end of moores law.IT companies in panic?

[endofIT]

hi i a am new here.After the 2009 idf where intel demoed 22nm i wanted to ask how far intel and other semicon companies can push the miniaturization ?

after 22nm it is 16nm but even at 16 nm quantum tunneling takes palce
http://en.wikipedia.org/wiki/16nm
. It has been claimed that transistors cannot be scaled below the size achievable at 16 nm due to quantum tunneling, regardless of the materials used.


11nm

At scales of ~10 nm, quantum tunneling, especially through gaps, becomes a significant phenomenon.[7] Controlling gaps on these scales by means of electromigration can produce
interesting electrical properties themselves

http://en.wikipedia.org/wiki/11_nanometer



so by intel road map 22nm will be out in 2011 and after that only 1 more fab would come out that is 16nm in 2013.What effect would it have on the economies?
I doubt that 16nm can even come out due to quantum tunneling.


I beleive that the continuous artifical stimuls to the economy that the IT industy provides will stop by 2013\2014 and then there would be no more advancement.
remember we are reaching the fundamental limits of atoms.nothing can be made smaller than a atom.all alternative theories propsed like dna\organic computing are lies\false expriments just to keep us comfortable.Also many IT people have proposed to have vertical scaling but that is unfeasable for home consumers due to heat and size but the bottom line it that there will only be 2 more fabs aka 22nm and 16 nm and then thats it.

I beleive that in 5 yrs the entire IT industry will collapse.

cnet article from 2003
http://news.cnet.com/Intel-sci...00-1008_3-5112061.html
In chips made on a 16-nanometer technology process, the transistor gate will be about 5 nanometers long.

in 2003 it was thought that 16nm will come out in 2018 but now it will come out in 2013 so its over in 5 yrs.so the 2014 stock market crash would be even more bad then the 2008.

 

[BTRY B 529th FA BN]

That is when the fun will begin.

 

[endofIT]

can you comment on my posts like will it be 16nm or 11nm? and what effect will it have on IT based economies?.No more advancement,no more powerful apps,games.

 

[Idontcare]

Welcome to the forums! :thumbsup:

Checkout this thread and the myriad of links and articles therein.

It is not intended to be a comprehensive "what are the limits of scaling" type thread but it does contain lots of links and good stuff that at least tells you how far we should be able to go with already existing pathfinding options, not to get into the more fanciful stuff like CNT', graphene xtors and 3D ICs.

 

[cusideabelincoln]

I'm no Physicist, so I'll just say 2013 won't matter because of 2012, baby.

 

[endofIT]

thank you but plz dont del this thread or merge it.I want to have a healthy discussion and not 2012 stuff.

whats your prediction.plz post

 

[Idontcare]

Originally posted by: endofIT
thank you but plz dont del this thread or merge it.I want to have a healthy discussion and not 2012 stuff.

whats your prediction.plz post

My prediction? Its in the thread linked above:

Originally posted by: Idontcare

Originally posted by: bgeh
Well we're primarily (as hardware enthusiasts) concerned with the leading edge (well everything perhaps chipsets), and my question to you would be: Do you forsee a slowdown on the leading edge then? [Let me also throw in a few assumptions: Assume China/India continues its path of growth, so you still have the prospect of rapidly growing markets with increasing demand, vs. the US/Europe with relatively stable demand]

Yeah a slowdown is inevitable, again not for the physics or the challenges but because of the economics of funding the resolution to their technical challenges.

Once Intel has a measured lead over the rest of the industry, I estimate around the 16nm node they will have iterated themselves to the point of being a full node ahead of the second place position in the industry by that time (i.e. Intel debuts 16nm at nearly the same time as competitor XYZ is debuting node N-1 while the rest of the industry is still on N-2), the management at Intel will have an ever more challenging situation of justifying paying an exorbitant premium on process technology development to maintain that technology lead and cadence ad infinitum.

At that point the shareholders will respond much more favorably to any indications by Intel's management that future EPS will be less encumbered with aggressive R&D charges. It reaches a point (mathematically around 20% of revenue) where R&D budgets become a sore point with shareholders and instead of being seen as an advantage it becomes an economically unjustified liability (not the technology, am speaking about the R&D budget needed to sustain the technology lead).

At Texas Instruments we critically monitored R&D budgets as a function of revenue and benchmarking to our peers...when we were running nearly 24% revenue for R&D budget it was made known to us in no uncertain terms by analysts that this was viewed as a negative and was resulting in depressing our stock price, something management responded to in part with their decision to take TI fabless for 45nm and beyond in CMOS (analog manufacturing is still done inhouse, so not fabless like AMD or Moto).

FWIW the doom-and-gloom scenario you postulate in your OP is the progenitor of the equally absurd 2012 comments...if you make silly comments in your OP you will invite silly comments to be made throughout the thread.

Originally posted by: endofIT
I doubt that 16nm can even come out due to quantum tunneling.

Originally posted by: endofIT
I beleive that the continuous artifical stimuls to the economy that the IT industy provides will stop by 2013\2014 and then there would be no more advancement.

Originally posted by: endofIT
but the bottom line it that there will only be 2 more fabs aka 22nm and 16 nm and then thats it.

Originally posted by: endofIT
I beleive that in 5 yrs the entire IT industry will collapse.

Originally posted by: endofIT
so the 2014 stock market crash would be even more bad then the 2008.

These kinds of comments will make it hard for the serious folks around here to take you seriously. I'm not saying that to be rude or mean, I'm trying to help you understand why your thread is eliciting the kind of responses it has thus far.

 

[endofIT]

fine but its my view.I asked your views.

 

[imported_Lothar]

I put your "end of IT" scenario about on par to the Y2K bug.
Much ado about nothing.

 

[Idontcare]

Originally posted by: endofIT
fine but its my view.I asked your views.

Well thanks for confirming my suspicions then. You just have yourself a super nice day now.

 

[alyarb]

insane thread. OP in middle school? "IT companies in panic" haha

 

[Soundmanred]

You might look into baming the "end of IT" on Obama, or maybe terrorists. Only then will people take you seriously.

 

[alyarb]

well the miniaturization of architectures is such that, once our litho methods get so small, the whole chip will disappear! how can we plug it into our boards if it's so small we can't even see it! dell/hp/ etc won't buy chips they cant see. so they'll stop making computers!

that's when the stock market collapses. it will be EVEN MORE BAD than 2008! also, i heard this buzz word called quantum tunneling and i thought i would drop it a couple times. seems to be an issue with nano stuff. you guys follow? that's my view. whats your view?

 

[Harvey]

Originally posted by: endofIT

Topic Title: end of moores law.IT companies in panic?
.
.
in 2003 it was thought that 16nm will come out in 2018 but now it will come out in 2013 so its over in 5 yrs.so the 2014 stock market crash would be even more bad then the 2008.

Remember the ancient Chinese proverb:

"Those who say it can't be done should get out of the way of those doing it."

Process reaches beyond the limits on Moore?s Law

From Electronic Products, September, 2009:

Attaching molecules to the surface of silicon could be a boon to electronics manufacturers

A team of researchers from Rice University and North Carolina State University have announced a new process of attaching molecules to semiconducting silicon. This method could help electronic manufacturers reach beyond the current limits on Moore?s Law as they promise to make smaller and more powerful microprocessors.

Moore?s Law states that the number of transistors that can be placed on an integrated circuit doubles about every two years. As electronic manufacturers pack more transistors onto integrated circuits by making the circuits even smaller, the problem becomes doping ? a process that has been essential to creating the silicon substrate that is at the core of today?s integrated circuits.

So the team?s challenge was to get past the limits of doping. Doping introduces impurities into pure crystalline silicon as a way of tuning microscopic circuits to a particular need, and it?s been effective so far even in concentrations as small as one atom of boron, arsenic, or phosphorus per 100 million of silicon.

The team suggests that by attaching molecules to the surface of the silicon (rather than mixing them in) affects the threshold voltage, or gate voltage, required to create a conductive path between the source and drain electrodes (blue) and turn the device on. The molecules influence the amount of charge carriers available within the device layer (red). This method serves the same function as doping, but works better at the nanometer scale.

James Tour, Rice?s Chao Professor of Chemistry, anticipates a wide industry interest in the process, in which carbon molecules could be bonded with silicon either through a chemical bath or evaporation.

The paper on this research can be found at http://pubs.acs.org/doi/abs/10.1021/ja9002537. For additional information, contact Mike Williams of Rice University at 713-348-6728 or e-mail mikewilliams@rice.edu.

Christina Nickolas

 

[cusideabelincoln]

Originally posted by: Harvey

Originally posted by: endofIT

Topic Title: end of moores law.IT companies in panic?
.
.
in 2003 it was thought that 16nm will come out in 2018 but now it will come out in 2013 so its over in 5 yrs.so the 2014 stock market crash would be even more bad then the 2008.

Remember the ancient Chinese proverb:

"Those who say it can't be done should get out of the way of those doing it"

My sentiments, exactly. We'll find some way to keep improving the "information age".

 

[TuxDave]

Originally posted by: endofIT
hi i a am new here.After the 2009 idf where intel demoed 22nm i wanted to ask how far intel and other semicon companies can push the miniaturization ?

after 22nm it is 16nm but even at 16 nm quantum tunneling takes palce

This may blow your mind but... quantum tunneling has been happening on the chips for a long time now. You're quoting 22nm and 16nm but do you have any idea how thin the gate oxide is? It's much much MUUUUCH thinner and electrons are tunneling through left and right through it. Thanks to high k, thick oxide transistors we managed to reduce (not eliminate) this on 45nm.

It's a guaranteed truth that Moore's Law will end. However the end is always up to debate and while you're bringing it up at th 16nm and 11nm nodes, people were talking about it way back at the 1um node. It will end, it's just not sure when.

 

[jlee]

Remember when people were saying that hard drives had reached their limit at 80Gb due to platter density, or somesuch?

Yeah.

 

[Phynaz]

Originally posted by: endofIT
I beleive that in 5 yrs the entire IT industry will collapse.

So very little of IT has anything at all to do with Moores Law that I think all will be fine. If for some reason processors stop becoming more powerful, the need to process information will not stop.

Or you could listen to Justin Rattner who thinks Moores Law has decades of life in it.

 

[Skott]

Even if Moore's Law did fizzle out would it really be bad? I'm thinking not. People would find other ways to improve things along other lines. IT collapsing? Nah. Alyarb is right this is a insane thread.

 

[Cogman]

Well, one things for sure, Moore's law WILL end at or before the 0.01 nm mark (width of an atom. Sorry, it just ain't gonna get smaller then that).

My moneys on the 1nm mark. I think they can get there.

 

[bfdd]

Originally posted by: JLee
Remember when people were saying that hard drives had reached their limit at 80Gb due to platter density, or somesuch?

Yeah.

are you telling me it didn't!?!? FFFUUUUUUUUU and I scoffed at all this terabyte talk!

 

[Harvey]

Originally posted by: Cogman

Well, one things for sure, Moore's law WILL end at or before the 0.01 nm mark (width of an atom. Sorry, it just ain't gonna get smaller then that).

My moneys on the 1nm mark. I think they can get there.

Of course. Physics and the rest of reality are like that. Gordon Moore postulated Moore's Law in 1965. Our understanding of the dimensions and scale of atomic physics has expanded since then, and in 2005, even Moore, himself said that Moore's Law is dead.

Moore's prediction may have been more metaphoric than realistic. OTOH, it's amazing how closely Moore's Law has tracked that prediction. Note the current sharp UP-tick. :Q

My point was that the OP's concluding statement was just so much unsupportable negativity about the creativity of human beings.

 

[alyarb]

he read a paragraph out of wikipedia and panicked. ooh, aah, so many difficulties @ 16nm. i can't imagine what it's like living on the brink of nervous breakdown because of something written by a wikipedia contributor.

if he had read the article on double patterning, likely written by a completely different wikipedia contributor, he'd find an article that implies intel could make 11nm chips today with current 193nm tools and techniques. it's all a matter of whether you want to believe something or not that makes it true, until the appropriate time a fact comes along and swats your truth out of the sky.

 

[Cogman]

Originally posted by: Harvey

Originally posted by: Cogman

Well, one things for sure, Moore's law WILL end at or before the 0.01 nm mark (width of an atom. Sorry, it just ain't gonna get smaller then that).

My moneys on the 1nm mark. I think they can get there.

Of course. Physics and the rest of reality are like that. Gordon Moore postulated Moore's Law in 1965. Our understanding of the dimensions and scale of atomic physics has expanded since then, and in 2005, even Moore, himself said that Moore's Law is dead.

Moore's prediction may have been more metaphoric than realistic. OTOH, it's amazing how closely Moore's Law has tracked that prediction. Note the current sharp UP-tick. :Q

My point was that the OP's concluding statement was just so much unsupportable negativity about the creativity of human beings.

most likely what will happen is we will get to the point where we just can't improve the x86 architecture any more, so intel will FINALLY decide to go with something more efficient and less clunky. We might go through another period of architecture flux in the future.