In 1965, a certain Gordon Moore wrote a certain paper that proposed that the number of transistors that could fit on an integrated circuit would double nearly every two years. This observation is what has come to be known as Moore's Law.
What does all that mumbo-jumbo mean, you may ask? Well, before getting to the root of the problem, I think the term "transistor" should be defined, mostly for my benefit. Bob Brown of Southern Polytechnic State University explains what a transistor is in his web lecture:
"A transistor is an electronic device with three elements called the base,
What does all that mumbo-jumbo mean, you may ask? Well, before getting to the root of the problem, I think the term "transistor" should be defined, mostly for my benefit. Bob Brown of Southern Polytechnic State University explains what a transistor is in his web lecture:
"A transistor is an electronic device with three elements called the base,
the collector, and the emitter. Transistors can function as analog devices
and are used in that way in radios, amplifiers, and similar gear. By
choosing suitable transistors and providing suitable inputs, they can also
function as digital devices -- switches that are either on or off."
Simply put, as Brown goes on to say, transistors are what allow computers to "operate as fast as they do." With this in mind, we can easily see the magnitude of Moore's Law; the exponential increase of transistors means that computers will become more advanced and the gratification from them will be nearly instantaneous.
One point of interest brought up by many regarding Moore's Law is its perpetuity. An article from Absolute Astronomy (strikingly similar to the one on Wikipedia) states that in his paper, Moore observed that the trend he would popularize began in 1958 with the invention of the integrated transistor and remained true up to the time he wrote it. Why shouldn't he predict that it would continue for years to come and gain world-acclaim for it?
Though I am quite ignorant on the subject, I believe Moore's Law will persist for the next couple decades (but not if the world ends in 2012), the reason being that there is no reason to stop. An article published by Gregory Johansson addresses the issue that engineers face now, which is the "running-out" of room on a chip they can not shrink for all the transistors they keep doubling. It's like having a party when your parents aren't home and each friend brings two more; eventually, you won't be able to accommodate for all of them. This is also one of the ultimate limits of Moore's Law.
Simply put, as Brown goes on to say, transistors are what allow computers to "operate as fast as they do." With this in mind, we can easily see the magnitude of Moore's Law; the exponential increase of transistors means that computers will become more advanced and the gratification from them will be nearly instantaneous.
One point of interest brought up by many regarding Moore's Law is its perpetuity. An article from Absolute Astronomy (strikingly similar to the one on Wikipedia) states that in his paper, Moore observed that the trend he would popularize began in 1958 with the invention of the integrated transistor and remained true up to the time he wrote it. Why shouldn't he predict that it would continue for years to come and gain world-acclaim for it?
Though I am quite ignorant on the subject, I believe Moore's Law will persist for the next couple decades (but not if the world ends in 2012), the reason being that there is no reason to stop. An article published by Gregory Johansson addresses the issue that engineers face now, which is the "running-out" of room on a chip they can not shrink for all the transistors they keep doubling. It's like having a party when your parents aren't home and each friend brings two more; eventually, you won't be able to accommodate for all of them. This is also one of the ultimate limits of Moore's Law.
Fortunately, these engineers have also come up with a solution, and that would be memristors. Johansson defines this joint-project of HP and Hynix as "memory resistors, which means that they can store information even in the absence of any source of power." He explains that this is great news "because they open up hope of exceeding the current chip capacity limits thought to [be] inviolable." Though not fully developed, knowing that a design to work around one of the foreseen problems of Moore's Law has been thought up, and the fact that it is being worked on, shows just how indebted and dedicated people are to this trend, and to the technological world as well. In this way, Moore's Law will hold in the future.
The main limit that butts heads with Moore's Law has been discussed - my coined "house-party" phenomenon - so we can move on to the futurists.
Some futurists view Moore's Law in a positive light, believing that it will continue to prosper (directly or indirectly) in the coming years. According to Absolute Astronomy, futurist Ray Kurzweil's understands that some new technology will take the place of our current "tech", but "that the exponential growth of Moore's law will continue beyond the use of integrated circuits into technologies that will lead to the technological singularity." This graph by Kurzweil illustrates the trend Moore had noted and predicted, along with a few additions (e.g. vacuum tubes and electromechanical computers). Not really relevant. I just thought a little variety might help. But reading Kurzweil's thoughts really made me look brightly at the future of technology. If Moore's Law applies to me later in life, I can only begin to fathom how awesome everything else will be.
Of course, there are others who are not entirely on the bandwagon, predicting the worst for Moore's Law without change. One such critic is Penn State's Suman Datta. This article from the Daily Galaxy shares with its readers Datta's belief that if Moore's Law is to survive, "some new technology will have to take over from silicon", whether it be carbon nanotubes or superconductors. Datta's words are not necessarily an attack against Moore's Law, but more of a cautionary wagging of the finger that if nothing is changed now, it will cease to change ever in the future.
And then there are those who are against it entirely, such as Martin Ford, who basically warns about a passive robot revolution; passive in that menial work will be taken over by them, and so people will become lazy and fat and the economy will take a nose-dive. Though it sounds kind of cool, I could also see this future, brought up in the Absolute Astronomy article.
Even still, I am sure that Moore's Law will continue to hold true. And I am sure our generation is certainly with me on that one. All about the iPhone and touch-screens, practically hand-held computers on their own, the teenage population of this era is, and always will be, ready to whip something out in the moment and get what they want fast.
No comments:
Post a Comment