The chip giant settled with AMD. But another rival is making noise about anticompetitive behavior.

Nvidia's Huang is gunning for Intel. Photo: Nvidia

You’d think Nvidia CEO Jen-Hsun Huang would be happy.

After bumping along as low as $7 a share at the beginning of the year his stock is up near $14. Several months ago Apple (AAPL) began using his graphics chipset – a group of circuits designed to work together – across nearly its entire line of Macs, giving him a very high-profile endorsement. And in the white-hot netbook segment, his Ion processors have won raves for turning underpowered laptops into HD video machines.

Problem is, both of these acclaimed Nvidia (NVDA) products might be dead in the water.

Why? Huang blames chip giant Intel (INTC). Nvidia’s graphics chipsets, which Steve Jobs liked enough to buy by the boatload, aren’t allowed to work with Intel’s latest offering, code-named “Nehalem” – and in the computer world no Intel compatibility means no mainstream future. Nvidia’s Ion chip is designed to work alongside the chip giant’s Atom processor, but lately it’s been priced out of the market by – you guessed it – Intel.

Feeling AMD's pain

All of this has piqued the interest of the Federal Trade Commission, which is looking into whether Intel has improperly used its power in the computer chip market to choke rivals. For those who are handicapping the chances that the FTC will bring charges, Nvidia’s gripes have recently taken on new importance: Advanced Micro Devices (AMD), which had been Intel’s main critic, recently settled its antitrust complaints with Intel for a tidy sum of $1.25 billion. That leaves Nvidia to bang the anti-Intel drum.

“I’m sympathetic to what AMD had to go through over the years,” Huang says.

Intel says it’s a tough but fair competitor, and that Nvidia has caused its own problems. To wit: The two companies signed a limited patent-sharing deal five years ago, and Intel says that if Nvidia had read the fine print, it would have noticed that designs like the current Nehalem chips weren’t covered in the agreement. And the Ion dispute? Well, Intel says it just lowered prices on its Atom chipsets to stay competitive. What’s wrong with that?

Nvidia tells a different story. Executives there say they’re a victim of the same kinds of tactics that got Intel into antitrust trouble in Europe and Asia. Nvidia believes it should be authorized to make Nehalem-compatible chipsets under the 2004 agreement; it claims Intel is just getting litigious to stop a competitive threat. And with Ion, Nvidia accuses Intel of unfair pricing that locks it out of the market. Intel says Nvidia just doesn’t understand the incentives it offers customers.

(The two companies are battling in court over the chipset agreement. Intel has asked a Delaware court to clarify whether Nvidia has the right to build Nehalem chipsets, and Nvidia has countersued for breach of contract.)

The courts will ultimately decide who’s right – that is, unless Intel decides to ink a settlement with Nvidia, too. Even then, that probably won’t be the end of Intel’s legal headaches. Maybe that’s the trouble with being a giant. There’s always some kid with a slingshot gunning for you.

follow me on Twitter: twitter.com/jonfortt

Corporate venture capital is one of the corners of the VC world that runs extremely hot and cold. When the startup world is gathering interest and money, practically every large company – even some small outfits – trots out its own venture investment group. But just as fast as they pile in with their corporate cash, the suits also run for the exits when times get dicey.

Take the previous tech boom-and-bust cycle. As the ‘90s ended and this decade began, corporate VC investment in startups soared from $468 million at the end of 1998 to $6.2 billion at the beginning of 2000, according to Thomson Reuters.

When the bottom fell out of the tech economy, the corporate cash crashed too, down to $848 million in the third quarter of 2001. Never mind that corporate VCs inevitably lose money on their deals, it appears that most public companies just don’t seem to have the stomach for it.

Still, a handful of tech companies have consistently stayed in the corporate VC game, including Microsoft (MSFT), Qualcomm (QCOM), and more recently Google (GOOG). For these tech companies, buying technology and talent early is worth the risk (they all also happen to be sitting on billions in cash to invest). But perhaps the most steadfast corporate VC is Intel (INTC).

The chip giant has invested some $9.5 billion in more than 1,050 companies since 1991. Its investments run the gamut, from seed investments of a few hundred thousands dollars in startups to buyouts of large companies like its recent $884 million acquisition of embedded software company Wind River Systems (WIND) (which, let’s be clear, is not VC; it’s a buyout). I had a chance recently to sit down with the head VC at Intel, Arvind Sodhani (his official title is executive vice president, and president of Intel Capital) to talk venture investing at the world’s largest chip company.

The problem with much of corporate VC is the reason for doing it. Is it strategic gains or investing to make a return? Without a clear purpose, corporate VC usually gets neither. Intel Capital’s reason for being is both, Sodhani says. “We need to return money to our shareholders, but we also need create demand for our technology through innovation and entrepreneurship,” he says. What the latter part boils down to is selling more chips.

In the early days of Intel Capital, the 1980s, the selling-more-chips part was accomplished by investing in specific software companies that could run on Intel architecture and promote the sale of more computers systems powered by Intel’s chips. One example was an early application that performed Chinese translation of PC software already tuned for Intel chips.

Today, Sodhani’s mission is a bit more diffuse. He and his team around the world are investing generally in companies that promote computing, the flow of more data through data centers, through smart electric meters or through the latest social network fad. “We benefit from growth in all those areas,” Sodhani says. “Take clean tech and the electric grid. As the grid becomes more intelligent, more computing will go into things like household meters. We want to get our Atom processor into meters, and there are 120 million households in the United States alone.”

At Intel’s recent CEO Summit in Huntington Beach, Calif., Sodhani announced seven new investments totaling some $25 million. The lucky companies included Joyent, a Bay Area-based cloud computing company that competes with Amazon’s EC2 service; a Japanese video conferencing startup, V-Cube; and Korea-based Crucialtec, which manufactures trackballs for mobile devices, among others. He also announced a series of follow-on investments. Among that class are Argentine social gaming company Vostu; India-based local entertainment events portal (Buzzintown.com), and an Israeli security software company (Safend).

It may be a winding route from social gaming in Argentina and trackballs in Korea to selling more chips, but that is the point, Sodhani says. That and making money on his deals. “We have had more than 400 portfolio companies exit,” he says, adding,  “We are in the top tier of investment returns.” For those who have spent any time in the VC world, that last statement ought to ring a bell. Every VC, according to their calculations, is in the top tier. Whether that is true or not in Intel’s case (it depends on what your definition of a tier is), is a bit beside the point. Sodhani and his team do have a different mission than most VCs, and their only limited partner is their corporate parent and their shareholders. Even so, it’s nice to hear a corporate VC brag, a sign that these are no dilettantes; they are in it to win.

A few years ago rivals mocked Jonney Shih, chairman of Asustek, and his purse-size laptop computers. Millions of netbooks later, Shih is having the last laugh.

Jonney Shih, CEO of Asus, in Taipei.

On a hillside above the Hsing Tian Kong temple in the northern reaches of Taipei, Jonney Shih sits on a wobbly stool next to an ornate low wooden table. Dressed in a taupe suit, white shirt, and silver tie emblazoned with jaguars, Shih, 57, cheerfully waves off three umbrella-wielding employees who try in vain to shield their boss from the hot sun and a swirl of menacing bees.

But Shih, who is waiting to be photographed for this magazine, sits serenely, perspiration-free in the sun, intent on a game of Chinese chess. "In Buddhism you learn to accept everything, to let it flow through you," Shih says. "Then you can slow down and think clearly."

It turns out the ferociously driven Shih is a less-than-model Buddhist. (Buddhists aren't supposed to be thinking about technology while they're meditating — something Shih is known to do.) But his ambition, combined with engineering skills and spot-on business instincts, also makes him the most brilliant technology executive you've never heard of.

He is the largest shareholder and chairman of Asustek (pronounced a-soos-tech), the $21-billion-a-year tech conglomerate that introduced the first netbook three years ago, ushering in a revolution in the stagnant PC industry. When it hit stores in the fall of 2007, Shih's $399 EeePC was derided by rivals as a low-power plaything. But Asustek, or Asus for short, went on to sell millions of the mini-notebooks and soon vaulted to No. 5 in worldwide PC market share.

Today virtually every PC manufacturer on the planet, including Dell (DELL), Hewlett- Packard (HPQ), and Toshiba, offers its own version of netbook. (The exception is Apple (AAPL).) But the biggest netbook maker, with 38% of the market, is another Taiwanese tech company, Acer, which also happens to be Shih's former employer. Asus, which had the market all to itself for about eight months, is now in second place, with a 30% share.

So Shih, after creating what has grown into a $10 billion category in two years, needs to come up with another breakthrough, and he'll apply his own flavor of Buddhism to the challenge. "Most people think Buddhism is passive or about escape," he says. "It's not. It's about confronting what's in front of you with a clear and flexible mind. That might be a hot day or your competition, but you accept it and do everything the best at that moment."

That sounds fairly magnanimous, but whether it's in Chinese chess or the PC world, Shih's best effort has a way of crushing the life out of whoever gets in the way. Shih's "not that well known in the West because he doesn't put himself first," says Jen-Hsun Huang, CEO of chipmaker Nvidia. "He's humble, but he always has a mental model for exactly what he wants his company to do."

The Giant Lion
Asus was started in 1989 by four former Acer engineers. (The name "Asus" comes from the mythical Greek horse, Pegasus.) At the time Acer, one of the original companies to transform the island of Taiwan into the center of computer manufacturing in the world, had already gone public on the Taipei exchange.

Many Acer employees took their stock gains and launched their own businesses. At a café in Taipei, four subordinates tried to persuade Shih, who was running R&D at Acer at the time, to join them in starting a company to design and manufacture motherboards — the central circuit boards in PCs that connect crucial components, including the processor and memory.

Shih demurred out of loyalty to his mentor Stan Shih (no relation), co-founder and chairman of Acer. But he did encourage his former reports to start Asus and took a stake in their new company. In 1994, after three years as president of Acer's business unit — selling Acer technology rather than designing it — Shih joined Asus as CEO.

When a chip company comes out with a new processor, it's up to the motherboard designers to integrate that chip into a standard circuit board that can run the computer. Whichever company can get its motherboard out first and squeeze the highest performance out of a chip set wins the business of the PC makers.

Back when Shih was at Acer, he made his reputation by building killer motherboards. When Intel (INTC) was rolling out its 386 processor in 1985, Shih and a group of engineers showed up at the Computer Electronics Show in Las Vegas ready to do battle. "We didn't sleep much that voyage," Shih recalls.

In the competition among motherboards, Dell's offering was the highest performing, but it wasn't a technology suitable for mass production. Shih's was, and it beat out the best from IBM (IBM), Compaq, and everyone else. (At the time, many PC makers produced their own motherboards, which they sold to other manufacturers as well.) The orders came pouring in, and Shih's reputation around Intel and the rest of the PC industry was made.

After moving to Asus, Shih continued his success with Intel's 486 processor, and computer makers such as Hewlett-Packard, Sony (SNE), and Dell found that if they used Asus motherboards, their computers performed better. By the mid-'90s, Asus sold more motherboards than anyone, and its revenue and profits climbed steadily throughout the decade.

But in 2001 other companies, ECS and Foxconn, started undercutting Asus's prices in the motherboard business. Asus's share in unit volume fell to No. 2, and annual profit dropped dramatically to $300 million in 2002, from $500 million the prior year. In response Shih launched what he called the "giant lion" strategy.

"You need to be a lion. A lion has position in the jungle," Shih says. "So we kept driving the performance, quality, and innovation of our motherboards — we kept our leading position in the jungle. But I realized that at the same time you have to have big market share. You need to be a giant lion." Shih founded a subsidiary, ASRock, to compete at the low end, leveraging Asus engineering and manufacturing.

The giant lion mauled the competition: Within two years Asus was back as the No. 1 revenue producer in the motherboard business, and its volume exceeded the output of the second, third, and fourth companies combined.

Meanwhile the company started making "white label" consumer electronics for the likes of Dell, HP, and Sony. Named Pegatron after the first part of "Pegasus" (the wholly owned subsidiary was spun out in 2008), it manufactured notebooks as well as routers, MP3 players, gaming consoles, and whatever else big brands wanted made. But Shih wasn't satisfied to be a mere contract manufacturer, and in 1997, Asus started making laptop computers under its own brand.

A computer for the masses
Asus is Taiwan's HP and Apple rolled into one. It is the No. 1 seller of notebooks there, but its laptops win for their performance, reliability, and style, not their discount prices. Asus has notebooks covered in leather, hand-polished steel, even bamboo.

But Shih's ambitions extend beyond what clearly is a maturing market. He wanted to build a machine for the next billion PC customers. His breakthrough notion was to provide a device that offered good enough performance to surf the web and do simple computing tasks in a very easy-to-use, affordable package.

Fortuitously, Intel at the time was working on a chip that would help Shih accomplish his goal. "Behind the scenes we had been working on Atom, our low-cost, lowpower chip," says Sean Maloney, Intel's executive vice president. "Jonney immediately wanted it." The question was how to package a machine around it.

For three months Shih and the head of Asus's motherboard business, Jerry Shen (now the Asus CEO), personally worked out the basic concepts: what features to include (Wi-Fi, a touchpad, and a solid-state drive) and what to throw out (Microsoft Windows, initially, and a full-size keyboard). Then they brought in a team of engineers to make their ideas real. At one point, as they struggled over the machine's software interface, Shen locked the team in a Taipei hot-springs hotel for two days. They finally emerged with their answers. When the first few thousand EeePC netbooks went on sale in Taiwan in October 2007, they sold out in 30 minutes.

Like other tech companies, Asus was hit by the global recession, and last winter it posted its first quarterly loss in the company's 20-year history. The company has cut costs through layoffs and salary cuts, and has scaled back its inventory. More recently, however, Asus has rebounded, blasting through analyst estimates for its third quarter, and its stock is trading at a 52-week high on the Taipei Exchange. (Asus has made Shih rich, but his only concession to his wealth is a chauffeur-driven Volkswagen Phaeton. "It's 80% of a Bentley and half the price," he jokes.)

Asus, which gets 40% of revenue from Asus-branded technologies, is forecasting a 30% increase year over year in netbook and notebook sales in 2010. Of course, rival Acer also forecasts growth, and the maker of the Aspire One model isn't likely to cede its No. 1 position in netbooks anytime soon. And so Shih is spending his time meditating about Asus's next industry-changing hit.

The next netbook
At Asus headquarters in a bright corner room filled with fabric swatches and beanbag chairs, the next phase of Shih's clear thinking is being prototyped. This is the company's top-secret design lab. Lying on counters are notebooks that look as if they are folded, origami-style, from sheets of aluminum. Others have keyboards that slide back and slightly up when the case is opened for a more ergonomic position. An international team of designers swap ideas on couches.

Shih's instinct tells him that the "next netbook" won't come from an engineering specification but from understanding how people use devices to communicate, get work done, and play. More than ever he is pouring company resources into design.

He pulls out a prototype of the forthcoming Eee Keyboard, an aluminum-clad keyboard with a touchscreen on one side. Via a wireless connection, it turns a flat-screen television into a websurfing, Facebook-friendly device. From his pocket emerges a smartphone that Asus developed with navigation company Garmin (GRMN).

The Asus-Garmin phone has been a dud, and the keyboard isn't out yet, but those items suggest that Shih is thinking about more easy-to-use, affordable products that are integrated as part of a digital lifestyle. "My competitors are doing their own version of the EeePC," Shih says, "but I don't know if they have the vision of how everything can work together."

Is Shih's insight about integrated technology the "giant lion" that will help Asus regain its leadership position in netbooks? It hardly sounds revolutionary, but by now rivals know better than to underestimate Shih, especially when this "bad Buddhist" is thinking clearly.

Settlement or no settlement, Advanced Micro Devices CEO Dirk Meyer still doesn’t trust Intel. That much was clear when I sat across the table from him last week at AMD headquarters in Sunnyvale.

I had a chance to talk to Meyer during the AMD analyst meeting last Wednesday. In a series of presentations that morning, Meyer and his lieutenants had made a cogent case that 2010 would be their comeback year. They have cut costs, pared non-core businesses, and returned to designing good chips and delivering them on time. Next year the company plans to begin selling several new chips, including a first: one that has a microprocessor and a graphics processor on the same chip.

But what about Intel (INTC), I asked him. In the past when AMD (AMD) had good products, you accused the chip giant of using its market dominance to unfairly squash you. (Intel has maintained that it plays by the rules.) Why should this time be any different? Do you think the playing field is level now?

Meyer knew what I didn’t: that in a few hours, he and Intel CEO Paul Otellini would announce a massive $1.25 billion legal settlement with Intel that’s supposed to put the companies’ acrimonious past behind them. That didn’t seem to change Meyer’s outlook on the chip giant, however.

“I would not tell any regulator around the world who’s still looking at Intel’s business practices to stop,” Meyer said. “The work that we’ve done needs to continue.”

Sounds to me like these two rivals might have put down the legal hatchet for now, but they haven’t quite buried it.

follow me on Twitter: twitter.com/jonfortt

Ben Foss, director of access technology in Intel's Digital Health Group, uses the Intel Reader to scan a book. Photo: Jon Fortt.

When Ben Foss's father in law was dying of liver cancer months ago, friends suggested Foss read "How We Die" to help the family with the grieving process. Foss has dyslexia, and finding an audio version of the book or scanning it into a computer typically would be an ordeal. But in this case he was able to plow through it at 250 words per minute.

Foss did it with an early version of the Intel (INTC) Reader, a $1,500 device he dreamed up along with colleagues in Intel's Digital Health Group. The device launches today as the first consumer product from the five-year-old group. And though its name seems to place it in the same category as trendy ebook readers from Amazon (AMZN), Sony (SNE) and Barnes & Noble (BKS), this reader is profoundly different.

This is not another thin tablet that displays text; instead it's more like a chunky digital camera that instantly captures the words on a printed page and pronounces them aloud. That makes it little more than a curio for mainstream gadget lovers, but a potential godsend for those who struggle to read standard text because of learning disabilities or vision problems.

The reader also poses a real risk for the world's largest chipmaker. Most of its previous forays into the electronics world have ended badly. Intel generally has thrived when it has focused on its core business of providing computing brains, to others and foundered when it has tried to design and market its own finished products.

But as demand for traditional PCs has slowed, Intel has resolved again to target non-traditional areas for growth. In healthcare technology, executives believe they have found a niche where they can build innovative products that medical professionals and consumers will pay a premium to use. The Intel Reader is poised to provide a high-profile test of that strategy – and it delivers a strong first impression.

The Intel Reader has a 4.3-inch LCD display, 4 gigabytes of flash storage and USB slots for adding more. It weighs 1.38 pounds. Image: Intel.

Perhaps that’s because for Foss, the reader is far more than just another project – it’s part of his life’s work. The 36-year-old's learning disability has driven him to work for decades on more convenient ways to get information off the page and into his head. “They say necessity is the mother of invention, but it’s not,” Foss says. “Frustration is.”

Foss lived as a frustrated reader for years. In high school, his mother read aloud to him to help him process the information in assignments. In law school, he scanned book pages onto his laptop and had text-to-speech software play them back to him as he rode a bike to class. Now technology has advanced enough that, along with a team of Intel designers, engineers and software architects, he was able to squeeze a scanner and a digital reader into a device about the size of a paperback novel.

On the outside, the Intel Reader doesn’t look like breakthrough technology. Next to an iPhone, it looks like a museum piece – maybe a mid-90's digital camera prototype. But unlike Apple's (AAPL) iPhone, Intel’s reader isn't designed for the mass market. The finished product is the result of untold hours of tweaking; Foss carved the first prototype from a Styrofoam beer cooler more than two years ago.

Inside there's some impressive technology, including an Intel Atom processor and a Linux-based operating system. Simple menus on a high-contrast screen cater to low-vision users. A thick battery guarantees 4 hours of unplugged use and 30 hours of standby time. A 5-megapixel camera on the bottom of the reader captures images of pages; optical recognition software converts the images into digital text. Text-to-speech software creates an audio file that users can transfer to another device or play aloud on the reader at an adjustable pace.

The slow setting goes below 120 words per minute. The fast setting blazes along at 250 words per minute – a less frustrating pace for speed listeners like Foss.

follow me on Twitter: twitter.com/jonfortt

The state of the state? "A train wreck," says one official.

If the world’s eighth-largest economy were a member of the proper religious order, it’d be time to call in a priest to administer last rites.

Name almost any serious malady and the state of California has it: the nation’s highest marginal tax rate coupled with an abysmal public education system; the most home foreclosures; a free-falling commercial real estate sector; lame-duck governor with no legislative support and a disdain for an annual budget process that he refers to as kabuki theater; unemployment somewhere between the official number of 12% and the whisper number of 18%; a 20% drop in year-over-year revenue; municipalities that have either declared bankruptcy (Vallejo) or are on the verge (Los Angeles); and a black-box permitting process that scares away business investment even while every week, 3,000 more taxpayers migrate to greener pastures.

Californians may be a can-do lot, but faced with all that evidence and much more, the political and economic leaders who spoke at the Milken Institute’s annual “State of the State” conference held yesterday at the Beverly Hilton could hardly have been more dour. “It’s a train wreck, and it’s getting worse,” said Bill Lockyer, California State Treasurer. Added former Assembly speaker Bob Hertzberg, now co-chair of governance reform group California Forward, “A high-speed train wreck.”

What’s this got to do with technology? Nothing, unless you consider that California is home to the many of the biggest tech companies on the planet (and 51 members of the FORTUNE 500), the bulk of the venture capital industry, many leaders of green-tech, two of the most patent-producing universities in the world in Stanford and UC Berkeley, and top thinkers across all spectra.

“California represents 10% of the population of the United States,” said Eric McAfee, chairman of McAfee Capital and CEO of AE Biofuels. “but probably 50% of innovation.”

Tech into ploughshares?

Ironically, Silicon Valley has built the tools and infrastructure to allow talented people to work anywhere in the world they choose, and as the state circles the drain, the fear is that businesses, entrepreneurs, and students will no longer feel the pull of the Golden State.

From a political and budgetary perspective, California has myriad problems – from Proposition 13 to direct-democracy ballot initiatives to abysmal credit ratings. But the biggest problem may be girth. The state increasingly seems too big not to fail.

“States were never intended to be the size of the entire eastern seaboard,” said Hertzberg. “What is the commonality between the folks in Calexico and Crescent City? This manifests itself in a politboro style of government in Sacramaento.”

None of the speakers–including gubernatorial candidates Gavin Newsom and Steve Poizner–offered a magic bullet, but there was some consensus on where to start.

Many called for an adjustment to the mandate that 2/3 of the legislature must approve a budget or a repeal of term limits that seem to enslave legislatures to special interests.

Others craved less (or more, take your pick) taxes and a lessened (or at least consistent) regulatory structure. Chevron's (CVX) John Watson, who will assume the CEO role in January, said the permit process for an upgrade to the company's Richmond, California, refinery took four years while an Indian company built its own entire refinery in half the time.

Still others offered hope that things will somehow work themselves out–because California has been in the dumps before and that the state will continue to be a magnet for the brightest immigrants.

“The combination of great science and great local universities and venture capital money started not just Genentech in 1976, but an entire industry,” says former Genentech president and current UCSF chancellor Susan Desmond-Hellmann, showing a slide of California tech companies that were either founded by immigrants or where immigrants played key roles in the early days, including Yahoo (YHOO), Sun (JAVA) , eBay (EBAY), Intel (INTC), Google (GOOG), Viewsonic, SanDisk (SNDK), and Nvidia (NVDA). “There’s something special about California: the atmosphere, the diversity, a wish to be curious and ask questions. The secret sauce behind Genentech, the Googles, Yahoos, and Amgens is the combination of entrepreneurial spirit, great universities, and a willingess to marry business and science.”

Desmond-Hellman warned, however, that Genentech has remained a biotech powerhouse because it has never lost sight of where its power comes from. “We always believed that the company would be no better than the people we recruited and we keep," she said. "We never took for granted that people had to work at Genentech. They have choices.”

So do the 3,000 Californians leaving the state every week.

Vista was too bloated for many corporate PCs to handle when it arrived three years ago. Its successor, Windows 7, is a better fit. Photo: Microsoft.

Rewind three years. One of the harbingers of doom for Windows Vista, Microsoft’s much-maligned operating system, was a survey that showed half of corporate PCs were too old or anemic to upgrade from XP to Vista. In other words, to get the new software, companies would have to spend a fortune on new computers.

Most never made the upgrade: 92% of PCs in the United States and Canada are still running XP, according to Softchoice, an IT consulting firm.

So it’s a good omen that Softchoice, the same firm that issued that fateful Vista study, is singing a very different tune about Windows 7. This time, its survey of 450,000 corporate PCs in the U.S. and Canada shows that 88% can handle the upgrade.

That matters because with today’s tight IT budgets, some companies will look to install Windows 7 on PCs they already own rather than buy new ones.

“Corporations, more now than ever, are trying to stretch a buck,” says Dean Williams, services development manager for Softchoice. And this time, they can. “Hardware requirements are not nearly the issue that they were three years ago when Vista was launched – it’s like night and day, really.”

Stats like this are fueling optimism in the tech world, since a popular new Microsoft (MSFT) OS can jump-start sales of other gear and services. Softchoice is among the hopeful: it offers software that helps companies plan their OS upgrades, and this survey seems to indicate that there’s demand for its services.

PC industry titans expect healthy sales of new Windows 7 PCs, too. Intel (INTC) CEO Paul Otellini said this week that the new OS has inspired some big companies to look at buying PCs again. Todd Bradley, executive vice president of Hewlett-Packard’s (HPQ) PC group, told Fortune last week that he expects Windows 7 to help retail sales this holiday season.

But for the tech industry, here’s the best Windows 7 stat of all: the average corporate PC is now four or five years old, since companies have delayed purchases in a down economy. That’s ancient by IT standards – old enough that it’s more expensive to maintain them that it would be to recycle them and buy something new.

Shaw is the latest non-technologist to ascend at Intel. Photo: Intel.

Who rules techland? Increasingly, it isn't the inmates.

In May, when Craig Barrett retired as chairman of Intel (INTC), the choice of his replacement marked a momentous occasion for the granddaddy of the semiconductor industry.

That Jane Shaw became nonexecutive chairman of Intel is a big deal, but not because she is Intel's first outsider to chair the board or because she is the first woman.

What makes her role noteworthy is that she is the first non-technologist in that seat. Yes, she has a science background, with a doctorate in physiology and a career in the pharmaceutical industry. But she's not a technologist in the Silicon Valley sense.

Considering that Intel's CEO, Paul Otellini, is the first non-technologist to run the company (see my 2005 profile of him) and that his most likely successor, Sean Maloney, isn't a silicon guy either, it's a remarkable turnabout.

Revenge of the anti-nerds

Now, this isn't a piece about Intel, though it could be.  Shaw is perfectly well qualified to be Intel's chairman, having served on the board — loaded with academics, retired government officials and three corporate types, not one of whom has a commercial semiconductor background — for 16 years. (She shows up in Brent Schlender's 2004 take on how Andy Grove was trying to re-make Intel's board and himself.)

The point is that Intel's experience represents one of the quietest yet fiercest battles going on across the Valley: the technologists versus the business people.

Time after time I come across people in the Valley who want to talk about this, most of whom are technologists bemoaning their loss of power. Hewlett-Packard (HPQ) hasn't had a gearhead at the helm for years. Yahoo (YHOO) ditched its engineer/founder/CEO who missed key turns in the Internet industry. Google (GOOG), firmly run by engineers, struggles to retain its top business talent because they know they can't move up.

Most careful Apple (AAPL) and Oracle (ORCL) watchers dread the day when their tech-savvy CEOs no longer are around to work their innovative magic.

Growing up, or selling out?

The company that seems to be biggest exception is Cisco, (CSCO) where salesman-in-chief John Chambers has skillfully guided the company for years.

Cisco, however, isn't lauded for its innovation. It is known for superior execution, deft acquisitions and a clear understanding of market opportunities — not the sorts of things that makes the hearts of engineers go pitter-patter.

The charitable explanation for all this is that Silicon Valley is becoming mature. It has grown up into  a real industry, not a frontier collection of maverick companies, and its leaders approach their tasks in an industrial fashion.

The less cheerful interpretation is that the glory days are gone, that Silicon Valley is little more than a bunch of careerists and, worse, venture capitalists.

The truth probably lies somewhere in the middle: Yes, Silicon Valley has become a destination for new MBAs who see the tech biz as a money-making opportunity rather than a passion to pursue. But the Valley will continue to produce interesting new companies – founded by whip-smart engineers and technologists.

If those founders want their companies to evolve into the next Intel, they're probably going to need a few of those spreadsheet jockies to help them get there.

Changes in computing mean software companies need hardware-savvy employees

By Sam Blackman, CEO and co-founder, Elemental Technologies

Blackman: Your next hire might need hardware chops. Photo: Elemental Technologies

Whether we knew it or not, we’ve all been relying on something called “Moore’s Law.”  Back in the 1960s, Intel (INTC)  co-founder Gordon Moore noticed that the number of transistors that could cheaply be placed on an integrated circuit had been doubling every two years.

That meant that central processing units, or CPUs — the chips that drive computer performance — were getting twice as fast in that same time period. That amazing rate of technological change has held up for more than 40 years.

Moore’s Law is why we take it for granted that the cell phone we carry around today is more powerful (and cost us less) than the top-of-the-line desktop computer we bought ten years ago. It is also why we’re not surprised that in less than a decade the Web has changed from a place to look at ugly text pages to a place to watch high-definition TV shows.

But after 40 years, Moore’s Law is slowing down. We’ve finally reached the point where faster processors consume too much power, and manufacturing them to achieve ever-higher frequencies gets to be too expensive. This technological pressure will radically reshape the way we build computers and write software in the years to come.

Going forward, computers will get faster by adding additional processors that work together to solve problems. That’s why we hear more these days about the number of cores in the CPU rather than how fast the processor is in our computer.  Giants like Intel and Nvidia (NVDA) are racing to create new “massively parallel solutions,” composed of as many as 240 individual processors designed to work in concert to solve problems.

Unfortunately, writing software that runs well on massively parallel systems is incredibly difficult.  Engineers need to figure out how to break big problems down into smaller pieces that individual processors can work on at the same time, how to keep all of the individual processors coordinated with each other, and how to assemble all of the work into a useful output.

At the recent Hot Chips microprocessor design conference in Palo Alto, Calif., John Hennessey, the president of Stanford University, called parallel computing “the hardest problem in computer science.”

To date engineers have only solved a small set of problems using parallel systems, and it’s not for lack of trying.  Microsoft (MSFT) and Intel think that figuring out parallel computing is so important that they’ve invested $20 million funding parallel computing research centers at the University of California Berkeley and the University of Illinois at Urbana-Champaign.

Difficult or not, the future of computing is going to be on massively parallel systems. Some savvy companies are already taking advantage of massively parallel systems to trade stocks, search for oil, and offer online video games. At Elemental Technologies, we’re building software to help professionals process video files faster and more economically than ever before.

To build the kind of team that can take advantage of these massively parallel systems, software companies are going to have to rethink the mix of engineers that they are hiring. They will need people with experience in hardware design and low-level “close to the metal” programming. Engineers who understand how these new massively parallel architectures work, and know how to parallelize problems.  Today, programmers with these skills are in seriously short supply.

There’s a pool of great engineers who don’t even realize that their future is working for software companies yet, though. They’re the digital hardware engineers who have spent their career working for chip companies and startups working on things like embedded systems and integrated circuits – where the parallel processing paradigm has been in use for years, since that is the way physical devices work. The smartest software companies will snap up as many of these engineers as soon as possible in the next few years and put them to work building software that can take advantage of the computers of the future.

Companies that don’t harness this resource will find themselves disrupted by faster, cheaper, and smarter software from competitors who did.

Blackman is CEO of Elemental Technologies, a Portland, Ore.-based company that develops massively parallel software to help computers convert video more efficiently.