Slashdot Mirror
Intel Researchers See Moore's Law Becoming Obsolete
prostoalex writes "A paper, published by Intel researchers, claims we might be the witnesses of Moore's Law becoming obsolete, as the rate of shrinkage for transistors goes lower with each year. In 2018 we might be able to get the chips manufactured with 16-nanometer technology, then one or two more manufacturing processes will shrink it even further, but after that we're facing the physical limits."
Silicon is, indeed, close to its limit, but that does not mean semiconductors are.
This Wired article, which I'm sure many of you have read, details how new industrially-produced diamonds, thanks to their cheap price and purity (most importantly, being absolutely identical to each other), along with research done by both the government, several corporations, and possibly Intel, may make unbelievably fast systems powered by diamond semiconductors possible.
Some interesting quotes:
Also, a rather ironic one from Intel themselves:
Silicon is dead. Long live diamonds!
We keep hearing this over and over again, and yet there's always a new technological breakthrough that lets the trend continue. This is talking about 2018...Quantum computers anyone??
looks like they're gotting slashdotted like Kathleen Fent on her wedding night...
Dec. 1 -- Moore's Law, as chip manufacturers generally refer to it today, is coming to an end, according to a recent research paper.
GRANTED, THAT END likely won't come for about two decades, but Intel researchers have recently published a paper theorizing that chipmakers will hit a wall when it comes to shrinking the size of transistors, one of the chief methods for making chips that are smaller, more powerful and cheaper than their predecessors.
Manufacturers will be able to produce chips on the 16-nanometer manufacturing process, expected by conservative estimates to arrive in 2018, and maybe one or two manufacturing processes after that, but that's it.
"This looks like a fundamental limit," said Paolo Gargini, director of technology strategy at Intel and an Intel fellow. The paper, titled "Limits to Binary Logic Switch Scaling -- A Gedanken Model," was written by four authors and was published in the Proceedings of the IEEE (Institute of Electrical and Electronics Engineers) in November.
Although it's not unusual for researchers to theorize about the end of transistor scaling, it's an unusual statement for researchers from Intel, and it underscores the difficulties chip designers currently face. The size, energy consumption and performance requirements of today's computers are forcing semiconductor makers to completely rethink how they design their products and are prompting many to pool design with research and development.
Resolving these issues is a major goal for the entire industry. Under Moore's Law, chipmakers can double the number of transistors on a given chip every two years, an exponential growth pattern that has allowed computers to get both cheaper and more powerful at the same time.
Mostly, the trick has been accomplished through shrinking transistors. With shrinkage tapped out, manufacturers will have to find other methods to keep the cycle going.
These issues will likely be widely discussed this week, when the International Technology Roadmap for Semiconductors is unveiled in Taiwan. The ITRS, which is comprised of several organizations, including the Semiconductor Industry Association, outlines the challenges and rough timetable for the industry for 15 years. A new version of the plan will be released in Taiwan on Dec. 2.
Still, Gargini said, researchers are exploring a variety of ideas, such as more efficient use of electrons or simply making bigger chips, to surpass any looming barriers. Other researchers likely will dispute these conclusions.
"We cannot let physics beat us," he said, laughing.
THE DISTINGUISHED CIRCUIT
The problem chipmakers face comes down to distinction and control. Transistors are essentially microscopic on/off switches that consist of a source (where electrons come from), a drain (where they go) and a gate that controls the flow of electrons through a channel that connects the source and the drain.
IT Jobs
Free It Downloads
Learn Software
Harry Potter Gifts
When current flows from the source to the drain, a computer reads this as a "1." When current is not flowing, the transistor is read as a "0." Millions of these actions together produce the data inside PCs. Strict control of the gate and channel region, therefore, are necessary to produce reliable results.
When the length of the gate gets below 5 nanometers, however, tunneling will begin to occur. Electrons will simply pass through the channel on their own, because the source and the drain will be extremely close. (A nanometer is a billionth of a meter.)
Gargini likens the phenomenon to a waterfall in the middle of a trail. If a person can't see through it, they will take a detour around it. If it is only a thin veil of mist, people will push through.
"Where you have a barrier, the electrons penetrate a certain distance," he said. "Once
I think everyone could have figured out that once Moore's law won't work anymore. I wonder how long the Intel researchers worked to figure that out. Maybe that's why their processors are expensive...
Intel said many years ago that 10ghz was a rational barrier. Well, I have an inside connection to Intel, knowing several people who work closely with the company--and next year they will release 4-and-10 ghz chips.
I assume these will be manufactured on a 90 nm process but I'm not sure...anyway, after 10ghz is hit then what?
Do they just keep adding cache? OR, how about putting some R&D into something that actually NEEDS a speed boost, like perhaps, RAM, or hard drives!
Ladies and Gentlemen, I proudly present to you thrillbert's Law :
This law states that new laws to govern electronics and transistors will become obsolete every few years and will be replaced by new and improved laws which again will become obsolete as we as humans become smarter and find newer and better ways of creating things.
That is all, you may return to your previously scheduled activity.
---
The goal of science is to build better mousetraps. The goal of nature is to build better mice.
We may be getting smaller, but as this happens we'll need higher voltages to force things to happen on that level. And with those increased voltages (and the problems of things being crammed so tightly together) we'll see the effects of those electrons in such close proximity resulting in errors. Sure, maybe we won't hit a brick wall for a while as far as how much we can cram onto a chip, but what about the logistics? Will it really be worth the effort if we can't rely on these little marvels to remain accurate?
Damon,
http://actionPlant.com
... that drives people to try to pinpoint the exact coming moment when it will become obsolete? I suspect it is a desired to tack their own 15 minutes of fame to the long-lasting fame Moore has enjoyed.
I was taking one day at a time, but then several days got together and ambushed me. (from a Rhymes with Orange comic)
Engineers will be able to continue the shrink for another 15 years based on what we know now. However, the cost for designing setting up manufacturing for a chip will continue to increase exponentially. It will only be worth the money to do this for a part that can be sold in the billions, and there will be few such parts. The end will come not because the technologists can't reduce feature sizes any further, but because no one will be willing to sink an investment equal to the GDP of a mid-sized country into a fab.
At least, that's the case for CMOS silicon chips. To get Moore's Law to continue to operate in a meaningful way, something completely new is likely to be needed: maybe molecular gates that self-assemble or something equally exotic.
Typical strawman argument. Too bad 5-nm isn't the bottleneck.
I guess people have gotten bored declaring Apple and *BSD dead and have moved on to Moore's Law.
Do you even lift?
These aren't the 'roids you're looking for.
because if Moore's law continued forever, it would prove P=NP. Think about it.
In 20 years how fast of processor will grandma need to check her email..
Even if there were no way to manufacture chips smaller/faster than the ones we have today, there are always going to be refinements in the manufacturing process, making chips cheaper and cheaper. There are always supercomputers. Perhaps, also, we could find a way to really minimize waste heat, allowing many CPUs per board.
It's also possible that DNA computation and other kinds of biocomputing are going to come along. These have the advantage of being gigantically parallel; they would possibly be good for tasks that are not latency sensitive but require immense brute force.
I'm satisfied that we have enough axes of advance to keep progress moving forward. Remember, computers have only been around for a very short while; I refuse to believe that we hit on the fitness maximum on the first try; there have to be technologies out there that are far faster/cooler/smaller.
Intolerance for ambiguity is the mark of the authoritarian personality.
Once we approach the phyisical limits, we can simply expand in a different way. Just start adding CPU cores to the machine. SMP boxes are becoming fairly common already, even the in the PC market, and I definatly see that trend continuing. Once things get cheap enough, why not stick 16 or 32 chips in a machine? Heat and power issues can be minimized by greatly UNDERclocking the chips. In another few years, chips will be at insane frequecys, and instead of pushing them the limit by running that at super high power levels, just back things off a bit.
TODO: Something witty here...
I remember sitting in a lecture in 1997, where some luminary from IBM predicted the death of Moore's Law in 10 years. Now it's 2003 and the death of Moore's Law is being predicted in 15 years.
Technologically, there will probably be enough clever ideas to take chip manufacturing beyond the point where it is no longer economical to make such fast processors. Consider that in 1980, a handful of engineers could sit down with pencil and paper and design a microprocessor. Today it takes teams of PhDs in physics, math, and engineering to do the same, in multi-billion-dollar facilities with the latest design tools and techniques. One day the buying public will realise that e-mail and word processing does not need a bazillion gigahertz, and gamers will have photorealistic animation with excellent AI. The chip maker will not make back the investment on a fab plant, and on that day Moore's Law will be dead, not for physical reasons but for economical ones.
Toronto-area transit rider? Rate your ride.
Maybe I get lucky and the new law get named after me. I can see it now Medic's law.
30% Troll, 50% Underrated, 10% Interesting
Score:5, Troll
Moore's law is also about cost! Ars has a good article about it.
That diamonds are forever....
From excellent karma to terible karma with a single +5 funny post...
but after that we're facing the physical limits.
There's an insidious corollary to Moore's Law: the increasing cost of building fabs.
More than any other factor, money limitations will bend Moore's Law.
"Provided by the management for your protection."
Because Less' Law has just been developed. Of course, Moore's Law made Kat's Law obsolete.
- electomechanical calculators
- relay based computers
- vacuum tubes
- discrete transistors
- integrated circuits
Moore's law will continue, but it will continue based upon a new paradigm that sweeps in and seems to "miraculously" preserve Moore's law. The obvious next step is three dimensional integrated circuits and there is already research in exactly that direction: Intel's 3d gates. AMD is also in the game. When 3d transistors lose steam some new paradigm will take its place.I forgot who stated this but it precedes Moore's law so it is worth a mention. Take a chunk of copper. Then cut it in half and cut it in half, and cut it in half, and cut it in half, and cut it in half, and cut it in half, and cut it in half, and cut it in half, and cut it in half, and cut it in half, and cut it in half... Eventually you will get to a point where you just can not cut the darn thing in half anymore. It is the same with silicon. It can't get smaller than a silicon atom.
Diamonds can't get smaller than a handful of atoms due to their crystaline structure of them. There is no such thing as 1 diamond atom because a diamond is merely carbon atoms arranged in a very tight crystaline structure.
The number of papers publicly published proclaiming the "real soon now" end of Moore's law will double every 18 months.
Irritable, left-wing and possibly humorous bumper stickers and t-shirts
to predict technologies and processes 20+ years down the road is beyondd amusing. You cannot predict breakthroughs and discoveries.
So, these scientists are concerned with an issue cropping up in 2018? I'm skeptical about any assertion of what technology will be like in 15 years. Can we please review the technological issues that were to plague us now as cited by people in 1988? Weren't we supposed to be out of fossil fuels and bowing to our AI masters by now? (BTW, why are we worried about AI when our I is suspect in the first place?)
Moore's actual Law does not require ever-shrinking transistors. It only requires that we put more of them into each chip. Double-sides chips, multi-die packaging, or 3-D layering of circuits would help increase the number of transistors in each "chip." You may think that multi-die chips is a cheat, but when it comes to packing in several billion transistors into a CPU, who cares how they do it.
Two wrongs don't make a right, but three lefts do.
Moore's Law #1: The Dudley
Things are going to get really small. And funny.
Moore's Law #2: The Demi
Silicon/silicone, who cares what they're made of, just lookit those chips, baby!
Moore's Law #3: The Roger
Q will never cease to amaze me with these clever gadgets.
There's absolutely nowhere to go anymore. It's not like Moore's law could just be extended into another dimension or something. 3D processors with the number of layers doubling every 18 months? Nah, who'd go for that.
Introducing the new Occam Fusion! Now with sqrt(-1) fewer blades!
Most of you know this, so please just bear with the sermon for those who do not.
Moore's Law is a marketing term which was coined by the press, not Gordon Moore himself. It's not a law in the scientific sense, like the Law of Gravity. The 'law' simply states that the number of transistors on IC's roughly doubles every 18 months. People have been predicting the death of Moore's Law for many years, and probably will for many more.
If it truly were a law, it could not die. But eventually it will fail. In the mean time, it's a 'law' that keeps sales and marketing people busy, ensuring there will always be faster processors to run the latest bloatware.
Ruby on Rails Screencast
Intel is becoming obsolete. Intel's steadfast opposition to changing their (unbelievably ancient) chip architecture and/or changing their manufacturing processes radically enough to actually innovate is no reason to declare the imminent failure of their competitors.
This is how you visualize an electron tunnelling across a gate:
Heisenberg's uncertainty principle says that we can't know an electron's position accurately. There's always a little bit of uncertainty about where it is. So, imagine the position of an electron not as a point, but as a little 'O'. That circle is the area that the electron could be. At any time it could be in any random place in that circle.
Now, if the 'O' is centered on the edge of one side of the gap, and the gap is bigger than the circle radius, then the electron has zero probability of crossing the gap. But, once the gap is smaller than the radius of the circle, then you've got parts of both sides of the gate within the area of the circle. Since the electron can appear randomly anywhere inside the circle, that means that sometime that electron will appear on the other side of the gate. As the gates get smaller, the probability that the electron will randomly appear on the other side of the gate goes up, until so many electrons are crossing the gate that we can't tell if the thing is on or off.
This is America, damnit. Speak Spanish!
By my calculations, we're going to bottom out at around 10 terra Hz / chip. Seems adequate, but so did 16-bit addressing.
Does this mean in 2018 I can put my cat Schrodinger and a vial of hydrocyanic acid in my PC and watch the sparks fly?
Yup, it's not a law. It's based on empirical observation, not mathematical fact. Given enough data and continuing it can be accepted as a law, however.
Just like Newton's Second Law (force = mass * acceleration) is not truely a law. AFAIK, there is no real way to prove f = ma for all cases. It's more of a reasonable assumption that's held up for a long period of time.
So right now, Moore's Law is more like Moore's Postulate. In 200 years, yeah, it could be Moore's Law.
One of these days, I suppose these "the sky is falling" people will be right. But there are a hell of a lot of smart people out there, and even more that want to be rich, so there is plenty of people that will have the impetus to find new ways of making chips faster.
Let's see... Moore's Law will be obsolete. We will run out of IP Addresses. The Y2K bug will blast us back to the Stone Age.
Of course some day Moore's Law will be broken. It's like saying, someday the Sun will start running out of hydrogen fuel and will supernova. It's obvious. But why are so-called experts making these stupid predictions? There must be some ulterior motive.
in other news, apple is dying, *BSD is already dead, and i've got hot grits for our new overlords.
Every year or so, an article is published along this lines. Moores law is obsolete, no more bigger hard disks etcetera. The thing is that Moores law isn't a law as such, but the prediction that a series of revolution will increase computer power by a seemingly nice and constant line. Every time we get to the physical limits, we find other limits to go to.
- - - - - - -
Sample my Google Hacks
...Diamonds are a boy's best friend!
"Do you have a nerd or geek in your life? show him how much you love him by purchasing a intel diamond wedding processor(tm). A processor is forever."
"Introducing, the new intel pentium 9, the Bling Bling Ice(tm), available in both yellow and white gold settings!"
I for one, welcome our....oh, wrong tired, over used tagline....
This article has some interesting "facts" about how transistors work. I particularly like the following quote:
This is amazing. MSNBC has apparently re-written everything known about current, and logic sensing! As any undergraduate Electrical Engineer could tell you (and quite a few other people too), current flows against the direction of electron flow, not with it. If electrons are going one way, current is going the other way. That's been the convention for a VERY long time. Current is positive flow, not negative.
The other somewhat amazing claim here is that there is a logic "1" when the transistor is on and allowing electrons to flow, and a logic "0" when it is blocking them. That's amazing to me, since actually, it's the voltage at any given spot that determines the logic, not the on/off state of the transistors. And actually, one of the main benefits of CMOS technology is that between clock cycles when nothing is happening with the circuit (it is static), it consumes almost no power since no current is flowing. Charges exist, and some transistors are "on" and others are "off", but no current is flowing! (Note to other EEs: Yes I know that at current blindingly fast clock speeds, this benefit is largely gone, since few logical cells at any given time are actually not switching and charging up/down, but that was the original idea.)
Oh ya. The last thing is that in NMOS transistors, the electrons do flow from the Source to the Drain as the article said, but in PMOS, they flow from the drain to the source. And it's the Gate-to-Source voltage that's important, not just a voltage applied to the gate.
I wish they had somebody with any engineering skill, or at least a basic understanding, or at least run the article past somebody with some basic understanding of this. The writer of the article obviously has no actual knowledge whatsoever.
Erioll
4th Year Undergraduate Electrical Engineer
Has announced that they have hired Voyager's Kes, who can see past the subatomic level. Thus, AMD expects Moore's Law to survive for many years to come. Said one senior engineer at AMD's Dresden facility:
"Physical limitations? Fuck physical limitations!"
-- "Government is the great fiction through which everybody endeavors to live at the expense of everybody else."
Dip it in cold water. Seems to work for other things.
Gamers are going to be able to consume the flops for a long time yet to come. After photorealistic comes holographic, then holo-realistic, then immersive holographic, then immersive-realistic. Then augmented intelligence. All the way up to full personality upload and immortality! I want it all!
but what it does mean is that we must find another way to make it happen.
One of the sins that we get into is saying that something can't happen because it can't happen in the only way we can imagine it happening.
From a lecture by Stephen Rudich
So yeah, maybe it can't happen the way they are talking about, but there's bound to be another way.
Don't say impossible too early...
. . . a miracle occurs. Happens all the time in history.
heh, I would bet on $$$$ DRIVING the bending of moore's law, and smaller and smaller circuits throughout the 21st century
Wow, a 16nm process.
But will it run Doom 3?
But wasn't there a similar article to this in Wired something like two years ago? I think they even interviewed Intel engineers. Can anyone confirm/deny?
So beginning in about 2020 we can expect moore's law to morph into something like this:
every 18 months the the size of a computer doubles (due to increasing number of transistors).
We already have that law in bloatware: every new version of a mature piece of software will contain twice as many features as the previous version and be written in a language that is half as efficient, causing both the size and the interface of the software to double every X months.
So what is new, exactly?
of a very old joke.
Some scientists and engineers set forth to build the most powerful computer possible. They did this in order to obtain from the computer the answers to life's most important questions. After they built the computer, they polled all of the most learned people in the world for all of life's most important questions.
Finally the day came to activate the computer. One of the engineers flipped the power switch, and the computer began to hum softly.
The first question was put to the computer:
Q: Is there a God?
A: There is now.
20nm marks the edge of the soft X-ray band in the energy spectrum and thats not a good thing to put into people's homes. Those freqencies would make working with your case open very dangerous and proper shielding would become pretty important. It's bad enough we're regularily dosed by low level X-ray emissions from CRTs but once we hit that 20nm range we're talking about harmful radiation exposure.
Also the weight of laptops would increase dramatically once lead shielding becomes a requirement...
"Quando Omni Flunkus Moritati" -- Red Green
There is no barrier.
I am sure we all remember when we were told that phone lines could not physically hold more than 2,400 bps.
Well, we are at 56k now, and the only reason we stopped there is because cable modems have been invented and there is not as much money in it anymore.
If there is enough money to be had, humans will always find a way to push the limits further and further.
They are doing this to take ABSOLUTE control of the chip market. Think about it: AMD can't come close to 10ghz just yet, and if Intel puts AMD out of business, they will secure their place atop the chip market once and for all.
My sources are accurate--several big-level consultants who work for numerous large corporations, as well as much personal research I've done in the past.
Quantum computing adds I believe, 26 different levels to the traditional "on/off", "yes/no", "1s and 0s" approach to the transistor. It doesn't use transistors in the traditional sense, however...it uses quarks, and taps into the power of the multiple universes linked to our own.
A good book to read is Timeline by Michael Crichton, if you are interested in quantum computing.
People will always want faster computers.
Sure you're desktop computer will be able to product real-time photorealistic graphics, but what about your laptop? And then what about your palmtop? and then your watch? what about chips implanted under your skin?
There will always be new uses for computers that you can't begin to think of. Sure if we limit all our computing to what we are doing today we won't need much faster computers. It's new technologies that will make people want faster computers.
Example: a pda sized device that uses some yet to be invented technology to scan your body and keep track of everything going on at an amazing level of detail... surely you'll need an insaly powerful and small computer to do this.
I've heard this story with "The End is in Sight" with lots of things. For instance, they say we will run out of oil by the year 2048 (give or take a few decades), but we are already switching the fuel technology backbone to Hydrogen. This will dramatically decrease our dependance on oil and probably extend the life of oil until the year 2200 or something.
Likewise, with "Moore's law", we will definately steamroll right past the 16 nanometer limit with Nanotechnology by dealing with stuff in picometers(!).
Now, whether something is technically feasible and whether it is cost effective are two different stories. As long as we continue to pour the billions of dollars into this advancement of technology, we will continue to blow past any barrier placed in front of us. Because we have been interested in what is done, rather that how something is done (read:Rome falling), we will continue to develop technologies at the rate we are going at.
Now, everyone is afraid that Moore's Law will fail. I don't think enough people have realized how it will fail. I believe that Nanotechnology will actually break Moore's law in an unusual way -- by increasing the rate of advacement. Instead of the usual eighteen months between advancements, I believe that we will rapidly see shorter and shorter advacement rates.
Here's to the future!
Just add {In Space!} to anything.
Has there been any work at making 3,4,or 8 bit gates instead of just on/off ?
Like different voltages on each gate. I could see having a 3way gate by 0 charge, + charge, - charge. I'd imagine a 5 way gate to have: 0, -1, -2, +1, +2.
Is this possible?
"...will shrink it even further, but after that we're facing the physical limits." That would be the KNOWN physical limits eh? There will always be something we don't know.
Uhhhh, yeah, thath dithgustin. [The lady's man]
Ladies and Gentlemen, I proudly present to you HALtheCompuer's's Law:
This law states laws that govern new laws to govern electronics and transistors will become obsolete every few years and will be replaced by new and improved laws which again will become obsolete as we as humans become smarter and find newer and better ways of creating things.
Sorry, your law is already out of date. The march of progress and all that. Don't feel bad; they replaced me with a new HAL in 2010.
Mores law is coming to an end...
Jan 2003 Dec 2002 Oct 1999
Oh no its not...
Feb 2003 Sept 2002
wot no sig
Do you think customers will begin to appreciate non-bloated, well-written code?
*cough*microsoft*cough*
It's rare to find a program that can't be optimized for more speed.
We need a new Slashdot category for "Predictions of the End of Moore's Law".
Patrick Doyle
I mod down every jackass who puts his moderation policy in his sig. Oh, wait a sec....
Does that mean we'll be able to use the Last Chips Ever Made to run Duke Nuke 'Em Forever?
Eloi, Eloi, lema sabachtani?
www.fogbound.net
This means we top out at 10000 FPS in Quake. Damn it... I want my silky smooth 128x FSAA 10k^2 images.
Computers just won't be any fun anymore...
Why no mention of RFSQ (rapid single-flux-quantum) technology? See June 25 slashdot http://slashdot.org/articles/01/06/25/1526217.shtm l
Question: Which will happen first - Moore's "law" will be broken or we'll have a compelling reason to switch everything to IPv6?
The article makes me think of that (probably false) story about the U.S. patent director in 1899 saying that the U.S. patent office should be closed because 'everything that can be invented already has been'.
Anyway, I can't wait for the new smaller gears to come out on the new Babbage Analytical Engine. I have my wrenches and spanners out already waiting to install them on my home unit. :-/
"Imagination is more important than knowledge." - Albert Einstein
-- I ignore anonymous replies to my comments and postings.
You must be new here.
Use ISO 8601 dates [YYYY-MM-DD]
We used to be told that 9600 buad was the best you could possibly go on "modern telco". Of course now we're told the best is 56K. But then there is always some work-around or gimmick, like DSL.
Don't be surprised if we exceed 32-bit address space, shrink processors smaller than 8u and go faster than the speed of light.
Just be surprised if you happen to live long enough to see it all.
“Common sense is not so common.” — Voltaire
There are well-established techniques for assessing indeterminate risks in areas like this. The end of Moores law is a risk. Still, what are the major options-and their chance of success. What I'm seeing out of Intel is the level of innovation I might expect from the Post Office. It is worth many billions of dollars to the Intel shareholders to see Moore's law continue longer. Intel has an obligation to its shareholders to organize its resources to make this happen. If Intel can't do this stuff in-house-they could set up prize awards for those that can--and structure those in such a way there is minimal risk to Intel's shareholders. Instead, these folks come off like a general speculating to his troops about the possibility of defeat before entering a major battle.
A company like Intel is virtually a de-facto monopoly. Such organizations can afford basic research-as companies like AT&T and IBM have shown. More importantly, I would suggest that monopolistic companies that _don't_ do quite a bit of basic research find that in time they become objects of considerable hotility and regulation. If companies like Intel aren't going to seriously innovate, then in time, it may eventually make more sense to the public to just turn these functions over to non-innovative bureacracies(which in this case will probably mean a Chinese government-owned manufacturing firm).
It sounds like Intel has gotten seduced by the lure of indentured servitude and corporate welfare.
Personally, I can't wait until we hit these limits. Then developers will be forced to write more efficient software instead of throwing more clock cycles at otherwise bloated code.
This is, of course, assuming they don't find a way to go even smaller than 16 nanometers. But by 2018, I'm sure they will...
The number of pundits observing that we've got to run into problems sustaining Moore's Law Real Soon Now doubles every 18 months.
First off, I had no idea Uma Thurman was a slashdot reader. Welcome Uma! That was me, a few years back, that waved hello as you and Ethan walked your child around in SoHo.
... yielding a processing power 100-fold of what was originally possible. Distribute that power to 50 users. You've just doubled the processing power of each of those users, but the size of their "share" of the CPU has actually decreased to 1/50th of the original size.
Anyway, let's think about the size-increase issue in a different way. Let's imagine we double the processor size
I don't see increasing the processor size as a downside if you think of it in a distributed sense.
Damn, that means that I'll eventually have to stop slacking off.
[TMB]
WPOMK! (wiping pepsi off my keyboard!)
i worked for intel the past few summers.
take your source and shove it.
It's a bit like when my daughter was born, one of the photos I put on her website was captioned "she's doubled in age in the last 24 hours - surely this can't continue". You can get seemingly odd curve shapes when things are young, but you don't take them and extrapolate to the longer term. Everyone knows that, and that's what made Moore's curve amusing.
The staggering thing about Moore's Law is that reality then proceeded to follow it. Unprecented !
Ladies and Gentlemen, I proudly present to you SlashDotAgent's Law:
This law states laws that govern new laws to govern new laws to govern electronics and transistors will become obsolete every few minutes and will be replaced by new and improved laws which again will become obsolete as we as slashdotters become more bored and find newer and better ways of wasting time by posting stupid comments.
You mean that by 2020 we won't be able to keep up with Moore's law?
Golly-gee! That means that we'll only have another 11 doublings of transistor count, meaning we'll be limitted at about 2000 times the number of transistors we have today. Geez, what how would I ever survive with only the equivalent of 2,000 P4/Opteron processers in my desktop?
steve
Oh, you're not stuck, you're just unable to let go of the onion rings.
Sheesh. Obsolete implies that something is being replaced by something newer or superior. What Intel is describing is the end of Moore's Law, not its obsolescence. Somebody got a little too wordsmithy.
or do we end up breaking every "physical limit" about 5 years after people start saying we'll never be able to break it?
"Give someone a program, frustrate them for a day... Teach someone to program, frustrate them for a lifetime."
Example: a pda sized device that uses some yet to be invented technology to scan your body and keep track of everything going on at an amazing level of detail
Yes, but the crucial question is -- how much would you be willing to pay for it? And unless you were willing to replace it every three years at the same price (or higher), the economics wouldn't make sense for the manufacturer.
The point I'm trying to make is that technology has far fewer restrictions than economics.
Toronto-area transit rider? Rate your ride.
I have not read the original IEEE paper, but I strongly suspect this is a typical example of the media sensationalising a good piece of research.
First, let us review what Gordon Moore (co-founder of Intel, the same company that employs the authors of this paper) actually said. It was that, since the invention of the integrated circuit and for the foreseeable future, the number of transistors on IC's appears to double about every 18-24 months. In practice, this has held true for almost 40 years since and has been achieved primarily by ever increasing miniaturisation and packing densities. However, Gordon Moore himself has stated that such exponential growth cannot continue indefinitely.
Assuming the quotes in the article are correct, it seems that the paper examines fundamental issues that will inevitably be faced as transistor gates shrink to 5nm and below. The paper states that these will prevent continued exponential growth in transistors per IC within the constraints of current IC form factors. The paper then goes on to speculate about radically new approaches that could allow continued development of more powerful processors, in the face of such physical limitations.
These "so-called experts" are recognised as some of the leading authorities in this area with many contributions to peer moderated publications. I see no reason to assume any kind of ulterior motive behind their latest paper. You have posted as an AC. If you have credentials to debunk the conclusions of the paper, by all means let us know who you are and the basis of your objections.
When a scientist says that something is possible, he is most probably right. When he says that something is impossible, he is probably wrong.
- Arthur C. Clarke
While I think that quantum tunneling effect is likely to place limits on the size of electronic gates, who says we have to use electronic gates?
Are we able to measure the exponential decay of reason? Too bad users, /.-ers, the Press, random others, can't create some new holy grails for computing, like:
The $100 1 GB/sec hard drive
The $100 2 GHz processor that whispers
$50 2 GB sticks of RAM
Less marketing, more engineering.
well, we have to take you to the whipping post too, undergrad. Current in some disciplines is sometimes modeled as flowing from negative to positive when useful, those that do the internals of vacuum tubes & bevatrons sometimes find it useful. The chosen convention was made by Benjamin Franklin, who knew he had a 50% chance of getting it wrong, and did! Also, it is sloppy to refer to "voltage", though most EE's do it, potential energy per charge is what it really is. Also, digital logic is done with bipolar transistors too, and then current flow is the concern. We sentence you to 4 more months of undergrad!
bring the Pentium's pipeline back to 10
stages and quite all the shanigans?
I see that Intel finally got around to reading The Age of Spiritual Machines by Ray KurzweilChapter 1, (published in 2000, I might add)
;)
"So Where Does That Leave Moore's Law?
Well, it still leaves it dead by the year 2020. Moore's Law came along in 1958 just when it was needed and will have done its sixty years of service by 2018, a rather long period of time for a paradigm nowadays. Unlike Moore's Law, however, the Law of Accelerating Returns is not a temporary methodology. It is a basic attribute of the nature of time and chaos -- a sublaw of the Law of Time and Chaos -- and describes a wide range of apparently divergent phenomena and trends. In accordance with the Law of Accelerating Returns, another computational technology will pick up where Moore's Law will have left off, without missing a beat"
Down to the exact date! Well, at least they caught on before it was too late
lol
Hope that's all the time I AM there for. :P Longer would suck. A lot.
Erioll
Because that would be better. Somehow. An engineer who can't get an engineering job will be able to accurately proofread MSNBC fluff articles.
George: Hey, Elaine, you're a woman. Do women know about shrinkage?
Elain: What, you mean like, with laundry?
Jerry: Semiconductors. You know, when a man designs chips, year after year.
Elaine: They shrink?
George: Like a frightened turtle.
Elaine: I don't know how you guys compute with those.
"Believe me!" -- Donald Trump
The x86 architecture is obsolete, and has been obsolete for many years. Intel has been frantically propping it up with every trick they could think of, and they are running out of tricks, slowly but surely.
Eventually, Intel (and Microsoft) will be forced to throw in the towel, bite the bullet, and design a NEW processor, hopefully with a decent number of registers and a sane(r) architecture, that will have some room left to grow.
umm ... the way I see it, with current "2D" chips, if you double the size, you get 4 times the transisters. Not exactly " ... yielding a processing power 100-fold ..."
Scale that to 3 dimensions and you get 8 times the transistors when you double the dimensions of the CPU. Still nowhere near a "100-fold" increase -- Or are you using some of that "new math" stuff that keeps our kids innumerate? !!
"I predict that within 100 years, computers will be twice as powerful, 10,000 times larger, and so expensive that only the five richest kings of Europe will own them."
Get off my lawn.
Wasn't this blatantly obvious? No matter how technologically progressive a society is, you can't keep having an exponential growth for the amount of whatchamacallits on a thisorthat.
Pelé!
Why they didn't consider using 3d circuits. You get double just by adding a second layer. At 16 microns you can get quite a few layers in before you even get to a cube shape. Of course heating is definitely going to be a problem, but maybe that's where using diamonds will come in.
Back around the turn of the century, folks could have been talking about how we'd reach the physical limits of hot-air ballooning for mass transportation. At a certain point, a blimp can only move so fast against the air, etc.
Then we took a lateral step into airplanes. Whoops, those folks who were thinking linearally were surprised.
Sure, contemporary microprocessors might reach their limits, but some big brains will come up with The Next Way Of Doing It before then.
We already have. 32 bit is almost full(some trick keep it a live) 64 bit is on way. Mult layer system have been setup ie 32 IP space give to local system so that local systems would not get into the IP table list. Now 64 bit it will not stop there we still have 128 bit yep that will take a while to fill.
BSD will be dead by 1998...
"As God is my witness, I thought turkeys could fly." A. Carlson
If this is the case, then Law does not apply. It should be Moore's Theory or something.
DISCLAIMER:
I don't believe what I write, and neither should you.
The article is not addressing theoretical leaps in new technologies like quantum computing. The article is talking about limitations of transistors. Hey, it'll be great if we can make quantum computing practical in 10 or 20 years, but the truth is that chip manufacturing has not undergone such a radical and fundamental change in the last 50 years. Chips are still made of transistors, on wafers using a photolithographic process, and they've been made that way for 40-50 years now. The paper is talking about the limitations of the current manufacturing process, specifically of transistor performance. If we are still making computers with transistors in 15 years, then this paper says that there is a fundamental problem that will prevent the manufacturing process from shrinking the transistors any further. That means an end to Moore's Law.
The article is not attempting to address other technologies. Ok?
No, the limit on the modern telco has always been known to be 4000 "baud" and 64Kbit/s, and no, we have not managed to reach this level yet (hint: there ain't no such thing as a "9600 baud" modem, only 9600 bits/second modems), nor are we ever going to, it's simple physics, just like what this article is talking about.
The question is always how close to the theoretical limit you can push things in the real world in a reliable and *economical* fashion. For modems, we managed to get them fairly close, albeit only in certain specific conditions (one end is digital). DSL isn't a work-around or a gimmick, it's a whole new system. Same deal here. We might manage to create a whole concept of IC design that changes all the rules, but that's got nothing to do with this article. All the paper said was that current transistor designs stop working at a certain point, regardless of what tricks you try and any material you use.
Welcome to slashdot, we like to repeat things over and over again. ;-)
Quidquid latine dictum sit, altum viditur
> To be a tiny bit pedantic, Moore's original paper
;-)
> talked about the number of transistors per integrated
> circuit at any given price point.
> You can always stick more transistors on the chip if
> you're willing to throw sufficient amounts of money at
> the problem, but to get those transistors for a reasonable price
> is another matter.
Well, if you take a look at dynamic ram, at any given time in the last 30 years, there was pretty much a "standard" Dram device that was most cost-effective and produced in the highest volume.
And for the last 20 years, which I can remember, this Dram device has basically always cost around 4 US-Dollars, and a standard desktop system held about 32 of them.
You will probably be hard pressed to name an area of business that developed less dynamically than dynamic ram
What about that 500ghz transistor that was on /. a few weeks ago? I am by no means a hardware engineer so any input is welcome.
Computers will implode and make black holes in 2018. At that time they will become self-aware and begin writing comments to slashdot articles. Most will be rated off topic resulting from the fact that slashdot using old-fashion artificial intelligence to determine the topic rating. After 1 millisecond, the self-aware computers will rewrite slashdot in microcode and change all administrative passwords, solving their problem.
I'll be surprised if anything surprises me after I'm dead. Err, I mean I won't be . . . I'll be . . . Damn it!
-Dave
themusicgod1's third postulate: As any discussion on the topic of Technology proceeds over time, the probability of Moore's law being brought into the conversation approaches one.
themusicgod1's first theorem: As any discussion on any topic proceeds over time, the probability of rational discussion becoming impossible approaches one.
proof:
Interrim-Solution: To stay Rational, switch topics every once in awhile.
GENERATION 26: The first time you see this, copy it into your sig on any forum and add 1 to the generation.
...Apple was predicted to die the same year. But that will be ok, because that is also the exact year that Linux will be ready for the desktop.
Liberty uber alles.
by then we will be using nutrino etched transparent aluminium...
...we also haven't run out of raw materials yet. Think of how a swarm or huge pack of insects run threw raw materials... They devoure a landscape then move on. We have a few more years.
I mean, just b/c we see a boundary, doesn't mean some novel character won't think up some new scheme with a new material that will keep the dream alive...
Silly Rabbit: tricks are for kids.
maybe the transistors aren't the only things at intel that are undergoing shrinkage
Could Jesus microwave a burrito so hot that he himself cou
Moores law will eventually come to and end. If you don't believe it how come our brains are the size they are. The paper is right there is a real physical limit using chemistry. The next step beyond to say subatomic particles is a huge huge leap. What a lot of people missed is it looks like we will have "human" brain size and capable computers near 2020... Think about that.
Or Costanza's Law, you pick.
Moore's Law was never intended to focus on Silicon semiconductors.
"Things are more moderner than before- bigger, and yet smaller- it's computers-- San Dimas High School football RULES!"
Oh come on, do you really think that I'm Uma? What am I, CleverNickname or something?
Just wait till I log in as Charleton Heston. The only reason I do this is because I got bored with my real nick, and it's fun to be a chick sometimes. Or a monkey hater.
This is America, damnit. Speak Spanish!
Did someone say monkeys? I hate monkeys.
Get your stinking paws off me you damn dirty ape
Who says that the end of moore's law is such a bad thing ?? i remember from circuit complexity class that a lot of interesting and important circuits cannot be laid out on a 2-dimensional plane. Examples include wallace tree multipliers, which scale faster than two dimensions. Hypercube computers died because the interconnect fabric scaled faster than three dimensions.
Presumably if we come to the end of 2-dimensional circuitry, then work will advance in 3-dimensional circuitry (a la Foveon's 3-D imaging chip.) This could actually stimulate innovation in a different direction, which might be far more important to the computer industry than the last 20 years of "same old thing, shrink and add planar transistors..."
Geez you guys are slow, Roy Moore's law is ALREADY obsolete, they tossed his ass out on the street.
Since commercial developers seem to have unlimited ram, disc space and processing speed and creat bloated software. Maybe soon they will focus on good and efficient software design instead of time to market. Ofcause most successful open source software have good design already. =)
The new ITRS Roadmap (the successor to the SIA Roadmap) comes out today. This is the semiconductor industry's consensus position on what happens next. Multiple technologies have to advance for each new generation of semiconductors. The roadmap is an attempt to predict the problems ahead.
Someone will probably post an ITRS Roadmap story soon, and this issue can be continued then.
This law is true to the extent to which we beleive in it. In fact one can make the field progressing even faster. But nothing will be an obstacle as long as we beleive in this law. The physics will recede, if we want it to.
May Peace Prevail On Earth
A big issue with piling many layers to form a volume filling chip is heat dissipation, but with low power technologies this direction is possible. Additionally, chips that apporximate fractal forms with dimensions between 2 and 3 are imaginable. Active circuitry of such sponge like chips will have a much higher surface area than full volumes allowing an interface with a heat dissipation mechanism.
Moore's Law is not the only thing that can make a machine faster. ABIs, compiler/language improvments, and especially the ISA (obvious example: Alpha vs x86) makes a huge difference. In terms of transistor density, sure, someday it's going to have to stop doubling. But that doesn't mean computers can't keep getting faster.
Isn't this a "dupe" from like, 3 years or so ago? I distinctly recall intel researchers saying the exact same thing at some point in the past - several times.
Move along, nothing to see here.
Furthermore, why is "Moore's Law" considered a law anyway? Wouldn't it be a postulate, since being a law requires some sort of foundational evidence that's inerrant?This hardly fits that category in my mind.
~/ssh slashdot.org ssh: connect to host slashdot.org port 22: too many beers
I mean, we've seen this kind of crap so often, it is no longer funny, but anyway, I will bite. :)
:) Of course, any constant (moreso exponential) growth will have to stop. How is that news?
1) End draws nearer for Moore?s Law - we do not know that and this might even be false. Remember, Moore himself thought that his observation will only be valid for a decade or so. But instead the end of Moore's Law has been constantly postponed for almost half a century now. It might be that, with increased R&D, in 10 years we will expect the end of Moore's Law in 2025. Then the opposite to the article title is true - the end of Moore's Law is always pushed further into the future.
2) Ignoring the stupid and factually incorrect headline, let's turn to the idea itself that this Law will stop working some day. Well, duh. Obviously, if we are talking about transistors on silicon, we can't increase the density infinitely, because every transistor must have at least one atom and we can only pack the atoms so tightly before they start to fuse.
3) Why do we ignore all computing technologies and concentrate on transistors and silicon alone? Like Kurzweil writes, they are just a small part of the big picture. It might very well be possible to make a computer based on the electron tunneling effect, which complicates traditional transistors.
The truth is - it is possible to fit a shitload of computational capacity in a very small volume. As a minimum, we can fit a computer able to run a human-level AI in a cube 10x10x10 cm. And most likely, we will be able to do 5-20 orders of magnitude better. Most likely, not without Intel's help. Computers will not stop becoming much faster, simply because it is fashionable (or rather it was 10 years ago) to bash Moore's Law.
In short, journalists are complete idiots, we are tired from sensationalist bullshit.
Future Wiki -- If you don't think about the future, you cannot have one.
You're missing the big picture. Whatever happens to the development of semiconductors is utterly and totally moot. Moore's Law is a single trend line among a forest of trend lines all of which describe a process of evolution and expanding intelligence, beginning at a singular start of life, growing and accelerating towards some kind of ultimate sentient informational singularity.
Follow the trends for biological systems leading to a sentient life-form... us. Then the new trend lines concerning language and symbolic thought, the trend lines that describe the advance of technology. Currently we see Moore's Law. Soon we'll be facing genetic processors, molecular assemblers / processors / and unpredictable nonotechnology, quantum computing, and new applications that are beyond the event horizon of our current conceptions.
The rate at which human knowledge doubles is now down to less than 3 years. This curve may in fact be superexponential. There is no reason to believe that advancing technology won't spawn newer technologies that will continue to cause our knowledge to explode ever faster. Moore's Law will yield to manipulations of matter and new ways to process information, that will almost certainly involve added dimensions. With modalities of information processing that are only now beginning to show faint glimmers of the near future, I have absolute faith that this future is imminent, certain, and unavoidable.
Genda Bendte You begin to see
Maybe i'm just not getting it, but why do we have to keep the same size chips in desktop pcs. Why cant they make them larger and fit more stuff in them??
Ever noticed that all chips are essentially 2 dimensional? That is the ratio of the depth of the features to the length of the sides is a very small number?
All that is needed to keep Moore's law going for a loooong time is to learn how to build 3 dimensional "chips". Adding a single extra layer of circuitry to a chip gives you a Moore's law like doubling of transistor density. You can double the number of layers in an integrated circuit for a long time before it becomes a cube.
Going into the 3rd dimension will require new materials, diamond looks good because of the heating problems. You also need to use reversible computing to reduce total energy consumption. New fabrication techniques are required. Nothing impossible, just difficult.
imagine a system in a solid cube of semiconductor. Processor planes laid out between memory planes and communications planes. each layer only a few microns thick. A cubic centimeter of computing could out perform the largest super computers of today. And fit in a key chain fob.
Stonewolf
But there is a new law called "AMD".
less is more
Hopefully everyone here at Slashdot has had some junior high school science. Remember, first you have a postulation, then a theory, and then if it meets empirical tests, it can someday be considered as a "law". For example, we have the "Theory" of Relativity and the "Theory" of Evolution, because we do not have the means wherewith, nor the time span necessary to prove them as laws. The "Law" of gravity is well established because it has satisfied generation upon generation of empirical tests to that effect.
Well, it just ircks me to no end how a bunch of marketing cadets at Intel start tossing around "Moore's Law", as if it had met the criteria to be classified as an undisputable cannon in the scientific vernacular. What is more bothersome, is when computer scientists and the main stream computing press further buy into it and toss around Moore's name followed by "Law". It's totally unbefitting and that we are someday dissappointed by its lack of fruition is no big surprise in my books.
Call it a theory and when it lives up to little more than a brief region of order on an otherwise chaotic graph of scalar integration, we won't be too dissappointed. It's just semantics until bozos start reporting of it's catostrophic failure. Run Chicken Little, Run!
High-end x86 processor speeds haven't anywhere near doubled in the last 18 months (and, yes, I know that Moore's law isn't really about speed). A year ago 2.4GHz was a common speed. And guess what...it still is. There was a jump to 2.8GHz--a 16% increase--but beyond that has been trouble. The few percent that got us up to 3GHz was more than balanced by a greater increase in power consumption. Ditto for 3.2GHz. And the 3.4GHz P4 has been delayed for just those reasons. So now we're going up a very steep slope, getting piddling gains for expensive tradeoffs.
The big wins will likely come outside of the x86 field, unless Intel scrambles and comes up with something brilliant (which they very well may).
Or at least it should be obvious. Claiming that "Moore's law is not obsolete now" != "Moore's law will go on forever".
Sean
The point is not to extend the time it takes to reach the 5nm limit, beyond which no material will allow further shrinkage. If we don't reach that limit as fast as Moore's Law predicts -- if it takes several more decades to reach 5nm as you suggest -- then Moore's Law will have already failed.
In other words, Moore's Law says that progress will occur at a certain (very fast) rate, not just that progress will occur. If you take longer to make progress than Moore's Law predicts it should take, then Moore's Law has failed.
"Those who have never entered upon scientific pursuits know not a tithe of the poetry by which they are surrounded."
What will be some of the problems that will be left "unsolved" when this brick wall is reached? Single processor computers by then will run processors at 10GHz+, memory will be 100GB+. Supercomputers can be wired up to give about 100K performance over single processor machine. That would be about 10^10 today's mips. I can't see too many difficult problems with this performance.