Comments
MIT Research Looks To Extend Moore’s Law
Newest First  |  Oldest First  |  Threaded View
melgross
50%
50%
melgross,
User Rank: Ninja
3/2/2014 | 8:03:02 PM
Re: Does Moore's law matter to end users?
A one nm line is about one or two atoms wide, an impossible task! Getting down to 10nm, will be extremely difficult. Even when the chips are made, efficiencies could be daunting. After a while, we run out of those clever workarounds. The laws of physics are like a thick steel wall that we're banging upon with a paper hammer.
Li Tan
50%
50%
Li Tan,
User Rank: Ninja
3/1/2014 | 8:31:14 PM
Re: Does Moore's law matter to end users?
I have no good vision to predict when Moore's law will reach its limitation. But for sure such kind of limit exists - the hardware producer can produce 20nm ICs but how about 2nm, 1nm, even .x nm? Trying to achieve the limit of Moore's law may not mean too much for end users. One day when the limit is achieved, the manufacture needs to think about other ways to improve the speed of computing.
Gary_EL
50%
50%
Gary_EL,
User Rank: Ninja
3/1/2014 | 12:54:45 PM
Re: Does Moore's law matter to end users?
Perhaps not for mobile and desktop PC users, but there are surely computationally intensive processes, where a lot happens in one place before anyhing gets placed on an external network. But, as melgross says, only so many atoms can line up in a given length, the limit is being reached, so I believe Moore's Law, too, will soon be reaching its limit.

 
melgross
50%
50%
melgross,
User Rank: Ninja
3/1/2014 | 2:02:19 AM
Re: Does Moore's law matter to end users?
The article is interesting but, it covers only half of the problem. So while it's true that manufacturing chips under 14nm will be difficult, and going below 7 may be impossible, the other problem is more important because the laws of physics are more important there. What is that problem? It's that there aren't very many atoms in the width of a line at these small process sizes. By the time we get below, we're talking of just 10 to 20 atom wide line features. It becomes almost impossible to keep electrons withing that line. Starting with 90nm we began to see the problem with Intel,s Netburst. Clever workarounds have been used since then to counter leakage, and other problems resulting from Heisenberg's Law of Uncertainty. Electrons tunnel out of a conductor. Larger conductors have this problem to a very small extent, but it begins to become significant as the surface become a large percentage of the total area of the conductor. At some point, it will be impossible to keep those electrons where they belong, and progress will have ended. Will people care, you ask? Certainly! We have been not only become accustomed to continued progress, but have become habituated to it. One might even say, addicted. When that progress begins to slow, and it has already, as Intel's last few generations have all come in behind schedule, it will become harder to sell new products. When it stops, what will happen then? We're waiting for new technologies to replace the silicon technology we've become accustomed to using. But alternatives aren't just around the corner, they're more than a decade away, and possibly a lot further than that, if possible at all. I expect more clever methods will allow better chips on the same processes, but large improvements depend on smaller processes every two years or so.. When that moves to three, then four, five and then no more smaller processes, only small improvements will be possible without an complete rethinking of computer processes. Whether that will be possible isn't known. Why but a new device if there are no major improvements? This would change everything. Network speeds aren't everything. They are only one part. We still do most of our computing on our devices, not over the network. It's not a world of dumb terminals.
Thomas Claburn
50%
50%
Thomas Claburn,
User Rank: Author
2/28/2014 | 5:08:10 PM
Does Moore's law matter to end users?
The main bottle neck for mobile and desktop PC users these days seems to be network speed rather than processor speed. Is there still much to gain from pushing processor semiconductor density?


The Business of Going Digital
The Business of Going Digital
Digital business isn't about changing code; it's about changing what legacy sales, distribution, customer service, and product groups do in the new digital age. It's about bringing big data analytics, mobile, social, marketing automation, cloud computing, and the app economy together to launch new products and services. We're seeing new titles in this digital revolution, new responsibilities, new business models, and major shifts in technology spending.
Register for InformationWeek Newsletters
White Papers
Current Issue
InformationWeek Tech Digest - August 20, 2014
CIOs need people who know the ins and outs of cloud software stacks and security, and, most of all, can break through cultural resistance.
Flash Poll
Video
Slideshows
Twitter Feed
InformationWeek Radio
Sponsored Live Streaming Video
Everything You've Been Told About Mobility Is Wrong
Attend this video symposium with Sean Wisdom, Global Director of Mobility Solutions, and learn about how you can harness powerful new products to mobilize your business potential.