Denary number	Binary equivalent
0	0
1	1
2	10
3	11
4	100
5	101
6	110
7	111
8	1000
9	1001
10	1010

As I mentioned earlier, a computer functions by manipulating 1s and 0s. As you have seen, you can represent any denary number in binary. It is also possible to represent any letter of the alphabet, or other character, using binary by simply assigning a code to it in the computer. For instance, there is an agreed representation of text known as International Alphabet Number 5 (IA-5) in which the letter ‘A’ is represented by the binary pattern 1000001. When I type the letter A, this binary number will be stored in my computer. I can later retrieve it and the letter A will be displayed on screen. This will only happen if the computer has received instructions to treat 1000001 as an IA-5 character. The same pattern could be used to represent the denary number 65. The computer knows what to do with the data because it has instructions from a program, and these instructions are themselves binary representations.

It is worth examining the difference between data and instructions. The data is the current information the computer program is working with. This might be some numbers I am adding up, or some text I am typing. It will vary from instance to instance. The instructions are what the computer does with the data. This must always be consistent, for example clicking on the Save button will always save the data.

So numbers and text can be represented using the binary system. What else can? Images can be represented using a technique known as bit-mapping. This divides an image up into thousands of cells and allocates a value to each cell. If the image is in black and white, each cell will have a value of 1 (indicating it is black) or 0 (indicating it is white). Colour can be represented by allocating more information to each cell to indicate the proportion of red, green and blue (RGB) values. A wide spectrum of colours can be created by varying the relative values of red, green and blue.

What else can be represented in binary? The answer is just about anything. Sound, like images, can be divided up into different segments and each given an appropriate binary value, which can then faithfully reproduce the sound. This is what your music CD player does.

You will hear people talk about computers being ‘digital’. Sound, light and other natural signals are usually analogue. The difference between digital and analogue is an important one as it underlies the advantage in using computers for many tasks. You will learn more about what is meant by analogue and digital in the next section.

So computers work by manipulating 1s and 0s. These are binary digits, or bits for short. Single bits are too small to be much use, so they are grouped together into units of 8 bits. Each 8-bit unit is called a byte. A byte is the basic unit which is passed around the computer, often in groups. Because of this the number 8 and its multiples have become important in computing. You will particularly encounter the numbers 8, 16, 32 and 64 in various contexts in computing literature, and this is usually due to the 8-bit byte being the basic building unit. The key point to appreciate is that although basing your entire system on only two digits may seem limiting, these two digits can be used to represent almost anything.

You will also hear people speak of kilobytes, megabytes and gigabytes or often just ‘K’, ‘meg’ and ‘gig’ as in, ‘This computer has 64 megs of RAM’, or ‘This file is 45 K’. Bits, bytes, kilobytes and megabytes are merely ways of measuring the size of things computers deal with. A kilobyte is 2 to the power of 10 bytes. This is actually 1024 bytes, but is close enough to a thousand to be given the prefix kilo, meaning a thousand. Similarly, a megabyte is 2 to the power of 20 (or 1 kilobyte squared), which comes out as 1,048,576 bytes. For the sake of convenience, this is called a megabyte, meaning a million bytes. A gigabyte is 1000 megabytes.

Digital devices are often faster and more efficient than analogue equivalents. So why is it that by limiting a system to just two values (1 and 0), it is possible to get an improved performance? On the face of it, it would seem that an analogue system, which could take any value, would have greater flexibility and power. There are two main reasons why digital systems are often superior: reliability and manipulation.

Reliability

There is an immediate problem in trying to reproduce an analogue signal. Since it can take any value, how do you know that the value you have reproduced is correct? This problem does not exist with digital signals. Because a digital value can only be one of a set range, it can be reproduced exactly.

This is why the telephone network has largely been switched over to a digital system. An analogue signal declines in strength when it is sent over a wire, so it has to be boosted (or amplified) at regular intervals. However, when an analogue signal is amplified, any incorrect or corrupted values in the signal (what is called noise) is also amplified because it is not possible to distinguish noise from the original signal. A digital signal is not amplified, it is regenerated, so the resulting signal is a faithful copy of the original. This is possible because of the use of discrete values.

Consequently digital signals can be sent over any distance you like (provided they are regenerated often enough), and the signal received at the end will be as good as the original.

Manipulation

Another reason why digital systems have become popular is related to the increase in speed and power of microprocessors, and their falling prices. Although microprocessors only deal with 1s and 0s, they do so so quickly and reliably that using a microprocessor to handle the light input into a camera, say, is now a viable proposition.

This means that things can be manipulated easily. Consider a digital image. It is represented in a computer as a series of 1s and 0s, which represent the colour, luminescence and hue of each of the image's picture elements (pixels). If I have a picture of the British Prime Minister shaking hands with the American President in a digital format, it is not a difficult job to alter it. All I am doing is altering the 1s and 0s underneath. So, using suitable software, it would not be difficult to replace the head of the British Prime Minister with my head (taken from a different digital image), and blend this in so that it would be very difficult to tell that it was a 'fake'. This would be very difficult to do with an analogue representation, such as an actual photograph. This ease of manipulation is very important and powerful, and is one of the reasons why working in a digital format is so attractive.

The key to your computer is a chip called the microprocessor. This is its brains, and is where most of the computing takes place.

Before the advent of the microprocessor, computers came mainly in the form of large mainframes which had a different circuit board for each function. At the core of a mainframe computer are three separate units linked together to form what is known as the central processing unit, or CPU. These three units are:

• The arithmetic and logic unit (ALU) - this is the unit which does the actual work of the computer. As well as the four basic mathematical functions of addition, subtraction, multiplication and division, it has comparison capabilities such as =, >, < (equals, more than, less than).

• The control unit - this unit controls the flow of data from the computer's memory into the ALU and to other devices.

• The memory.

A microprocessor combines the ALU and control unit on one silicon chip, which is why it was at one time referred to as the ‘computer on a chip’. In mainframes the CPU includes memory, but this is separate in microcomputers, so I shall use the term CPU to refer to just the combination of ALU and control unit. I shall describe microprocessors in more detail later, but you should appreciate for now that they can perform a variety of functions. Inside your computer, in addition to the microprocessor which forms the CPU there are other microprocessors that are used to control the graphics card, modem and other devices.

The CPU microprocessor is housed on a circuit board called the motherboard. Also on the motherboard is the clock chip which acts as a metronome for the computer so that all its actions can be synchronised. There may also be one or two ROM chips. ROM stands for Read Only Memory, which means that the data on these chips cannot be altered, it can only be read. These chips often contain some important programs which come supplied with the computer and which are needed for it to function properly. This is why they are made to be read-only; it would be very unfortunate if an unsuspecting user altered them.

As well as the CPU microprocessor there are devices which can be used to enter data into the computer, and which it can use to output data. These are called input/output devices (usually referred to as I/O devices) and might include a keyboard and mouse (for input) and a monitor and printer (for output). The data for these devices will often pass via a slot-in circuit board (called a card) inside the PC which plugs into a slot on the motherboard. These cards perform a number of functions, such as converting data to a form usable by that particular brand of device.

The CPU also sends and receives data to and from the computer's memory, which is usually referred to as RAM (random access memory). The RAM consists of chips similar in construction to the CPU. This is the memory which stores all the data the computer is currently using. You can think of the RAM as the computer's desk. When it needs to work on something it will retrieve it, for example a file from the hard disk, and put it into the RAM - just as you might get a file from a drawer and place it on your desktop. The computer will also load into the RAM the programs it needs to work with this file, in the same way you might place your calculator on your desktop. Like a desktop there is a limit to how much can be placed in the RAM. The contents of RAM are often lost when the computer is switched off.

The microprocessor will also have to read from and write to a permanent data storage device in the form of a hard disk drive. Other permanent storage devices include CD-ROMs (again the ROM in the name indicates that these CDs cannot be written to), optical disks, tape storage or floppy disks (which may be read only or writable).

To receive and send all of this data the microprocessor is connected to cables, which are referred to as buses. This term comes from the Latin omnibus, meaning ‘for all’. Buses are designed to carry all kinds of data, rather like the buses you might use to travel around a city (their name is also derived from omnibus). Thinking of your computer as a street layout, with traffic and buses transporting data to various destinations, is not a bad way of visualising what is actually happening.

This is a simple overview of a computer's architecture, but it is sufficient for this course. The architecture is summarised in the figure below.

The second factor that determines a computer's power is its clock speed. Each action of the CPU can be thought of as occupying one cycle - although most modern CPUs can perform several tasks at the same time. Therefore the greater the number of cycles per second, the faster the computer.

The third factor is the size of the words that both the microprocessor and the buses can accommodate (go back to the previous segment on computer architecture if you need to refresh your memory about these). You may sometimes hear people refer to a computer - or particularly a games console - as a 32-bit, 64-bit or 128-bit machine. These terms refer to the size of word that the microprocessor can manipulate. The larger the word size, the more information each word can contain. A 32-bit word can contain twice the data of a 16-bit word. Therefore increasing the word size improves both the complexity (more data can be manipulated) and the speed (because it takes the same time to interpret each word).

A fourth factor that affects a computer's power is the amount of memory, or RAM, available. RAM acts as the computer's working memory, so it contains the information the computer needs for its current session. This will include various operating system commands (e.g. what a mouse double click means), the programs currently running (e.g. Microsoft Word with two documents open), current display configurations (e.g. that one Word document is in front of the other and the mouse pointer is in the top left of the screen) and so forth. An increase in the size of the RAM increases the amount of data it can store at any one time. This improves the complexity of the computer because it can run programs that require a lot of data to be handled, and several programs can be run simultaneously. It also increases the speed of the computer because when the RAM becomes full the computer will temporarily store data on the hard disk. This will be retrieved when it is required, which takes more time than having it currently available in memory. So more RAM provides more speed.

The power of computers now is such that their performance is fast enough to meet most needs, and the software you use does not really necessitate the rapid upgrade it once did. This gradual easing of demand poses a problem for the computer industry.

Mainframes were the dominant form of computing before microcomputers. They are usually very expensive, powerful and operate specialist software.

Mainframes are typically used by large companies, public authorities and universities for their data handling tasks. These tasks are typically:

• File maintenance: This is perhaps the most common use of mainframes. Maintaining records is a huge task for institutions. Records can contain information on sales, credit card status, payroll details, social security details, health records, stock inventory, etc. These either need to be accessed by different people in real-time (for instance a travel agent booking an airline ticket) or updated in batches (for instance warehouse stock levels at the end of each day). It is necessary in such cases to have the data stored centrally and then accessible by those who need it. A lot of minicomputers are now capable of performing these tasks in medium-sized companies.

• Simulations: Many physical and engineering problems cannot be solved without the help of complex computer simulations. These require intensive mathematical work, and so take advantage of a mainframe's computational power. Examples include weather forecasting, or calculating the position of astronomical bodies with extreme accuracy. Many minicomputers or workstations are now used for this type of problem.

• General purpose: Many universities used a mainframe to act as a general purpose computing facility. Each user can then be given their own area on the mainframe to store files, and different departments can use its resources to perform different tasks, e.g. predicting bird populations in the Biology department and calculating metal stress in the Engineering Department. PCs are now used to perform many of these tasks.

You will have noticed that after each task I mentioned that other types of computer could now be found performing these tasks. This indicates that the general growth in the mainframe area has slowed and is even in decline, as smaller computers have become more powerful. Mainframes are still required by many institutions, however, to perform large data handling tasks. They are particularly useful when data needs to be held centrally, with different people needing access to it. This is illustrated below:

Minicomputers are powerful, special-purpose computers. They were originally viewed as small mainframes - hence the prefix ‘mini’. However, they have become increasingly powerful and have replaced mainframes for many functions. Examples include Digital Equipment's VAX machines and IBM's AS/400s.

Tasks minicomputers might be used for include:

• Plant control: Many industrial plants require a central computing facility to collect data from various sensors and then to act accordingly. For example, in a chemical engineering plant, as the pressure in one vat increases the computer registers this, and opens a release valve slightly while also adjusting the boiler temperature.

• Network control: Many computer networks need a central computer which provides storage space and controls the network using special network software. This is known as a server. The other computers which access the server are called clients. Such machines can also act as the interface to the Internet, accepting Internet messages and hosting e-mail and World Wide Web facilities. Powerful PCs can also be used to perform these functions.

• Databases: As mentioned above, the role of mainframes in file maintenance is increasingly being taken by minicomputers. Minicomputers can hold databases of records which appropriate people can access.

For the general public it is PCs which tend to symbolise computers. However, most ‘heavy duty’ computing is performed not by PCs but by minicomputers. With the growth in networking computers in most institutions the role of minicomputers has grown. It is in this market that some of the largest software companies, such as Novell who provide networking software and Oracle who supply database software, do most of their business.

The network model is shown below.

Workstations are based on specialised microprocessors and can be thought of as powerful PCs. They are typically used for specialist engineering tasks. Workstations use a special type of microprocessor known as a RISC (reduced instruction set computer) chip. This technology, developed at IBM, removes many of the complex instructions from a microprocessor and has instead a set of basic instructions, which perform their tasks very quickly. This approach increases the speed and the power of the microprocessor, particularly when dealing with numerical problems. The workstation market is dominated by SUN Microsystems. They typically cost in the £10,000s, and can cost £100,000s.

Tasks workstations might be used for include:

• CAD/CAM: Computer Aided Design and Computer Aided Manufacturing have been growth industries since the mid-1980s. These technologies allow engineers to design complex machine parts without having to produce an actual physical model. The necessary complex, 3-D graphics require a lot of computing power and good quality video capabilities.

• Animation: Similarly to CAD/CAM, animations require a lot of processing power which must be performed quickly, and are ideally suited to the RISC technology.

• Simulations: Performing simulations of processes, such as the behaviour of an industrial plant, requires considerable computing power.

• Multitask programming: Any complex programming which is deemed to require more power than is offered by a PC, and which needs to perform more than one task at a time (called multitasking), is often performed on a workstation. Recent PC operating systems such as Windows NT can also perform multitasking.

The workstation will typically be part of a network.

Microcomputers are based on a microprocessor and are intended for individual use; hence they are called personal computers, or PCs. They were initially standalone machines, but are increasingly connected to a network. They are ideal for tasks such as those listed below where the user requires individual computing power.

• Word processing: Word processing programs allow a user to produce professional-looking documents, with different fonts, styles and pictures. The production of complex and attractive documents is much easier with the use of word processing software than it was in the days of the typewriter.

• Spreadsheets: Spreadsheets allow a user to create mathematical models. These are particularly useful in financial planning and can be used to answer questions such as ‘What happens if we raise the cost of our product by 1%?’.

• Desktop publishing: This allows people to create drawings, manipulate images and combine them with text to produce professional graphic design work, for instance magazine layouts, posters, book covers, etc.

• Games: Although this may seem a non-serious use of computers, games represent a large market. By making use of the PC's computing power very good quality graphics and complex game play can be achieved.

• Servers: As mentioned above, many powerful PCs are now being used as servers to control a network.

Generally the tasks performed by the mainframes have been taken over by minicomputers. PCs initially created a new type of usage for the computer; for instance, spreadsheets and desktop publishing took over from activities previously performed by hand. As PCs have increased in power they have begun to be used for some of the tasks that required minicomputers and even mainframes in the past.

With the increase in computer power the distinction between types of computers is often difficult to maintain. When does a PC become a workstation for instance? Remember, though, that having an appreciation of the different types of computers and the tasks required of them is important, but you should not get too bogged down in trying to categorise machines.

In this Unit you have covered:

• What software and hardware are.

• The different categories of software.

• What binary is, and how it relates to denary.

• How a computer uses bits to represent information.

• What analogue and digital mean.

• The main components of a computer and their functions.

• The factors that affect the power of a computer.

• The difference between mainframes, minis, workstations and PCs.

• Computer terminology, such as RAM, ROM, CPU, bytes, kilobytes, data, instructions, I/O, motherboard, buses and words.