Moore's Law to Hit a Brick Wall at 18nm



+ Add a Comment


Moores Law will Prevail.

There are ways to work around and solve the smallest semiconductor barrier.

We have a Processor that we are completeing and testing, the MegaMegaProcessor.

It runs between 20GHz - 100GHz.

We are also building Desktop and Laptop PC prototypes now with these Processors,

that run from 100GHz - 1THz (1000GHz).

- Ready for the PC SPEED ?!


Topic Site:


Company: JLC Robotics



Appreciate the info, it’s good to know

charter of human rights



It's a good thing that Intel is doing research on Silicon Photonics or laser based CPUs'.



It's kind of funny that today this article is released saying that moores law will soon come to an end due to manufacturing costs/limitations below 18nm when yesterday you posted an article about how toshiba has made a breakthrough and we will be seeing 16nm ahead of market estimates.



Toshiba is doing that for flash memory, which are substantially less complex circuits than CPUs, IIRC, and they also don't need to run nearly as fast.  the faster you go, the more power CMOS eats and the more heat you need to deal with.



You are right, I overlooked the fact that the previous article was in reference to flash memory. But it just goes to show that noone can predict what kind of manufacturing breakthroughs we are going to have in the next 10 years.



It has always killed me that this is revered as Moore's "Law"...  It should really be Moore's Hypothesis Which Just Happens To Have Been Approximately Right Thus Far.  Or whatever.  The point is that there's no real scientific reason the hypothesis *should* be true...  It's more a self-fulfilling prophesy than it is anything else.


Anyway, I agree with the other comments pointing out how it's impossible to know what could be the case in five years.  You might as well just go polish your crystal ball if you want to predict transistor sizes that far into the future. 

EDIT: It seems that link below to the Halfhill article goes on that rant for me. 



how many times have we heard this?  Seems like once a year we hear that Moore's law will end when X happens, and then it happens and they're like "oh, well we found a way to get around that, now Moore's law will not apply when Y happens..."


Mosher previous article seems to show that by 2015 we will have much smaller capacitys than that.(unless I am not understanding how the memristor works prperly)



That's still five years away from now. Who knows what crazy stuff will be invented by then.



What would you have done in mid 2004 if someone said that people would be able to buy quad core, 45 nm computers with videocards that have power measured in terraflops, storage measured in Terrabytes, etc.



didnt people think it was going to get debunked before?  But has not.  Just never know till it happens.



Huh? I thought you guys debunked Moore's Law. Or Tom Halfhill did it for you. Either way, it's been broken for a while.



Try reading the article you linked to. 

Moore's law describes component integration on integrated circuits that are economical to manufacture.  It doesn't describe clock frequency or other aspects of processor performance. Actually, Moore's article doesn't mention microprocessors at all, because they weren't invented until six years after his article appeared.

Which what this little article is about, anything smaller than 18nm is not economical to manufacture.  Not about the speed or anything.



I was actually referring to the debunking of its truth:

Myth: Moore's law is uncannily accurate. Fact: Moore's law is way off. If the original 12-month period had held true since 1965, today's chips would have more than 27 trillion transistors. The 24-month revision predicts 37 million. Actual progress: Intel's Prescott Pentium 4, a relatively economical chip, has 129 million transistors. 

 But very nice of you to call me an "R-tard" without bothering to understand it yourself. 

Log in to MaximumPC directly or log in using Facebook

Forgot your username or password?
Click here for help.

Login with Facebook
Log in using Facebook to share comments and articles easily with your Facebook feed.