3D Computing Is On the Way
One of the worst parts of this job is having to admit the occasional mistake. A writer’s slip-up can be seen by thousands. And today, we must start with such a correction. It wasn’t my own mistake, but rather Doug Hornig’s on March 22nd.
This error wasn’t something Doug could have possibly foreseen, but nonetheless, we have to correct his statement. In the issue, Doug noted:
I remember my colleague David Galland once opening a Casey conference by saying: “So who is Lindsay Lohan, anyway, and why do we care how much she drinks?” Good question. I’m sure, given the intellectual quality of our readership, your answer would be, “We don’t.” But far more Americans are fascinated by her trials than have even heard of the supremely important one down in North Carolina. And that’s sad. Because von NotHaus’s fate matters, to all of us, an awful lot more than whether LiLo is headed for jail or rehab.
Doug’s words made plenty of sense at the time. But a few days ago, this statement was suddenly turned on its head by Lindsay Lohan herself. On June 27th, she tweeted: “Have you guys seen food and gas prices lately? U.S. $ will soon be worthless if the Fed keeps printing money!”
Doug and I can write articles about the Fed until our fingers turn blue, but we’ll probably never reach the number of people she just did with this message. According to her Twitter profile, she has 2,132,759 followers. And the majority are probably thinking, “What’s the Fed?” and “What’s printing money?” for the first time in their lives.
In a way, this shows the power of the anti-Fed message. The fact that the Fed is destroying the U.S. dollar isn’t some grand mystery. It doesn’t take a Ph.D. in economics to understand it. Why, one can even be a Hollywood drug addict and get what’s happening.
Furthermore, it shows how bad things are really getting. If even the LiLo tabloid culture can’t avoid the realities of our economic condition, we are in serious trouble.
Before we get started, I want to let everyone know about another Doug Casey appearance. He’ll be speaking in Vancouver at the Capitalism & Morality Seminar organized by one of our former employees, Jayant Bhandari. Along with Doug, there will be a few other speakers of interest to our readers, including Rick Rule and Butler Shaffer.
Next, I’ll turn it over to Doug Hornig who will surely “redeem” himself with an article about 3D computer chips. Then, Andrey Dashkov will go off the beaten path and share his insights on mining opportunities in Kyrgyzstan.
By Doug Hornig
Somewhat surprisingly, computer architecture has always been two dimensional.
Chips are designed much like the wooden mazes of yore that rats used to run through in science experiments. Walls constrict the movement of the charge, and gates throughout the chip allow it to progress toward an exit. Nothing moves in three dimensions – there is no “up” inside a CPU.
Even when adding multiple cores to a processor – common practice nowadays, with the majority of chips now shipping with two or more cores on the same die – components have just been packed next to each other in the same plane.
That’s about to change.
First, 3D arrays for multiple-core processors are on the way.
Chips stacked in 3D are not merely as good as the same number of chips placed side by side: They’re better, way better. That’s because they’re not running independently, but are fully integrated across all of their functions. According to IBM researchers working in Switzerland, this dramatically shortens the distance information needs to travel on a chip to just 1/1000th of that on 2D chips, while allowing the addition of up to 100 times more channels (or pathways) along which that information may flow. The result: higher speed and far more power packed into a comparable space.
The arrival of 3D arrays has awaited the resolution of some technical problems. For one thing, these things get hot, and conventional air-cooling techniques don’t work. Developers had to construct a liquid, interlayer cooling system. According to IBM, it had to be “a system that maximizes the water flow through the layers, yet hermetically seals the interconnects to prevent water from causing electrical shorts. The complexity of such a system resembles that of a human brain, wherein millions of nerves and neurons for signal transmissions are intermixed but do not interfere with tens of thousands of blood vessels for cooling and energy supply, all within the same volume.”
That’s quite a challenge, but they succeeded with the structure pictured in the diagram above.
In addition, the designers had to work within an accuracy of ten microns – ten times more accurate than for interconnects and metallizations in current chips. That they have also accomplished.
Building a machine based on 3D architecture is now under way. And what a machine. In early March 2010, IBM agreed to cooperate on a four-year project – called CMOSAIC – with the École Polytechnique Federale de Lausanne and the Swiss Federal Institute of Technology Zurich (ETH).
These folks are not thinking small. From the project description: “The CMOSAIC project is a genuine opportunity to contribute to the realization of arguably the most complicated system that mankind has ever assembled: a 3-D stack of computer chips with a functionality per unit volume that nearly parallels the functional density of a human brain.”
Yet these techniques only involve taking the same basic chip designs and stacking them more closely together to be more efficient – like building a skyscraper to replace 40 one-story office buildings. Same use, less space, and some efficiency gained in wiring and HVAC.
However, in 2008 at the University of Rochester, an effort led by Professor Eby Friedman helped to create the first true 3D chip – or as they have dubbed it, the first processor “cube.”
The Rochester chip is not simply a number of regular processors stacked on top of one another. It was built specifically to optimize key processing functions vertically through multiple layers of silicone, the same way ordinary chips lay out logic horizontally along that maze.
Manufactured at the Massachusetts Institute of Technology, the chip has millions of small holes essentially drilled into the insulation that separates the layers. When the layers are assembled, millions of vertical connectors can be added in that insulation to connect the transistors on different levels.
For years, chip designers have postulated the eventual end of miniaturization, whereby their attempts to push more transistors in the same area will quickly come to a screeching halt. From there, the only way to go is up. However, until recently there have not been any truly successful models for a 3D chip that would be friendly to use with chip design software and could accommodate the different impedances and power requirements of the differing functions each layer would provide, or handle complex tasks like synchronicity.
Since each layer could be a different processor with a different function – from decoding digital audio and video to performing complex calculations – the 3D chip is essentially an entire circuit board folded up into a tiny package.
Using this technique, the group at Rochester claims the chips inside something like an iPod could be compacted to one-tenth of their current size yet run at ten times the current speed.
Mr. Friedman says, “Are we going to hit a point where we can’t scale integrated circuits any smaller? Horizontally, yes. But we’re going to start scaling vertically, and that will never end. At least not in my lifetime. Talk to my grandchildren about that.”
Unfortunately, 3D chips are still in the development stage. We’ll have a long wait before they get out of the lab and into the marketplace. But in the meantime, Intel has taken an interim step forward, by announcing on May 4, 2011 that it will soon begin high-volume manufacturing of chips featuring Tri-Gate, the world’s first three-dimensional transistors. According to the company, they will allow for the prolongation of Moore’s Law for at least another two years.
Call it a mini-revolution. And we do mean “mini.” These things are small. Transistors in Intel’s current-generation chips – the planar (2D) type that have been the standard since their invention in 1959 – are already only 32 nanometers (nm) in size. 3D Tri-Gates are only 22nm. (A nanometer is one-billionth of a meter. More than six million 22-nm Tri-Gate transistors would fit on the period at the end of this sentence.)
This image shows the vertical fins of Intel’s Tri-Gate transistors passing through the gates:
PCMag describes how it works (here’s a visual presentation by Mark Bohr):
The Tri-Gate technology gets its name from the fact that transistors using it have conducting channels that are formed on all three sides – two on each side, one across the top – of a tall and narrow silicon fin that rises vertically from the silicon substrate. On a traditional [planar] transistor, the gate runs just across the top. But on the vertical fin, transistors can be packed closer together. This provides enough extra control to allow more transistor current to flow when the transistor is on, almost zero when it is off, and gives the transistor the ability to switch quickly between the two states.
The Tri-Gate’s capabilities give chip designers the flexibility to choose transistors targeted for low power or high performance, depending on the application. They can provide up to a 37% performance increase at low voltage versus Intel’s 32-nm planar transistors. Or, when run at the same performance level, they consume less than half the power required by the 2D transistors. All this has an added manufacturing cost of just 2-3% for each wafer from which chips are cut.
Intel’s factories will have to be upgraded to produce the processors (to be called Ivy Bridge) based on the new tech. That’s no small task when you manufacture about five billion transistors per second, as Intel does. But the company says that the actual changes being implemented will not be more significant than have been required for previous process improvements.
We can expect to see processors and devices using Ivy Bridge technology by the end of 2011, with product shipping in early 2012.
By Andrey Dashkov
Louis James and I visited Kyrgyzstan last week to update our assessment of the place for investment purposes. It’s off the beaten track, but an important place, having once been one of the major mining suppliers in the Soviet era. Kyrgyzstan produced gold, silver, lead, coal, uranium, rare earths, and other minerals. The economic ties with the rest of the former Soviet Union largely fell apart in the early 1990s, and as of today the country is the second poorest in Central Asia.
Kyrgyzstan’s natural endowments have long been seen as one of the bullish factors for the local economy. However, legislative instability and corruption still remain as major obstacles to foreign investment in the country.
A recent example is Kentor Gold, an Australian company that owns the Andash copper-gold deposit. The company’s production plans have been delayed by at least four months, and a recent hearing in the parliament resulted in a recommendation that the project not go ahead. The company’s managers suspect that corruption may be one of the reasons Andash got suspended. Moreover, that instance may have been a part of a bigger picture: The government seems not to be keeping its promises to provide a level playing field.
On an international scale, Kyrgyzstan ranks quite poorly on the Corruption Perceptions Index: In 2010 it was ranked 164th out of 178 countries surveyed. As a matter of comparison, neighboring Kazakhstan fares a little better at 105th, but Uzbekistan and Tajikistan are down there with Kyrgyzstan at 172nd and 154th, respectively.
The picture this information outlines is quite grim: In terms of corruption, the Kyrgyz Republic does not seem to be an attractive jurisdiction for foreign investment.
Looking at the data provided by the World Bank, however, we see that there are improvements: The country’s overall “Doing Business” score is 44th among 183 surveyed economies. The country is up three positions from last year’s 47th place. This is quite interesting, especially given that “Investor Protection” – one of the most important areas to us – is where Kyrgyzstan fares very well: It’s ranked 12th on that scale. The booming Kazakhstan is 44th on the same scale.
The numbers, as you see, are quite controversial. Anecdotal evidence of local protests is not unique to Kyrgyzstan. It can mean pain for mining companies and their shareholders, but as the dust from the recent overhaul in the country settles, we may observe Kyrgyzstan becoming a more attractive jurisdiction than it once was. We met with government officials on our recent trip to the country and could see that even if the results might be lagging, the administration is trying to improve its business climate. Although Kyrgyzstan remains a risky jurisdiction in which to operate, we note that there’s an election coming up this fall; if the country can make that transition smoothly, and the new government remain as pro-business as the current one, that would be a positive sign for early adopters to have a new look for opportunities in the country.
((Kyrgyzstan is just one of the exotic places Louis James has visited this year in order to ferret out the best speculative mining opportunities. Put his diligence to work for you with a risk-free subscription to the Casey International Speculator. Learn more here.))
One-on-One with Peter Thiel (CNBC)
Here’s a good video of Peter Thiel, co-founder of Paypal, discussing education, technology, and regulation. Thiel points out that innovation has slowed to a crawl in the U.S. – largely due to government regulation at every step. The most successful sectors for technology have been computers and the Internet – both areas that enjoy minimal regulation. Coincidence? I think not.
BofA, Goldman Among Banks Cutting Jobs (Bloomberg) and Goldman Sachs Plans to Hire 1,000 in Singapore While Cutting U.S. Jobs (Huffington Post)
I’ve mentioned this trend before regarding the financial industry. We can’t let our anger destroy it permanently. In the past few decades, we’ve been complaining about the loss of good-paying manufacturing jobs to competition overseas. I certainly don’t want to be similarly complaining about college-degree jobs a decade from now. At this rate, we’ll all be flipping burgers and serving sodas to Asian and European tourists.
As the Huffington Post article notes, “The jobs are ‘high-paying, skilled positions in sales and investment banking.’” This isn’t the operations department handling the transactions and the paperwork: These are the big money-making positions, and they’re leaving New York.
Poll: 4 in 10 See “Permanent Decline” (Politico)
The poll linked to in the article goes beyond the U.S. economy and has some interesting results on the wars as well. However, here are some of the more interesting economic results:
Nearly 4 in 10 Americans say they think the economy is in permanent decline, a new polls [sic] shows as deep pessimism about the economy becomes more widespread. ….
Overall, 81 percent of Americans think the economy is in bad shape, while 18 percent say it is in good shape.
That’s it for today. Thank you for reading and subscribing to Casey Daily Dispatch.
Casey Daily Dispatch Editor