The basic architecture of today’s electronic computers originated in a design proposed in the 1830s by Charles Babbage (1792–1871), a British professor of mathematics. The ideas that he wanted to embody in a huge steam-powered mechanical computing device were finally realized only in the mid-twentieth century, using relays and electronic circuits. Since then, computers have evolved from room-sized mainframes to desk-sized mini-computers to today’s laptop and palm-sized  microcomputers and the miniature digital signal processors found in many appliances.

From the standpoint of the computer user, the most important aspects of the computer are still those recognized by Babbage: input, processing,  storage, and output. Anyone who wants to use a computer for information processing is primarily concerned only with getting the information in, getting it processed, storing intermediate values, and using the results. Achieving these results depends on the interworkings of transistors, electric currents, coding systems, binary numbers, computer programs, programming languages, multiprocessing, and operating systems.

Input, Processing, Storage, and Output

All electronic computers use the same fundamental architecture that Babbage recognized nearly 200 years ago. One or more input devices, such as a keyboard or mouse, handle entry of information (data) into the machine.

A processor determines or computes new values from entered values and from other previously computed values. Storage elements retain entered and computed data values. Finally, output devices send entered and computed values to a printed page, an electronic display, a telephone line, a loud speaker, or wherever they can be used by people or other machines.

In today’s personal computers, the input devices may include a keyboard, mouse, joystick, trackball, light pen, touch screen, infrared detector,  telephone line, microphone, scanner, magnetic diskette, and CD-ROM disk. Input can take several forms—audio, video, text, numbers, and graphic images. New input devices are also continually being developed.

Processing reflects the computer’s ability to combine and manipulate data to convert it into useful information. The main processor is an  integrated circuit (IC) chip or set of chips, often called a CPU (Central Processing Unit).

Other IC chips do some processing as part of the input and output devices.

 Storage, either on a short-term or long-term basis, is provided by IC chips that only store data and by magnetic disks (hard or fixed disks and removable diskettes), magnetic tapes, and writeable compact disks. Output, or the results of data manipulation, is provided by means of data terminal screens, printers, telephone lines, infrared ports, and loudspeakers.

Hardware

Input and output (I/O) devices described earlier are all part of the hardware of the computer—that is, they are realized in pieces made of plastic, metal,  glass, and special materials (such as silicon and germanium) that can be seen and touched. I/O devices contain a wide variety of parts, large and small, many of which could be seen in a disassembled printer, mouse, or other I/O device.

The basic electronic element in integrated circuit (IC) chips—whether used for processing or for storage—is the transistor. One IC chip may contain thousands or millions of transistors. The great power of the modern computer comes from being able to put enormous numbers of very tiny  (miniaturized) transistors into a single plastic case, often no bigger than a pencil eraser.

The basic function of a transistor is to control the strength of a small electrical current using a small electrical voltage. Transistors in a radio or audio player control the production of sound from radio waves, magnetic tapes, or compact disks. Transistors in a computer control electrical currents that represent information in the form of numbers and letters.

A radio or an audio player may produce sound imperfectly if there are electrical disturbances nearby, such as fluorescent lights or electric motors.

These imperfections must be avoided in a computer if it is to produce correct results. Therefore, transistors in an integrated circuit chip are designed so that they control electrical currents by turning them on or off.

The strength of an electronic current varies over time, as shown by the dashed line in the figure below. A transistor in a computer can examine the current strength at fixed time intervals and decide at each time whether the current is on or off, as shown by the solid line in the figure. A current above a certain strength is considered to be on; anything below that level is treated the same as no current, and is considered to be off.

A signal (current) that has only two states—on or off—is called a binary signal. All modern computers use binary signals so that they will be insensitive to electrical disturbances. A binary signal representing a series of 0s and 1s is also called a digital signal. Electronic computers that use such signals are known as digital computers.

Binary Coding

In a computer, a current that is considered to be either on or off at any time can represent a series of binary digits: 0s and 1s. When the current is off,  the value is 0, and when it is on, the value is 1.

The two-digit binary numbering system is the basis of machine representation of numbers, letters, characters, and symbols. The smallest unit of information is the binary digit, or bit. Eight bits (known as a byte) is considered a complete unit for storage or computation.

By considering the binary digits in fixed-length groups, the electrical current can be interpreted to mean a series of decimal digits or letters. The coding scheme used in most small computers today is called the ASCII character set. By assigning a unique eight-digit binary number sequence (a byte)  to each member in an alphabet of letters, decimal digits, and symbols, the ASCII set can represent each character on a keyboard by 0s and 1s.

Storage

Data values and commands being processed in a computer are represented as electrical currents that vary with time and are interpreted as a series of binary digits (0s and 1s). Data values and commands that are stored in integrated circuits are represented by currents and are “remembered” only as long as the circuits are powered.

In contrast, data values and commands that are stored for long periods must survive when power is off. They are also represented as a series of binary digits, but those digits are not realized as electrical currents varying over time. In a disk, diskette, or tape storage unit, binary digits are represented as magnetic spots on metallic coatings. Each spot is magnetized with its north pole in one direction or in the opposite direction.

Special electromagnets write and read the spots under the control of the processor.

Processors and Calculation

A processor is an integrated circuit chip that combines transistors in such a way that it can do arithmetic, make comparisons, make decisions, and control the input, output, and storage elements of the computer. These operations are performed using data values that are entered into the computer,  stored and processed in it, and sent as output from it.

Because the computer handles information in the form of letters and numbers (ultimately represented as binary numbers realized as electrical currents), it is natural to control its operation with a string of letters and numbers. For example, a certain series of letters and numbers fed to the processor on a particular circuit may cause it to fetch an item of data from a specified place in the storage unit. Such a series of letters and numbers is called a command.

Another command (a different series of letters and numbers) may cause two data values to be added together. Other commands can control movement of data values to and from storage, from input and to output, and can compare two data values and determine what command should be performed next.

Software and Programs

A series of commands stored in the computer’s storage unit is called a program. The same series of letters and numbers may be interpreted as commands if used in one way and interpreted as data values when used in another way. Because a computer can (1) store commands in the same way as data values and (2) modify its own commands based on the results of processing data, a computer can be called a “stored program machine.”

These two abilities give electronic computers great power.

Computer programs consisting of a series of commands are called software. Software is as essential as hardware in accomplishing the work that computers do. Software is realized as letters and numbers written or printed on a piece of paper or stored in magnetic form on disks and diskettes.

Multiprocessing and Interrupts

Several different command sequences, or programs, may be stored in a computer at one time. The processor may do all of the processing specified by one program and then do the processing specified by another. Under this arrangement, if the first program calls for some input or output task that takes considerable time (such as reading a tape), the processor might remain idle for quite a while. This time could be used to do some of the tasks specified by another program.

Most computers allow for one program to be interrupted and another one to be started or resumed under specified circumstances. This capability is called multiprocessing. Multiprocessing allows one to start an operation, such as sending a message via e-mail, and then switch to another operation, such as word-processing, while the first operation is being completed.

Many modern computers contain numerous processors and can store huge quantities of data. Different processors may be doing different things at the same time, under the control of different programs, each referencing input, storage, and output units in turn or as needed.

Machine Language, Programming Languages, and Applications

A series of commands consisting of letters and numbers (stored as binary digits, or bits) that the hardware processing unit can understand is said to be in machine language. Most processors understand very detailed commands, each of which does a small task. These can be combined in many varied ways to provide flexibility and hence great power.

For example, one command may call for fetching the data value stored at location A in the storage unit. The next command can ask data value stored at location B to be fetched. A third can cause the two values to be added together. A fourth would send the result to a display unit. The person creating the program probably thinks of this series of commands in terms of a single function, “display AB,” and is not interested in the individual commands that must be performed to cause this to happen.

Writing computer programs to do useful work has been greatly simplified by creating other programs that interpret instructions written in more user-friendly terms than machine language. Such programs are called programming language processors. Popular programming languages are BASIC, C, ADA, and PASCAL.

Most people who use computers do not want to learn any programming language. Instead, they want to communicate with the computer in terms of the job they wish to do. Word processing is a typical example: A person does not want to write computer instructions just to compose a simple letter.

Programs that do particular types of data processing are called applications or application programs. Application programs have been written to do every kind of information processing imaginable: from playing solitaire, preparing tax returns, processing payrolls, and sending e-mail messages, to designing computer circuit boards, tracking airplanes, and controlling rockets. 

______________________________________________________________________________________________

Reference

Davis, Gordon. Introduction to Computers, 3rd ed. New York: McGraw-Hill, Inc., 1977.

Parker, Charles. Understanding Computers Today & Tomorrow. Fort Worth, TX: Dryden Press, 1998.