Intel Has 5 nm Processors in Sight

Status
Not open for further replies.

DRosencraft

Distinguished
Aug 26, 2011
743
0
19,010
Should be very interesting to see how this turns out. That will be some impressive tech. Better start saving now and hope they're right about not hitting any roadblocks, although they probably have no real functional idea otherwise they'd just make the effort to go straight there. Nothing wrong with the step-by-step approach though. Wish them luck in overcoming Moore's Law.
 

balister

Distinguished
Sep 6, 2006
403
0
18,790
Unfortuneately, Intel is about the hit the wall in how far they can go. 1 nm is pretty much the wall as that's about 5 atoms in width. Quantum effects start to take over once you get to that level and it is not as easily dealt with due to things like Heisenberg's Uncertainty Principle and how the Strong and Weak forces start being a much bigger factor.
 

ben850

Distinguished
Apr 19, 2009
325
0
18,780
LOL I'm still rocking a Phenom II x6 on my AM2+ setup.. Looks like my next upgrade could be an extremely huge leap in CPU technology :)
 
[citation][nom]balister[/nom]Unfortuneately, Intel is about the hit the wall in how far they can go. 1 nm is pretty much the wall as that's about 5 atoms in width. Quantum effects start to take over once you get to that level and it is not as easily dealt with due to things like Heisenberg's Uncertainty Principle and how the Strong and Weak forces start being a much bigger factor.[/citation]
Agreed, quantum computing is where I'd be investing my R&D right now. I think they will have problems before 2019.
 
Six to eight years from now is not really "in sight" in this industry. Anything and everything can change in less than half of that time as far as this information is concerned. Saying that this will happen by 2019 doesn't mean that it won't be delayed until around 2025 or something like that. I'm also not convinced that 14nm will even hit the market for Intel until 2014 given that it would mean that Intel has about a year to get both Haswell and the 14nm die shrink of it (Broadwell if I remember correctly) and the chances of Haswell having a full line-up by the end of 2013 aren't even looking very good given that Ivy only got CPUs from all families Celeron and up a few days ago.
 
  • Like
Reactions: Exploding PSU

TeraMedia

Distinguished
Jan 26, 2006
904
1
18,990
@balister:

The problem is worse than that. If you consider that the typical doping level for an n-channel or a p-channel is on the order of 1:1000 atoms, then that means you need to actually have 1000 atoms for one of them to be doped (replacing Si with some other atom in the lattice). So a structure that is only 25 atoms wide by 25 atoms long is only 625 atoms per atomic layer. I remember reading somewhere that they need a minimum number of doped atoms for there to be any actual electrical effect, but I don't remember what that number was.
 

jkflipflop98

Distinguished
[citation][nom]balister[/nom]Unfortuneately, Intel is about the hit the wall in how far they can go. 1 nm is pretty much the wall as that's about 5 atoms in width. Quantum effects start to take over once you get to that level and it is not as easily dealt with due to things like Heisenberg's Uncertainty Principle and how the Strong and Weak forces start being a much bigger factor.[/citation]

Well good thing YOU were here to point that out!!! I'm sure none of the engineers and scientists at Intel have been poring over this exact problem for the last 10 years or so. Jesus P. Christ on a water buffalo I better got those dudes on the horn and share your revelation!
 

GabZDK

Honorable
Aug 12, 2012
44
0
10,540
[citation][nom]gigabyter64[/nom]If I remember correctly, wasn't Terminators based on 12nm chips, we are just about there!![/citation]

Yeah well, we better start loonking for a John Connor, we have like 5 years from now
 
[citation][nom]jkflipflop98[/nom]Well good thing YOU were here to point that out!!! I'm sure none of the engineers and scientists at Intel have been poring over this exact problem for the last 10 years or so. Jesus P. Christ on a water buffalo I better got those dudes on the horn and share your revelation![/citation]

There are these things called laws of physics and unless they are wrong, engineers have to deal with them. Getting to extremely small scales can change how things interact, especially concerning electrical properties. Engineers can't change how a material behaves in a given situation, so unless they make some sort of breakthrough, approaching sizes such as 5nm and lower can be significant problems for some things. For example, we've been working on space travel for over fifty years, yet we still don't (at least not officially) have light speed travel, let alone faster than light travel. Engineers aren't magic. If there is no solution to a problem or at least no solution that works with current technology, then they can't solve that problem, period. That's probably what balister was talking about, whether or not Intel has managed to overcome the difficulties involved in continuing to shrink the transistors and the distance between them.
 

darkavenger123

Distinguished
Feb 22, 2010
353
0
18,780
[citation][nom]balister[/nom]Unfortuneately, Intel is about the hit the wall in how far they can go. 1 nm is pretty much the wall as that's about 5 atoms in width. Quantum effects start to take over once you get to that level and it is not as easily dealt with due to things like Heisenberg's Uncertainty Principle and how the Strong and Weak forces start being a much bigger factor.[/citation]

I am sure nobody in Intel has thought of this! You should go apply a job at Intel! I am sure they'll make you CEO!
 

jkflipflop98

Distinguished
[citation][nom]blazorthon[/nom]There are these things called laws of physics and unless they are wrong, engineers have to deal with them. Getting to extremely small scales can change how things interact, especially concerning electrical properties. Engineers can't change how a material behaves in a given situation, so unless they make some sort of breakthrough, approaching sizes such as 5nm and lower can be significant problems for some things. For example, we've been working on space travel for over fifty years, yet we still don't (at least not officially) have light speed travel, let alone faster than light travel. Engineers aren't magic. If there is no solution to a problem or at least no solution that works with current technology, then they can't solve that problem, period. That's probably what balister was talking about, whether or not Intel has managed to overcome the difficulties involved in continuing to shrink the transistors and the distance between them.[/citation]

LOL it's a LAW OF PHYSICS. . . unless you can make a breakthrough? LOL.
 

Gulli

Distinguished
Sep 26, 2008
1,495
0
19,310
[citation][nom]TeraMedia[/nom]@balister:The problem is worse than that. If you consider that the typical doping level for an n-channel or a p-channel is on the order of 1:1000 atoms, then that means you need to actually have 1000 atoms for one of them to be doped (replacing Si with some other atom in the lattice). So a structure that is only 25 atoms wide by 25 atoms long is only 625 atoms per atomic layer. I remember reading somewhere that they need a minimum number of doped atoms for there to be any actual electrical effect, but I don't remember what that number was.[/citation]

They'll replace silicon with carbon at some point and then they can build 1nm transistors while at the same time drastically increasing clockspeeds. Then computer hardware will have reached a dead end (it will be many times faster than today's hardware however, probably fast enough to support the same level of intelligence as the human brain), unless quantum computing turns out to be feasible one day.
 
Well there have been functioning examples of single atom transistors since the 90s but never commercialized (hints at the tech that is still hidden in private and government labs) but 5nm is good. Can't wait for the tech to hit the market.
 
[citation][nom]darkavenger123[/nom]I am sure nobody in Intel has thought of this! You should go apply a job at Intel! I am sure they'll make you CEO![/citation]

Thinking about something and actually being able to do anything about it with the technology are two very different things. Intel will probably have to use a different type of technology because even if Intel manages to use make transistors out of individual atoms and having them placed less than 1nm apart, they would be pretty much entirely incapable of going any further in this scale. Whether or not there are other things that can then be done instead of shrinking the scale is not what balister was talking about. Also, engineers generally aren't CEOs. That's why they're engineers instead of business men.
 
This just has to do with the transistor width, not the actual size of the chip itself. To avoid some of the nastier QM effects they'll have to pad in extra space between the transistors, that will prevent the chips from getting "smaller" though they'll still use less power and run faster. At ~15nm is where QM starts effecting signaling integrity, ECC inside the CPU becomes an absolute necessity.
 
[citation][nom]jkflipflop98[/nom]LOL it's a LAW OF PHYSICS. . . unless you can make a breakthrough? LOL.[/citation]

... Make a breakthrough as in a technology that isn't bothered by the same laws of physics as much (IE such as quantom computing), not somehow circumventing them like magic, which is impossible unless they are wrong, in which case, as palladin9479 said, that'd be a job for scientists to fix first and for engineers to implement later. Seriously, try putting some actual thought into what you're saying and discrepancies such as this misunderstanding of your shouldn't happen, at least not nearly as often.
 
[citation][nom]jkflipflop98[/nom]LOL it's a LAW OF PHYSICS. . . unless you can make a breakthrough? LOL.[/citation]

Physics are rather immutable. Thing is, we don't understand all of them yet so there may very well be some special case scenario around QM. That is a problem for the scientists, then after that the engineers take that documented science and build an engineering solution out of it. That all takes time, lots of it, we've been understanding the physics around fusion for decades yet the engineers still can't make a viable fusion reactor.

So far the EE's working at CPU manufacturers have been working inside known physics, all their problems have been engineering and efficiency problems not science problems. Very soon they'll have exhausted the amount of knowledge regarding controlling electrical signals through a medium, then it becomes a science problem. I have no doubt that it'll eventually get solved, it just may not be in a way we expect.
 
G

Guest

Guest
There were reports about "Moore's Law" going to end in 5 - 10 years when I got my Everex 386/20Mhz computer (according to pc magazine, it was the fastest personal computer for 3 months) back in 1988. The techs and scientists keep finding solutions to the problems that were claimed to be sure to kill Moore's observation.
 
Status
Not open for further replies.