the CPU.

Magnetic disk. A hard disk drive (HDD) stores bits as magnetised spots on rotating aluminium or glass platters coated with magnetic material.

Each surface is divided into concentric tracks, each track into sectors (traditionally 512 bytes, now usually 4096 bytes in Advanced Format). Tracks at the same radius on all surfaces make a cylinder.

Access time = seek time + rotational latency + transfer time. Drives spin at 5400, 7200, 10000 or 15000 rpm.

In 2026 desktop HDDs of 1–8 TB are ordinary and data-centre drives using HAMR technology reach 30–36 TB, with 40 TB+ models shipping to hyperscale customers. Magnetic tape is still used for cheap archival backup.

Flash memory is EEPROM-based solid-state storage with no moving parts: pen drives, memory cards (SD, microSD) and SSDs.

It is fast, silent, shock-proof and low-power, but has a limited number of write cycles and costs more per gigabyte than an HDD. Consumer SSDs run to 8 TB (M.2 NVMe); enterprise flash drives of about 245 TB exist as of 2026.

Optical disk. Data is written as microscopic pits and lands along one long spiral track, read by reflecting a laser beam off the surface.

Disk	Laser	Capacity	Notes
CD-ROM / CD-R / CD-RW	infrared, 780 nm	700 MB	audio and small data
DVD	red, 650 nm	4.7 GB single layer, 8.5 GB dual layer	film, software
Blu-ray (BD)	blue-violet, 405 nm	25 GB single layer, 50 GB dual layer	HD video
BDXL	blue-violet	100 GB / 128 GB	archival

External storage devices connect through a port rather than sitting inside the case: external HDD and SSD (USB or Thunderbolt), pen drive, memory card, network attached storage (NAS) and cloud storage (Google Drive, OneDrive).

1.12 Input Devices

Device	How it works	Typical use
Keyboard	each key closes a switch; the keyboard controller sends a scan code	text entry; QWERTY layout, 104 keys
Mouse	optical sensor or ball tracks movement; pointing device	GUI pointing, drag and drop
Scanner	a light source and CCD/CIS sensor digitise a page into an image (measured in dpi)	copying documents and photographs
Light pen	a photodetector in a pen senses the screen and reports position	drawing directly on the screen, CAD
OMR (Optical Mark Reader)	detects the presence of pencil marks in fixed positions by reflected light	multiple-choice answer sheets, Loksewa and entrance exams
OCR (Optical Character Recognition)	software recognises the shapes of printed or written characters and converts them to editable text	digitising books and forms
BCR (Bar Code Reader)	a laser scans the bar pattern; reflected light gives a digit string (e.g. EAN-13)	supermarket billing, library, inventory
MICR (Magnetic Ink Character Recognition)	reads characters printed in magnetic ink in the E-13B font	bank cheque clearing, used by Nepali banks
Touch screen	resistive or capacitive layer detects the position of a finger	ATMs, smartphones, kiosks, POS
Microphone	a transducer turns sound pressure into an electrical signal, digitised by the sound card	voice input, recording, online class
Digital camera / webcam	a lens focuses light onto a CCD/CMOS sensor; resolution in megapixels	photographs, video conferencing

Other input devices worth naming: joystick, trackball, graphics tablet, biometric (fingerprint, iris) scanner, QR code reader, and sensors of all kinds.

1.13 Output Devices

Output is either soft copy (temporary — screen, sound) or hard copy (permanent — paper).

Monitor. The Visual Display Unit. Quality is described by resolution (pixels, e.g. HD 1366×768, Full HD 1920×1080, 4K UHD 3840×2160), refresh rate (Hz), dot pitch and colour depth.

	LCD	LED	OLED
Backlight	cold cathode fluorescent lamp (CCFL)	light-emitting diodes	none — each pixel emits its own light
Thickness	thicker	thin	thinnest
Power	moderate	lower	lowest for dark images
Contrast / black level	moderate	better	best (true black)
Cost	lower	moderate	highest

Strictly, an "LED monitor" is an LCD panel with an LED backlight.

Printers. Divided into impact (the head strikes the paper through a ribbon) and non-impact.

	Dot matrix	Inkjet	Laser
Type	impact	non-impact	non-impact
Mechanism	9 or 24 pins strike an inked ribbon	tiny nozzles spray liquid ink droplets	laser draws a charged image on a drum; toner is fused by heat
Quality	low (visible dots)	good, photo-quality	excellent, sharp text
Speed unit	characters per second (cps)	pages per minute (slow)	pages per minute (fast)
Noise	noisy	quiet	quiet
Running cost	very low	high (cartridges)	low per page
Best for	multi-part carbon bills, bank passbooks	photos, home use	offices, high volume

Also: plotters (large engineering drawings), 3D printers, projectors (LCD/DLP).

Speaker. Converts the digital audio signal — after the sound card's digital-to-analogue converter — into sound waves. Output is described by power (watts) and frequency response; headphones and earphones are personal versions.

1.14 Hardware Interfaces: ports, slots and standards

Interface	Transmission	Connector	Speed	Status
Parallel port	8 bits side by side on 8 wires	DB-25 female	up to about 2.5 MB s⁻¹ (ECP)	obsolete; was used for printers
Serial port	1 bit at a time on one wire	DB-9 (RS-232)	up to 115.2 kbps	obsolete; was used for modems, mice
USB	serial, differential pair, hot-swappable, supplies power	Type-A, Type-B, micro, Type-C	1.1: 12 Mbps · 2.0: 480 Mbps · 3.2 Gen1: 5 Gbps · Gen2: 10 Gbps · Gen2×2: 20 Gbps · USB4: 40 Gbps · USB4 v2: 80 Gbps	current standard
HDMI	digital video and audio on one cable	19-pin Type-A	1.4: 10.2 Gbps · 2.0: 18 Gbps · 2.1: 48 Gbps · 2.2: 96 Gbps	current standard
Expansion slot	parallel/serial lanes into the system bus	ISA, PCI, AGP, PCI Express	PCIe 4.0 ×16 ≈ 32 GB s⁻¹; PCIe 5.0 ×16 ≈ 63 GB s⁻¹	PCIe only

Despite the name, a parallel port is slower than a modern serial port: pushing eight bits simultaneously down eight long wires causes skew and crosstalk, so parallel links cannot be clocked fast.

Modern high-speed links (USB, SATA, PCIe, HDMI) are all serial for exactly this reason.

USB Power Delivery now carries up to 240 W, enough to charge a laptop over the same cable that carries the data.

HDMI version 2.2, released in June 2025, raised the maximum bandwidth to 96 Gbps and requires the new Ultra96 certified cable; it supports up to 12K at 120 Hz.

Expansion slots are connectors on the motherboard into which add-on cards (graphics, sound, network, capture) are plugged.

The historical order is ISA → PCI → AGP (graphics only) → PCI Express, which uses point-to-point serial lanes (×1, ×4, ×8, ×16) instead of a shared parallel bus.

(a) The Intel 4004, the first microprocessor (1971)

(b) Transistors per chip, 1971 to 2020 (log scale)

Chapter summary

• A computer accepts input, processes it under a stored program, stores data and produces output: the IPO-S cycle. Its characteristics are speed, accuracy, diligence, versatility, storage, automation and reliability — with no IQ.
• Five generations: vacuum tube → transistor → IC → microprocessor → AI/ULSI. Babbage is the Father of Computer; the Intel 4004 (1971) began generation four.
• Speed units: Hz/MHz/GHz, MIPS, FLOPS; time units ms, μs, ns, ps with T = 1/f. Storage: 8 bits = 1 byte, and 1\ \text{KB} = 2^{10} bytes, each further step \times 2^{10}.
• Classification by power: supercomputer > mainframe > mini > microcomputer.
• Architecture = what the machine does (programmer's view); organization = how it is built (implementation).
• Block diagram: input → CPU (CU + ALU + registers) ↔ primary memory ↔ secondary storage → output, with control signals from the CU to every unit.
• An n-bit address bus addresses 2^{n} locations; the address bus is unidirectional, the data bus bidirectional, the control bus carries commands.
• RAM is volatile (SRAM = cache, DRAM = main memory); ROM is non-volatile (MROM, PROM, EPROM, EEPROM, Flash). Cache speeds repeat access; a buffer matches unequal speeds.
• Secondary storage: magnetic disk (tracks, sectors, cylinders), flash/SSD, optical (CD 700 MB, DVD 4.7 GB, BD 25 GB), external and cloud.
• Serial interfaces (USB, HDMI, PCIe) have replaced the old parallel and RS-232 ports because long parallel links cannot be clocked fast.

Practice questions

Group A — Multiple choice (1 mark each)

1. Which of the following is not a characteristic of a computer? [1]aDiligencebVersatilitycIntelligencedAccuracy
2. The microprocessor that started the fourth generation of computers is [1]aIntel 8085bIntel 4004cIntel 8088dMotorola 6800
3. If the address bus of a CPU is 20 bits wide, the maximum number of memory locations it can address is [1]a20b2^{20}c20^{2}d2^{10}
4. Which memory is volatile? [1]aROMbEPROMcRAMdFlash memory
5. A device that reads characters printed in magnetic ink on a bank cheque is [1]aOMRbOCRcMICRdBCR
6. The capacity of a standard single-layer DVD is about [1]a700 MBb4.7 GBc25 GBd8.5 GB
7. Which port carries digital video and audio together on one cable? [1]aVGAbRS-232cHDMIdIEEE 1284

Answers to Group A

1. (c) — a computer has no IQ of its own; it only follows instructions.

2. (b) — the Intel 4004 (1971) was the first microprocessor.

3. (b) — an n-line address bus selects 2^{n} distinct addresses.

4. (c) — RAM loses its contents when power is removed.

5. (c) — MICR reads the E-13B magnetic ink characters on cheques.

6. (b) — DVD-5 single layer holds 4.7 GB; 8.5 GB is dual layer.

7. (c) — VGA is analogue video only; HDMI carries both digital video and audio.

Group B — Short answer (5 marks each)

1. Define a computer. Explain any four characteristics of a computer with one example of each. [5]
2. Differentiate between computer architecture and computer organization with two examples of each. [5]
3. What is a bus? Describe the three types of bus in a computer system, stating the direction of each. [5]
4. Differentiate between RAM and ROM. Name and describe any three types of ROM. [5]
5. A computer has a 32-bit address bus and a 64-bit data bus. Its clock runs at 4 GHz and one memory transfer takes 5 clock cycles. Find(a) the maximum addressable memory in GB if each location holds one byte,(b) the clock period, and(c) the peak transfer rate in GB s⁻¹. [5]
6. Distinguish between OMR, OCR and MICR, giving one application of each. [5]

Answers to Group B

1. A computer is an electronic device that accepts data as input, processes it according to a stored program, stores the data and results, and produces useful information as output. Four of its characteristics follow.

Speed it performs billions of operations per second, so a result file of lakhs of students can be graded in minutes.

Accuracy it repeats arithmetic without random error, and a wrong output means wrong input (GIGO); a bank adds interest to every account to the same precision.

Diligence it never tires or loses concentration, so a billing system can process every meter reading month after month.

Versatility the same hardware does word processing, accounting or games when the program is changed. Other characteristics are storage capacity, automation and reliability.

A computer has no intelligence or feelings of its own, so it can only do what the program tells it.

Together these properties explain why computers have replaced manual work in banks, offices, schools and hospitals, and they also show the limits of the machine: it is fast and accurate, but it is not able to think.

2. Computer architecture describes the attributes of a computer that a programmer can see and that decide what it can do, while computer organization describes how those attributes are physically built and connected.

Architecture answers what the machine does; organization answers how it does it.

Basis	Architecture	Organization
Concerned with	Logical, programmer-visible design	Physical implementation
Examples	Instruction set, number of registers, data types, addressing modes	Control signals, interfaces, memory technology, bus design
Question	What operations are available?	How is the operation carried out in hardware?
Changes	Rarely, because software depends on it	Often, with each new chip model

Example 1: whether a processor has a multiply instruction is an architectural decision; whether it is done by a dedicated multiplier circuit or by repeated addition is an organizational one.

Example 2 a 64-bit instruction set is architecture, but the number of cache levels used to implement it is organization.

Two processors from different makers can share one architecture (the same instruction set) while having very different organizations, which is why the same software runs on both.

Thus architecture is the design plan, and organization is the construction that follows it.

4. RAM (Random Access Memory) is the main working memory: it is read-write and volatile, so the contents are lost when the power is switched off. ROM (Read-Only Memory) is normally only read and is non-volatile, so it keeps its contents without power.

Basis	RAM	ROM
Operation	Read and write	Mainly read
Volatility	Volatile	Non-volatile
Use	Running programs and data	Start-up (BIOS) program, firmware
Speed and cost	Fast, costlier per bit	Slower to write
Types	SRAM, DRAM	PROM, EPROM, EEPROM

Three types of ROM: (i) PROM (Programmable ROM) is blank when made and can be programmed once by the user with a special device; after that it cannot be changed.

(ii) EPROM (Erasable PROM) can be erased by exposing its quartz window to strong ultraviolet light and then reprogrammed, but the whole chip is erased at once.

(iii) EEPROM (Electrically Erasable PROM) can be erased and rewritten electrically, byte by byte, without removing the chip; flash memory in pen drives and SSDs is a development of it.

A computer keeps its BIOS or start-up firmware in ROM-type memory so that it always knows how to boot, while RAM holds everything that changes while you work.

3. A bus is a set of parallel wires carrying bits between the units of a computer. The address bus is unidirectional (CPU → memory/I/O) and carries the address of the location being accessed; n lines give 2^{n} locations.

The data bus is bidirectional and carries the actual data or instruction; its width normally equals the word length.

The control bus carries individual timing and command signals (READ, WRITE, CLOCK, RESET, INTERRUPT) in both directions.

5.(a) Locations = 2^{32} = 4{,}294{,}967{,}296; at one byte each that is 2^{32}/2^{30} = 4\ \text{GB}.(b) T = 1/(4\times10^{9}) = 0.25\ \text{ns}.(c) One transfer takes 5 \times 0.25 = 1.25\ \text{ns} and moves 64 bits = 8 bytes, so the rate is 8/(1.25\times10^{-9}) = 6.4\times10^{9}\ \text{B s}^{-1} = 6.4\ \text{GB s}^{-1}.

6. OMR senses whether a mark has been made in a fixed position by measuring reflected light — used for MCQ answer sheets.

OCR recognises the shape of printed or handwritten characters and converts them to editable text — used to digitise books and forms. MICR reads characters printed in magnetic ink in the E-13B font — used for clearing bank cheques.

Group C — Long answer (8 marks each)

1. Draw the block diagram of a computer system and explain the function of each unit. Clearly distinguish the data path from the control path, and state the function of the control unit, the ALU and the registers. [8]
2. (a) What is a microprocessor? Explain clock speed and word length, and state how each affects performance. [4](b) A processor has a clock speed of 3.2 GHz, a 16-bit data bus and a 28-bit address bus. Calculate the clock period, the memory it can address in MB, and the number of instructions executed per second if each instruction takes 8 clock cycles. [4]
3. Explain the memory hierarchy of a computer. Compare cache, primary memory and secondary memory on speed, cost, capacity and volatility, and explain with an example why cache memory improves performance. [8]

Answer outline to Group C

1. Draw input unit → CPU → output unit with primary memory above the CPU and secondary storage below. Inside the CPU show CU, ALU and registers.

Solid arrows = data and instructions; dashed arrows = control signals radiating from the CU to input, memory, ALU and output.

Then describe: input converts data to binary; the CU fetches, decodes and issues control signals but performs no arithmetic; the ALU performs arithmetic and logic; registers hold operands and results; primary memory holds the running program; secondary storage keeps data permanently; output converts binary back to human-readable form.

2. (b) T = 1/(3.2\times10^{9}) = 0.3125\ \text{ns}. Addressable locations = 2^{28} = 268{,}435{,}456; at 16 bits = 2 bytes each, total = 2^{28}\times 2 = 2^{29}\ \text{B} = 512\ \text{MB}.

One instruction takes 8 \times 0.3125 = 2.5\ \text{ns}, so the rate is 1/(2.5\times10^{-9}) = 4\times10^{8} instructions per second = 400 MIPS.

3. Sketch the pyramid: registers, L1/L2/L3 cache, RAM, SSD, HDD, tape/cloud. Going down, capacity and cost per byte improve but access time worsens by roughly a factor of ten per level.

Cache is small, fast, volatile SRAM; primary memory is larger, slower, volatile DRAM; secondary memory is largest, slowest and non-volatile.

Cache works because of locality of reference — a loop executed a million times is fetched into cache once and then read at cache speed, so a 90 % hit rate can cut the average access time of a 1 ns cache and a 60 ns RAM to about 0.9(1) + 0.1(60) = 6.9\ \text{ns}.

Group D — Long answer (10 marks each), with full model answers

1. Describe the generations of computers. Compare them in a table. [10]
2. Explain computer architecture with a neat block diagram of a computer system. [10]
3. What is a computer? Explain its characteristics and any five applications, with examples from Nepal. [10]

Model answer D1: generations of computers (10 marks)

Introduction. A generation of computers is a stage in the history of computing that is defined by the main electronic technology used for switching.

Five generations are recognised, and each is smaller, faster, cheaper and more reliable than the one before. (Draw the timeline figure of section 1.2.)

Explanation. The first generation (1946-1958) used vacuum tubes; ENIAC and UNIVAC-I were huge, hot and unreliable, programmed in machine language with a magnetic drum for memory.

The second generation (1959-1964) used transistors, magnetic core memory, assembly language and early FORTRAN and COBOL; the IBM 1401 is an example, and a machine of this type processed Nepal's 2028 BS census.

The third generation (1965-1970) used integrated circuits, introduced operating systems and multiprogramming, and keyboards and monitors; the IBM 360 and PDP-8 are examples.

The fourth generation (1971-present) is based on the microprocessor, starting with the Intel 4004; it gave us the personal computer, GUI, networks and the Internet.

The fifth generation (present and future) uses ULSI chips, parallel processing, artificial intelligence and quantum technology to make machines that can learn and understand speech.

Generation	Device	Memory	Language	Example
First (1946-58)	Vacuum tube	Magnetic drum	Machine	ENIAC, UNIVAC-I
Second (1959-64)	Transistor	Magnetic core	Assembly, FORTRAN, COBOL	IBM 1401
Third (1965-70)	IC (SSI/MSI)	Core, early RAM	BASIC, Pascal, OS	IBM 360, PDP-8
Fourth (1971-now)	Microprocessor	Semiconductor RAM	3GL, 4GL, GUI	IBM PC
Fifth (emerging)	ULSI, AI	Very large RAM, optical	Natural language	AI systems

Merits and demerits. Each step brought lower size, power use and cost with higher speed and reliability. The early generations suffered from heat and failure, and the latest raise issues of privacy, e-waste and jobs.

Conclusion. The history of computers is the history of a shrinking switch: from a glass tube to a transistor to a chip with billions of transistors. This progress made computers affordable for ordinary students in Nepal.

Model answer D2: computer architecture and block diagram (10 marks)

Introduction. Computer architecture is the design of a computer as seen by the programmer: its instruction set, registers, memory and the way the units are connected.

The basic plan used almost universally is the stored-program (von Neumann) model, in which instructions and data share one memory.

(Draw the block diagram of section 1.7: input unit, CPU with CU, ALU and registers, primary memory, secondary storage, output unit; solid arrows for data and dashed arrows for control.)

Explanation of the units.

1. Input unit accepts data and instructions from the user (keyboard, mouse, scanner) and converts them to binary.
2. Control unit (CU) fetches each instruction, decodes it and sends control signals to the other units; it does no calculation.
3. Arithmetic and logic unit (ALU) performs addition, subtraction, comparison and logical operations.
4. Registers are very fast storage cells inside the CPU that hold the current instruction, operands and results.
5. Primary memory (RAM and ROM) holds the programs and data that are being used now; RAM is volatile.
6. Secondary storage (hard disk, SSD, pen drive) keeps programs and data permanently, but is slower.
7. Output unit converts binary results to a form people can use (monitor, printer, speaker).
8. Buses (address, data, control) are the wires that carry information between these units.

Working. The program is loaded into memory. The CU fetches an instruction, decodes it, the ALU executes it using registers, and the result goes to memory or to the output unit. This fetch-decode-execute cycle repeats millions of times a second.

Path	Carries	Direction
Data path (solid)	Data and instructions	Between input, memory, CPU, output
Control path (dashed)	Timing and command signals	From the CU to every unit

Conclusion. Because instructions and data live in one memory, a computer can be reprogrammed simply by loading a different program. This idea is why one machine can serve as a billing terminal in the morning and a game console at night.

Model answer D3: computer, characteristics and applications (10 marks)

Introduction. A computer is an electronic device that accepts data, processes it under the control of a stored program, stores the result and produces information as output. It works in the cycle Input, Process, Output, with Storage.

Characteristics.

1. Speed: billions of operations per second; an examination board grades lakhs of students in minutes.
2. Accuracy: no random errors; wrong output comes from wrong input (GIGO).
3. Diligence: no tiredness, so a billing system runs month after month.
4. Versatility: one machine runs many programs.
5. Storage capacity: terabytes in a small box, retrieved instantly.
6. Automation: runs a program to completion without help.
7. Reliability: consistent results over long periods.
8. No intelligence: it follows instructions literally.

Applications (five examples from Nepal).

1. Banking: ATMs, mobile banking, ConnectIPS transfers and Fonepay QR payment.
2. Education: e-learning, online result publication and student records.
3. Government: national ID, passport, driving licence and the Nagarik App.
4. Weather and science: the Department of Hydrology and Meteorology uses computer models for forecasts and flood warnings.
5. Health: hospital records, CT and MRI scanning, and telemedicine for remote districts.

Area	Computer does	Benefit
Banking	Processes transactions	Fast, 24-hour service
Education	Stores and publishes results	Less paperwork
Government	Keeps citizen records	Fewer queues
Weather	Runs forecast models	Earlier warnings
Health	Stores records, forms images	Faster diagnosis

Conclusion. Computers are fast, accurate, tireless and versatile, but they have no intelligence of their own. Used properly, they improve almost every service that a Nepali citizen depends on, from banking to weather warnings.