Each generation of computer is characterized by amajor technological development that fundamentallychanged the way computers operate, resulting inincreasingly smaller, cheaper, more powerful and moreefficient and reliable devices. Read about eachgeneration and the developments that led to thecurrent devices that we use today.
First Generation - 1940-1956: Vacuum Tubes
The first computers used vacuum tubes for circuitryand magnetic drums for memory, and were oftenenormous, taking up entire rooms. A magnetic drum,also referred to as drum, is a metal cylinder coatedwith magnetic iron-oxide material on which data andprograms can be stored. Magnetic drums were once usedas a primary storage device but have since beenimplemented as auxiliary storage devices. The tracks on a magnetic drum are assigned to channelslocated around the circumference of the drum, formingadjacent circular bands that wind around the drum. Asingle drum can have up to 200 tracks. As the drumrotates at a speed of up to 3,000 rpm, the device'sread/write heads deposit magnetized spots on the drumduring the write operation and sense these spotsduring a read operation. This action is similar tothat of a magnetic tape or disk drive.They were very expensive to operate and in additionto using a great deal of electricity, generated a lotof heat, which was often the cause of malfunctions.First generation computers relied on machine languageto perform operations, and they could only solve oneproblem at a time. Machine languages are the onlylanguages understood by computers. While easilyunderstood by computers, machine languages are almostimpossible for humans to use because they consistentirely of numbers. Programmers, therefore, useeither a high-level programming language or anassembly language. An assembly language contains thesame instructions as a machine language, but theinstructions and variables have names instead of beingjust numbers.
Programs written in high-level languages aretranslated into assembly language or machine languageby a compiler. Assembly language programs aretranslated into machine language by a program calledan assembler.
Every CPU has its own unique machine language.Programs must be rewritten or recompiled, therefore,to run on different types of computers. Input wasbased on punched cards and paper tape, and output wasdisplayed on printouts.
The UNIVAC and ENIAC computers are examples offirst-generation computing devices. The UNIVAC was thefirst commercial computer delivered to a businessclient, the U.S. Census Bureau in 1951.
Acronym for Electronic Numerical Integrator AndComputer, the world's first operational electronicdigital computer, developed by Army Ordnance tocompute World War II ballistic firing tables. TheENIAC, weighing 30 tons, using 200 kilowatts ofelectric power and consisting of 18,000 vacuum tubes,1,500 relays, and hundreds of thousands of resistors,capacitors, and inductors, was completed in 1945. Inaddition to ballistics, the ENIAC's field ofapplication included weather prediction, atomic-energycalculations, cosmic-ray studies, thermal ignition,random-number studies, wind-tunnel design, and otherscientific uses. The ENIAC soon became obsolete as theneed arose for faster computing speeds.
Second Generation - 1956-1963: Transistors
Transistors replaced vacuum tubes and ushered in thesecond generation of computers. Transistor is a devicecomposed of semiconductor material that amplifies asignal or opens or closes a circuit. Invented in 1947at Bell Labs, transistors have become the keyingredient of all digital circuits, includingcomputers. Today's microprocessors contains tens ofmillions of microscopic transistors. Prior to the invention of transistors, digitalcircuits were composed of vacuum tubes, which had manydisadvantages. They were much larger, required moreenergy, dissipated more heat, and were more prone tofailures. It's safe to say that without the inventionof transistors, computing as we know it today wouldnot be possible.The transistor was invented in 1947 but did not seewidespread use in computers until the late 50s. Thetransistor was far superior to the vacuum tube,allowing computers to become smaller, faster, cheaper,more energy-efficient and more reliable than theirfirst-generation predecessors. Though the transistorstill generated a great deal of heat that subjectedthe computer to damage, it was a vast improvement overthe vacuum tube. Second-generation computers stillrelied on punched cards for input and printouts foroutput.
Second-generation computers moved from cryptic binarymachine language to symbolic, or assembly, languages,which allowed programmers to specify instructions inwords. High-level programming languages were alsobeing developed at this time, such as early versionsof COBOL and FORTRAN. These were also the firstcomputers that stored their instructions in theirmemory, which moved from a magnetic drum to magneticcore technology.
The first computers of this generation were developedfor the atomic energy industry.
Third Generation - 1964-1971: Integrated Circuits
The development of the integrated circuit was thehallmark of the third generation of computers.Transistors were miniaturized and placed on siliconchips, called semiconductors, which drasticallyincreased the speed and efficiency of computers. A nonmetallic chemical element in the carbon family ofelements. Silicon - atomic symbol "Si" - is the secondmost abundant element in the earth's crust, surpassedonly by oxygen. Silicon does not occur uncombined innature. Sand and almost all rocks contain siliconcombined with oxygen, forming silica. When siliconcombines with other elements, such as iron, aluminumor potassium, a silicate is formed. Compounds ofsilicon also occur in the atmosphere, natural waters,many plants and in the bodies of some animals.Silicon is the basic material used to make computerchips, transistors, silicon diodes and otherelectronic circuits and switching devices because itsatomic structure makes the element an idealsemiconductor. Silicon is commonly doped, or mixed,with other elements, such as boron, phosphorous andarsenic, to alter its conductive properties.
A chip is a small piece of semiconducting material(usually silicon) on which an integrated circuit isembedded. A typical chip is less than ¼-square inchesand can contain millions of electronic components(transistors). Computers consist of many chips placedon electronic boards called printed circuit boards. There are different types of chips. For example, CPUchips (also called microprocessors) contain an entireprocessing unit, whereas memory chips contain blankmemory.
Semiconductor is a material that is neither a goodconductor of electricity (like copper) nor a goodinsulator (like rubber). The most common semiconductormaterials are silicon and germanium. These materialsare then doped to create an excess or lack ofelectrons.
Computer chips, both for CPU and memory, are composedof semiconductor materials. Semiconductors make itpossible to miniaturize electronic components, such astransistors. Not only does miniaturization mean thatthe components take up less space, it also means thatthey are faster and require less energy.
Instead of punched cards and printouts, usersinteracted with third generation computers throughkeyboards and monitors and interfaced with anoperating system, which allowed the device to run manydifferent applications at one time with a centralprogram that monitored the memory. Computers for thefirst time became accessible to a mass audiencebecause they were smaller and cheaper than theirpredecessors.
Fourth Generation - 1971-Present: Microprocessors
The microprocessor brought the fourth generation ofcomputers, as thousands of integrated circuits werebuilt onto a single silicon chip. A silicon chip thatcontains a CPU. In the world of personal computers,the terms microprocessor and CPU are usedinterchangeably. At the heart of all personalcomputers and most workstations sits a microprocessor.Microprocessors also control the logic of almost alldigital devices, from clock radios to fuel-injectionsystems for automobiles. Three basic characteristics differentiatemicroprocessors:- Instruction Set: The set of instructions that themicroprocessor can execute.
- Bandwidth: The number of bits processed in a singleinstruction.
- Clock Speed: Given in megahertz (MHz), the clockspeed determines how many instructions per second theprocessor can execute.
What in the first generation filled an entire roomcould now fit in the palm of the hand. The Intel 4004chip, developed in 1971, located all the components ofthe computer - from the central processing unit andmemory to input/output controls - on a single chip.
Abbreviation of central processing unit, andpronounced as separate letters. The CPU is the brainsof the computer. Sometimes referred to simply as theprocessor or central processor, the CPU is where mostcalculations take place. In terms of computing power,the CPU is the most important element of a computersystem.
On large machines, CPUs require one or more printedcircuit boards. On personal computers and smallworkstations, the CPU is housed in a single chipcalled a microprocessor.
Two typical components of a CPU are:
- The arithmetic logic unit (ALU), which performsarithmetic and logical operations.
- The control unit, which extracts instructions frommemory and decodes and executes them, calling on theALU when necessary.
As these small computers became more powerful, theycould be linked together to form networks, whicheventually led to the development of the Internet.Fourth generation computers also saw the developmentof GUIs, the mouse and handheld devices
Fifth Generation - Present and Beyond: ArtificialIntelligence
Fifth generation computing devices, based onartificial intelligence, are still in development,though there are some applications, such as voicerecognition, that are being used today. Artificial Intelligence is the branch of computerscience concerned with making computers behave likehumans. The term was coined in 1956 by John McCarthyat the Massachusetts Institute of Technology.Artificial intelligence includes:- Games Playing: programming computers to play gamessuch as chess and checkers
- Expert Systems: programming computers to makedecisions in real-life situations (for example, someexpert systems help doctors diagnose diseases based onsymptoms)
- Natural Language: programming computers to understandnatural human languages
- Neural Networks: Systems that simulate intelligenceby attempting to reproduce the types of physicalconnections that occur in animal brains
- Robotics: programming computers to see and hear andreact to other sensory stimuli
In the area of robotics, computers are now widely usedin assembly plants, but they are capable only of verylimited tasks. Robots have great difficultyidentifying objects based on appearance or feel, andthey still move and handle objects clumsily.
Natural-language processing offers the greatestpotential rewards because it would allow people tointeract with computers without needing anyspecialized knowledge. You could simply walk up to acomputer and talk to it. Unfortunately, programmingcomputers to understand natural languages has provedto be more difficult than originally thought. Somerudimentary translation systems that translate fromone human language to another are in existence, butthey are not nearly as good as human translators.
There are also voice recognition systems that canconvert spoken sounds into written words, but they donot understand what they are writing; they simply takedictation. Even these systems are quite limited -- youmust speak slowly and distinctly.
In the early 1980s, expert systems were believed torepresent the future of artificial intelligence and ofcomputers in general. To date, however, they have notlived up to expectations. Many expert systems helphuman experts in such fields as medicine andengineering, but they are very expensive to produceand are helpful only in special situations.
Today, the hottest area of artificial intelligence isneural networks, which are proving successful in anumber of disciplines such as voice recognition andnatural-language processing.
There are several programming languages that are knownas AI languages because they are used almostexclusively for AI applications. The two most commonare LISP and Prolog.
Voice Recognition
The field of computer science that deals withdesigning computer systems that can recognize spokenwords. Note that voice recognition implies only thatthe computer can take dictation, not that itunderstands what is being said. Comprehending humanlanguages falls under a different field of computerscience called natural language processing. A number of voice recognition systems are available onthe market. The most powerful can recognize thousandsof words. However, they generally require an extendedtraining session during which the computer systembecomes accustomed to a particular voice and accent.Such systems are said to be speaker dependent. Many systems also require that the speaker speakslowly and distinctly and separate each word with ashort pause. These systems are called discrete speechsystems. Recently, great strides have been made incontinuous speech systems -- voice recognition systemsthat allow you to speak naturally. There are nowseveral continuous-speech systems available forpersonal computers.Because of their limitations and high cost, voicerecognition systems have traditionally been used onlyin a few specialized situations. For example, suchsystems are useful in instances when the user isunable to use a keyboard to enter data because his orher hands are occupied or disabled. Instead of typingcommands, the user can simply speak into a headset.Increasingly, however, as the cost decreases andperformance improves, speech recognition systems areentering the mainstream and are being used as analternative to keyboards.
The use of parallel processing and superconductors ishelping to make artificial intelligence a reality. Parallel processing is the simultaneous use of morethan one CPU to execute a program. Ideally, parallelprocessing makes a program run faster because thereare more engines (CPUs) running it. In practice, it isoften difficult to divide a program in such a way thatseparate CPUs can execute different portions withoutinterfering with each other.
Most computers have just one CPU, but some models haveseveral. There are even computers with thousands ofCPUs. With single-CPU computers, it is possible toperform parallel processing by connecting thecomputers in a network. However, this type of parallelprocessing requires very sophisticated software calleddistributed processing software.
Note that parallel processing differs frommultitasking, in which a single CPU executes severalprograms at once.
Parallel processing is also called parallel computing.
Quantum computation and molecular and nanotechnologywill radically change the face of computers in yearsto come. First proposed in the 1970s, quantumcomputing relies on quantum physics by takingadvantage of certain quantum physics properties ofatoms or nuclei that allow them to work together asquantum bits, or qubits, to be the computer'sprocessor and memory. By interacting with each otherwhile being isolated from the external environment,qubits can perform certain calculations exponentiallyfaster than conventional computers.
Qubits do not rely on the traditional binary nature ofcomputing. While traditional computers encodeinformation into bits using binary numbers, either a 0or 1, and can only do calculations on one set ofnumbers at once, quantum computers encode informationas a series of quantum-mechanical states such as spindirections of electrons or polarization orientationsof a photon that might represent a 1 or a 0, mightrepresent a combination of the two or might representa number expressing that the state of the qubit issomewhere between 1 and 0, or a superposition of manydifferent numbers at once. A quantum computer can doan arbitrary reversible classical computation on allthe numbers simultaneously, which a binary systemcannot do, and also has some ability to produceinterference between various different numbers. Bydoing a computation on many different numbers at once,then interfering the results to get a single answer, aquantum computer has the potential to be much morepowerful than a classical computer of the same size.In using only a single processing unit, a quantumcomputer can naturally perform myriad operations inparallel.
Quantum computing is not well suited for tasks such asword processing and email, but it is ideal for taskssuch as cryptography and modeling and indexing verylarge databases.
Nanotechnology is a field of science whose goal is tocontrol individual atoms and molecules to createcomputer chips and other devices that are thousands oftimes smaller than current technologies permit.Current manufacturing processes use lithography toimprint circuits on semiconductor materials. Whilelithography has improved dramatically over the lasttwo decades -- to the point where some manufacturingplants can produce circuits smaller than one micron(1,000 nanometers) -- it still deals with aggregatesof millions of atoms. It is widely believed thatlithography is quickly approaching its physicallimits. To continue reducing the size ofsemiconductors, new technologies that juggleindividual atoms will be necessary. This is the realmof nanotechnology.
Although research in this field dates back to RichardP. Feynman's classic talk in 1959, the termnanotechnology was first coined by K. Eric Drexler in1986 in the book Engines of Creation.
In the popular press, the term nanotechnology issometimes used to refer to any sub-micron process,including lithography. Because of this, manyscientists are beginning to use the term molecularnanotechnology when talking about true nanotechnologyat the molecular level.
The goal of fifth-generation computing is to developdevices that respond to natural language input and arecapable of learning and self-organization.
Here natural language means a human language. Forexample, English, French, and Chinese are naturallanguages. Computer languages, such as FORTRAN and C,are not.
Probably the single most challenging problem incomputer science is to develop computers that canunderstand natural languages. So far, the completesolution to this problem has proved elusive, althougha great deal of progress has been made.Fourth-generation languages are the programminglanguages closest to natural languages.
sumber : http://www.techiwarehouse.com
No comments:
Post a Comment