If Moore's Law hits a sinkhole...

[tw]

Previously posted was that Moore's Law for computers may be about to hit a brick wall. Fundamental to the problem is that transistors have shrunk to about as small as atomic physics permits. On 15 Aug 2005, EE Times wrote this front page story:

Pity the poor MOSFET. Once the star of microelectronics - the ideal blend of elegant simplicity that made the global industry possible - the planar metal-oxide-semiconductor field-effect transistor appears to be approaching the end of its useful life. It leaks where it shouldn't. It's plugged up where it should be open. And repeated attempts to keep it functioning have left the vital parts dangerously thin.

But if the planar MOSFET is in trouble, what comes next? Some experts say there really is no replacement - that new materials, more advanced fabrication techniques, maybe a new substrate will keep the little guy going well into the next decade. Others believe only a move to a radically different, three dimensional structure can keep device scaling on track much beyond the 45 nanometer chip manufacturing mode ...

High-k dielectric materials were suppose to ride to the rescue about now, improving the electric-field strength in the channel and thus reducing leakage for a given dielectric thickness ...

But both high-k and metal gate technologies have proven far from easy to integrate. "This path calls for new, hostile materials to be introduced into the process flow, each with its own fundamental problems.

This materials include Zirconium oxides, Hafnium oxides, or Lanthanum oxides. Not commonly known materials because many are rare as well as little understood.

The upcoming Intel processors will soon be doing 45 nanometer transistors. As the EE Times article points out, simulations for a 32 nanometer transistor result in massive leakage problems. You have already seen how bad the leaking is. Hold you finger on a Pentium processor without a heatsink. No, the Pentium will not burn up as so many claim. But your finger will. Transistors are already so thin that Pentiums can produce more heat than a 100 watt light bulb.

Intel recently announced a fundamental change in design philosophy. No big surprise. Everyone was predicting it. Traditionally, we put all critical computer functions in one chip so that communication between those functions is fastest. Once we send a signal between two ICs, the message gets really slow. However Intel must now separate CPU functions - because transistors are leaking and therefore consuming so much electricity. The trick is to separate CPU functions that need not communicate so fast. That literally means changing the entire CPU architecture. Last time a change this radical was performed, it was called the original Pentium.

Intel has also set new corporate limits on how much electricity a chip can consume. IOW CPU speed is no longer the mantra. Once it was all about doing the same computations only faster. Trying to maximize CPU processing verses minimal electric consumption is a new corporate wide directive. All this to avoid the brick wall.

I am reminded of the last scenes in James Bond's Thunderball where the hydrofoil is desperately swerving to avoid rock outcroppings. Eventually, a crash into rocks as hard as bricks. In the CPU business, are we about to witness the same end? And who will get the girl? Oh. Back then James never got the girl. He was rescued. And then he was replaced by Roger Moore. Curious coincidence.

[Undertoad]

They also want to save power because notebooks are the boom business now, not laptops.

And if Moore's law does finally hit the wall, we will see more multi-CPU solutions, so each chip better take less power. It will be difficult to justify 1000 watt desktops with 4 processors in order to run Office 13.

I don't think computing power is driving any innovation right now because all the desktops are faster than we need already. Desktops are about 10 times more powerful than they have to be in order to run Windows, Office, a browser, and full motion video all at the same time. A minority of people do the things that require a really powerful system: Battlefield 2, Unreal Tournament 2K4, video editing and massive photo manipulation.

Even when DOS was prominent and programmers had to do weird and interesting things to get access to more memory, there were more applications pushing the edge. Nowadays, the cheapest system will run Windows and Office and everything they need to do, with practically no difference from the most expensive system.

[richlevy]

Originally Posted by Undertoad

A minority of people do the things that require a really powerful system: Battlefield 2, Unreal Tournament 2K4

So in other words, the really important stuff.:p

I'm only half joking here. A huge amount of innovation is driven by entertainment. It has been argued that porn provided the early impetus to the VCR industry. More effort has been expended on developing, marketing, and getting health coverage for using Viagra than on many cancer medications. More Hummers were sold to soccer moms that the US military ever bought.

Collectively speaking, 350 million people in the US (or a few billion worldwide) looking for fun have more economic clout than military, government, or even business. If there's a solution there it will be found. One possibility is moving from electricity to light as in photonic computing. Another is by building a ternary computer instead of a binary computer, which would be more efficient. To do this might require a combination of materials to allow for three states instead of two.

[Elspode]

I'm headed out to MicroCenter to buy my quantum computer this morning. Soon, I'll be posting WTF photos from a kick ass machine! I sure hope they don't force me to buy it with Vista on it, though.

[mbpark]

TW,

How does it feel to be proven right, and that smack-talker Radar proven wrong?

Mitch

[richlevy]

They are already looking into optics for data storage. They might do the same for CPU's.

[Radar]

Gaming is what drives innovation in the computer industry and that won't change. The motherboard I'm buying has 2 PCI Extreme Ports to add 2 high end video cards in SLi mode. This means one video card will render the current frame, the other will render the next frame, allowing an ungodly amount of frames per second and amazing performance.

I wouldn't bet on Moore's law failing. Right now Intel, AMD, and others researchers and universities are working on Quantum computer chips. They will be an order of magnitude faster than our current chips. These chips are right around the corner.

I've actually heard the chips are already done, but the government won't allow them to be released until encryption methods are improved. The new chips would eat right through current encryption. It would normally take 30+ years even with supercomputers to break 1024 bit encryption. A Quantum chip could do it in minutes.

I'm sure if the chips are available already, the government is using them. I know the U.S. government recently purchased the world's largest solid state storage chip. It holds more than 400 terabytes. This could hold an operating system which could run anything almost instantly. Couple this with the Quantum chip and you've got a machine that would allow the government to get into anyone's computer at anytime.

[tw]

Originally Posted by Radar

I wouldn't bet on Moore's law failing. Right now Intel, AMD, and others researchers and universities are working on Quantum computer chips. They will be an order of magnitude faster than our current chips. These chips are right around the corner.

Yeaah. And oil companies are still keeping the 1000 MPG carburetor off the market. And US nuclear subs are actually stored in secret caves beneath NYCs bedrock.

It helps to first learn of reality before posting. Quantum computers are but fragile advance science toys that, last I had heard were up to 4 bits. It takes something like 8 years to double those bits. And still they are not even stable enough for a controlled computer room. Meanwhile if encryption security was a reason for quashing quantum computers, well, quantum encryption is currently being tested in a network in MA. Quantum encryption is already far ahead of quantum computing.

Quantum computer research is not ongoing in 'application' research labs such as AMD and Intel. It is still in 'basic' research labs such as University of CA (Santa Barbara), National Institute for Standards and Technology, CERN in Switzerland, Paul Drude Institute in Berlin Germany, etc.

If you know about this 400 Terabyte memory chip, then share details such as who developed it, where, using what technology, manufactured in what quantities, etc? Why not? Radar ... you have again done exactly as Rush Limbaugh types would do. Hype a lie in hopes that the naive will believe propaganda. Warning message to the naive. Radar is lying because he provides no underlying facts, no supporting numbers, and contradicts 'state of the art' science. Most damning - he provides no details hoping that is enough for you to 'know' something.

Optics for data storage, as Rich Levy notes, was suppose to first appear on the market this year from IBM. I suspected that was why IBM sold their hard drive division to Hitachi. Optics for CPUs has long been an objective to eliminate those power hungry data busses inside a CPU. But currently no one has made a successful or useful optical switching transistor. As EE Times noted, these would be breakthrough technologies that could replace the FET transistor. Unfortunately, as EE Times said, "there really is no replacement".

[Radar]

I shouldn't be surprised by the fact that tw has a woefully inadequate education with technology. Over the years, he's proven himself stupid in a number of areas. But what's really funny is that this idiot actually has the gall to accuse me of being dishonest. Tw isn't fit to stand in my shadow when it comes to computer technology, and for that matter, pretty much everyone else on this board is in that boat. It's not an insult, it's just the truth. I'm one of the best all around network engineers in the business and that's not bragging.

I've never lied about anything I've ever said here, or anywhere else on the internet. That is a fact.

I remembered reading the article about the Government buying the world's largest RAM disk a long time ago. After doing some research when being challenged by tw (as if he's a challenge for anyone), I found that the government did indeed buy the world's largest RAM disk and that they intend to use it in ways that I thought, but that it's not 400 Terabytes. It's only 2.5 Terabytes per unit. I remembered the 400 Terabyte number for a reason. I'm guessing it's because they ordered enough of the units to equal 400 TBs. Either way, I admit when I'm wrong and being wrong is not the same thing as being dishonest. It's been something like a year since I read the article and even then I was glancing through it while at an airport.

http://www.techworld.com/storage/new...fm?NewsID=1176

http://www.atsnn.com/story/37239.html

http://www.physorg.com/news6149.html



Now let's move onto tw's next retarded contention. He stupidly claims that Quantum computers are a long way off. I guess for a child like him 3 years is a long time.

http://www.engadget.com/entry/1234000933047853/

http://physicsweb.org/articles/world/15/4/4/1

http://www.physorg.com/news6149.html

Here's something about Quantum Computer Memory...

http://www.eurekalert.org/pub_releas...-nst100604.php


tw did accidentally get something right. Most of the research for Quantum computers is being done in universities. Who is paying for this research? Large computer companies like IBM, Intel, and AMD.

It must suck being such a bitter, sad, pathetic, little jackass like tw. What a horrible life he must live. Still, my pity of his stupidity and dishonesty wouldn't stop me from knocking the shit out of him in person if he ever called me a liar. It's a good thing we have an internet so he can talk shit and act brave. He can't do it anywhere else.

[Radar]

Here's are most of the places around the world doing research on Quantum Computing.

http://www.qubit.org/phpscripts/places.php

[Undertoad]

You talked at first about a 400 TB chip but revised it to a 1.5 TB solid state ram disk when you were questionned.

From a practical and engineering standpoint, we can't really tell the difference between lying, being poorly trained, making bad assumptions, reading too fast, having a personality problem that causes you to insist you have the high ground, or simply being mistaken. All an engineer would know is that you were wrong, and in the real world it probably cost somebody money and/or time.

[Radar]

No, I talked about a 400 TB chip, which is how I remembered it after having barely glanced at an article a year ago. I admitted to being wrong about that. And did state the correction. Also, it's 2.5 TB of a RAM Disk, which is actually faster than most solid state chips such as USB storage.

In either case, my point about the U.S. government buying the world's largest high speed storage to be used against Americans was correct and so was what I said about Quantum computing. In otherwords, I was correct about it being highly unlikely that Moore's law would fail anytime soon.

An engineer (I am one) would not assume dishonesty, even if one did find someone else to be mistaken. They would only bring up the mistake.

[dar512]

Just for the record, Radar - are you predicting that quantum computers will be available in retail for the desktop in 3 years?

[classicman]

Originally Posted by Radar

An engineer (I am one) would not assume dishonesty, even if one did find someone else to be mistaken. They would only bring up the mistake.

No an engineer would only be interested in facts, not opinions - Period.

[tw]

Originally Posted by classicman

No an engineer would only be interested in facts, not opinions - Period.

Where does classicman address the topic of quantum computers or D-wave? Where does he offer using information or insight? Oh. classicman instead attacks the messenger. Again, classicman has nothing useful to offer - just posts belittling remarks.