Overclock.net banner

1 - 20 of 31 Posts

·
Registered
Joined
·
883 Posts
Discussion Starter #1
Or so says Intel. After that, we may see one or two more, but thats it.

Article is a little old, but a good read.

Quote:
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.
http://news.com.com/2100-1008-5112061.html
 

·
Registered
Joined
·
193 Posts
What do we do after 16 nm then? Overclock the hell out of them?
 

·
Registered
Joined
·
827 Posts
I was at an intel seminar 5 years ago when they said this. They where prototyping 25nm at the time and explained that this phenomenon was already present at the 25nm process. Silicon is an old material to be using and electron pushing should of gone out years ago since we've had prototype photon CPUs since the turn of the century. I'd say Intel has a rough quantum processor mockup already, and they are just slowly releasing processors into the market to extend their usuable life (although atm they seem more set on eviscerating AMD with the amount of releases they've made recently) By 2050 i'd say we'll be using 16bit Quantum PCs and they'll be powerfull enough to finally create the transporters from star trek
 

·
404 Title Not Found
Joined
·
2,328 Posts
After 16nm they'll just have to make more and more cores. I'd imagine there would be no reason to OC by then but we'll do it anyway.
 

·
Registered
Joined
·
2,338 Posts
Quote:

Originally Posted by accskyman View Post
After 16nm they'll just have to make more and more cores. I'd imagine there would be no reason to OC by then but we'll do it anyway.
Heck yea, we'd still do it! Imagine the TDP of those suckers!
 

·
Banned
Joined
·
1,195 Posts
Quote:

Originally Posted by Burn View Post
Yeah, I wonder what the next step in computing will be, after 16nm comes out...
Theoretically, 11nm, but they've pushed silicon way too far. It's a miracle that 45nm is possible, so they seriously need to consider new technologies. Quantum processing....likely.
 

·
Premium Member
Joined
·
5,282 Posts
From what I've read quantum computing won't show performance gains for the kinds of tasks we want CPU's to perform. Mainly they are for optimization and things that are too complex for traditional algorithms - the example the article used was optimizing mail routes. Once you get over 20 or so houses there are too many possibilities for modern algorithms/CPU's to handle. A quantum comp would naturally form the best route almost instantly. But for gaming and the like there are few benefits from quantum computing.
 

·
Registered
Joined
·
2,338 Posts
God said "Let there be Battlefield 9142"

and it was buggy
 

·
Premium Member
Joined
·
7,397 Posts
Quote:

Originally Posted by Sideburns View Post
light.
We'll be doing carbon nanotubes before light photons.
 

·
Registered
Joined
·
827 Posts
Quote:

Originally Posted by Burn View Post
Well, if we were to go to quantum processing, would silicon still be a viable substrate?
No, because quantum processors inherently use photons instead of electrons to transport data, due to certain properties of light (watch Dr. Quantum on Youtube for a quick headsup), it would have to be some form of optical medium.

EDIT: Sorry I should of said, electron particles do exibits the same characteristics as photons, it's just, lets say, easier to harness the properties of photons for their quantum characteristics.

Oh and there is only so far they can go with multiple cores as they are bandwidth limited by the buses. Plus they really need to start pouring more money into actual software development atm before they concern themselves with getting better CPUs.
 

·
Premium Member
Joined
·
4,807 Posts
By that time, they'll probably get battling for half-nodes so I think they'll be able to force a 13nm or so. At that process, they have SO much real estate on the silicon to play with. I can't even begin to imagine what they could do to fill that up. We can only use so many cores, stream processors, and L2 cache.... Either that or the chips will be so small that even an IHS would crush it.
 

·
Registered
Joined
·
939 Posts
Nanotubes work really well over long distances, but the traditional transistor doesn't work with them.

Light, on the other hand, has the useful property of wave interference, so it would be possible to do AND, NOR, NAND, OR, XOR, NOT gates based on placing obstacles to change the wavelength temporarily.
 

·
Premium Member
Joined
·
4,746 Posts
What does quantum processing entail?
 

·
Registered
Joined
·
1,769 Posts
silicon has really come to its limits. They sure have milked alot out of it though.
 
1 - 20 of 31 Posts
Top