THE DEVELOPMENT OF COMPUTERS

Modern computers come in an enormous variety of sizes and shapes, ranging from the smallest personal computers to huge machines filling warehouse-sized rooms. Nearly one hundred fifty years ago there were no such things as computers — at least in the sense we are using the term now. There have been calculating aids for millennia. Knotted ropes, marks in clay, the abacus, and the soroban are all methods of keeping track of numbers. But the stored-program computer really did not come into existence until the 1830.

A score of years after the war of 1812, an English inventor and mathematician Charles Babbage was commissioned by the British government to develop a system for calculating the rise and fall of the tides.

Dozens, even hundreds of clerks busily calculating away throughout their lifetimes could not get their job done, let alone do it without errors. Babbage decided to build a device he called an analytical engine.

He designed the first programmable computer, complete with punched cards for data input. Incidentally, the punched card was not invented for use with the computer but was used as early as the 1700s by Bouchon and in the 1800s by Jacquard to control automatic looms (станок). Babbage adapted the idea for his computer, and it has been with us ever since.

Babbage gave the engine the ability to perform different types of mathematical operations. The machine was not confined to simple addition, subtraction, multiplication, or division; it had its own " memory" and, because of this " stored program", the machine could use different combinations and sequences of these to suit the purposes of the operator. It became an autonomous machine, able to perform on its own, once commanded to do so as were the automated looms and the common clock.

The machine of his dreams was never realized in his lifetime.

Yet Babbage's idea didn't die with him. Others made attempts to build mechanical, general-purpose, stored-program computers throughout the next century. In the process it became clear that mechanical methods of general-purpose computing on all but the most modest scale were simply not practical.

In 1941 a relay computer was built in Germany by Conrad Zuse. It was a major step toward the realization of Babbage's dream. The logical operations of the computer were alterable by changing the interconnections among the relays. At the same time, in the United States, International Business Machines (IBM) built a machine in cooperation with scientists working at Harward University under the direction of Prof. Aiken during the years from 1939 to 1944. The computer, called the Mark I Sequence-Controlled Calculator, was built to perform calculations for the Manhattan Project, which led toward the development of the atomic bomb.

The relay computer had its problems. Since relays are electromechanical devices, the switching contacts operate by means of electromagnets and springs. They are still fairly slow and very noisy. They also consume a lot of power, if their contacts become dirty or corroded, they are unreliable.

The gadget (приспособление) that was the basis for the first computer revolution was the vacuum tube, an electronic device invented early in the twentieth century. The vacuum tube was ideal for use in computers. It had no moving parts, or at least no mechanical moving parts. It switched flows of electrons off and on at rates far faster than possible with any mechanical device. It was relatively reliable, lasting hundreds of hours before failure. Previously, computer designers could think only in terms of hundreds of calculations in a program to be run on a mechanical computer. Now they could easily conceive of programs with thousands of related computations using a vacuum-tube computer. The first vacuum-tube computer was built at Iowa State University at about the same time as the Mark I. It was the beginning of the revolution. It was called ABC (Atanasoff-Berry Computer). From the ABC a number of vacuum-tube digital computers evolved.

A splendid example of these first generation electronic computers is ENIAC (an acronym for Electronic Numerical Integrator and Calculator). ENIAC was over 90 tons and bulging into 3000 cubic feet and costing millions. Its 18 thousand vacuum tubes demanded 140 kilowatts of electrical power, enough to supply a block of buildings of respectable size. With its 16, 000 bytes of random access memory and its 100-kilohertz clock, it was not quite up to the basic computer capability of modern computers. Since its programs were hardwired – that is, the programs operating the computer were established by physically changing the patterns of the wires interconnecting the vacuum tubes – it was not so flexible in its operation.

From the university laboratories the computer finally entered the wider world in 1951 with the delivery of the first UNIVAC I (Universal Automatic Computer).

In 1948 the next key element in spreading the practical - and impractical — applications of computers, the transistor, came into existence. The potential advantage of the transistor over the vacuum tube was almost as great as that of the vacuum tube over the relay. A transistor can switch flows of electricity as fast as the vacuum tubes used in computers, but the transistors use much less power than equivalent vacuum tubes, and are considerably smaller. With the transistor came the possibility of building com­puters with much greater complexity and speed than was considered even remotely possible just 10 years before.

The integrated circuit constituted another major step in the growth of computer technology. Until 1959 the fundamental logical components of digital computers were the individual electrical switches, first in the form of relays, then vacuum tubes, then transistors. In the vacuum tubes and relay stages, additional discrete components such as resistors, inductors, and capacitors were required in order to make the whole system work. These components were generally each about the same size as packaged transistors. Integrated circuit technology permitted the elimination of some of these components and " integration" of most of the others on the same chip of semiconductor that contains the transistor. Thus the basic logic element - the switch, or " flip-flop", which re­quired two separate transistors and some resistors and capacitors in the early 1950s, could be packaged into a single small unit in I960. The chip was a crucial development in the accelerating pace of computer technology.