TUCKED IN A CORNER of a hangarlike space along North Austin’s high-tech corridor, fewer than a hundred dedicated techno-nerds spent most of last year logging eighteen-hour days at used metal desks. Short on sleep and wired on Diet Coke, they scurried about in a warren of cubicles, trying not to be distracted by the banging on the assembly line on the other side of the building or the chatter from the telemarketers upstairs. The walls were bare, the floor was concrete, and amenities—like windows—were nonexistent. There wasn’t even a sign on the door.

Yet out of these spartan surroundings came one of 1995’s hottest high-tech launches: Power Computing, the first company to manufacture and sell Macintosh clones under license from Apple Computer. Just as two other Texas companies, Compaq and Dell, came from nowhere to dominate the market for IBM clones in the eighties, Power hopes to do the same for Macs in the nineties, and already it is off to an impressive start. Last spring a skeleton staff introduced three inexpensive Mac clone models and, a few months later, rolled out another clone that is faster than anything Apple has ever offered—all to rave reviews from industry bibles like MacUser and InfoWorld. By this summer, Power will have shipped 100,000 personal computers, nearly twice as many as Compaq shipped in its first year.

Power’s arrival on the scene comes at a fortuitous time. The success of Compaq, Dell, and others in marketing knockoffs of IBM PCs suggests that consumers care more about what a computer does than who makes it—they’d rather pay less for a machine that performs the same functions. Yet the demographics of the clone market have changed. In the eighties most young computer buyers had grown up with IBMs, so it made sense to clone IBMs. Today most young buyers have grown up with Macs, so Power is well positioned for a sales boom. Another factor in Power’s favor is the war between Apple and the increasingly dominant “Wintel” camp (PC makers that use Microsoft’s Windows software and the Intel chip). Apple is desperate to maintain and expand its market share—which includes sales by its licensees—so it will do whatever it can to help Power prosper.

Of course, good timing would mean nothing were it not for good management—and that means CEO and president Steve Kahng (pronounced “Kong”), a moonfaced 46-year-old native of South Korea. An engineer by trade, he displays none of the brash arrogance of many entrepreneurs, but he is clearly a savvy businessman. For one thing, at the same time that he was negotiating for the license with Apple’s executives, he was raiding its top talent. Half of the eighteen engineers at Power’s outpost in Cupertino, California, are Apple defectors, including vice president of engineering Jon Fitch and his deputy, Carl Hewitt, two key architects of Apple’s Power Macintosh line. “Apple has a lot of engineers,” Kahng says in a heavy Korean accent. “Nine or ten isn’t that many.”

Kahng has also turned Power Computing into a magnet for young, talented Mac lovers eager to get in on the ground floor of a new venture. Hewitt is only 29. Marketing director Michael Rosenfelt, who held the same position at a Kansas City—based start-up called APS Technologies, is 28; a University of Texas at Austin dropout, he talks at only one speed—fast forward—and preaches about Power’s mission to bring the Mac to the masses. The chief of operations, 29-year-old Ric Ralston, also came from APS, where he had set up that company’s technical-support, product-development, and customer-service operations. Brian Combs, Power’s 26-year-old on-line evangelist and Webmaster, is a high-tech hero who created the popular Austin City Lim-its Internet site. These and other rela-tive youngsters serve as Mac mentors for Power’s grown-ups, many of whom are veterans of IBM clonemakers like Compaq, Tandy, and Digital Equipment Corporation.

Perhaps Kahng’s greatest strength as a manager, though, is his tightfistedness. “It’s no secret that Steve is a skinflint,” says Henry Hirvela, Power’s chief financial officer. Indeed, although Kahng is a multimillionaire, he lives modestly—he dresses mostly in polo shirts with the red-and-black Power logo—and expects his company to do the same. The whole reason he put down roots in Austin last February, he says, is that “California is too expensive.” The city won out over Portland, Oregon, because with companies like Apple and Dell already doing business here, a less expensive infrastructure was in place for manufacturing Mac clones and setting up technical-support and mail-order operations. And Power’s unglamorous first home in Austin—an unused corner of struggling IBM clonemaker CompuAdd’s factory floor—came rent-free. “Research and development is a priority,” Kahng says. “Fancy offices are not.” He jumped when CompuAdd’s CEO, Rick Krause, offered space and labor to Power in exchange for a fee for every clone that rolled off the assembly line. It meant that Kahng could spend his start-up money to pay top engineers rather than build a plant and hire an assembly crew.

Power’s employees love to swap stories of Kahng’s penny-pinching ways, like the time Hirvela came to Austin for his job interview. Instead of springing for a hotel room, Kahng insisted on playing host at his place: a two-bedroom apartment near the Arboretum shopping mall. “Steve neglected to tell me that he didn’t have a bed,” recalls Hirvela, who ended up sleeping on the guest room floor. Then there’s Kahng’s car, a wheezy 1984 Mercedes with more than 215,000 miles on it. To draw attention to Kahng’s frugal ways, the company tried to raffle it off last year at an industry trade show. (The winner took a computer instead.) Employees travel in pairs to take advantage of friends-fly-free discounts. Until last summer they all shared Kahng’s business cards—crossing out his name and substituting their own.

Yet Kahng’s cost-cutting doesn’t mean he’s unwilling to part with his money. “Steve spends huge amounts where it makes sense,” says Geoffrey S. Burr, Pow-er’s vice president of sales and marketing, noting that Kahng has already poured more than $10 million into engineering and launched an irreverent $7 million advertising campaign in the Macintosh trade press (promotional T-shirts featured Kahng’s face and the slogan, “I love the smell of silicon in the morning”).

Kahng’s approach to business is an outgrowth of his upbringing. He was born in Seoul six months before the Korean War began to a housewife and a professor of mechanical engineering who managed to support his family comfortably despite wartime shortages. “We were relatively rich by Korean standards thirty years ago,” he says—but certainly not by American standards, as he discovered when he came to the U.S. with his family at age seventeen. “Americans waste a lot,” he says.

After earning a master’s degree in computer engineering from the University of Michigan, Kahng began designing giant mainframe computers for IBM in Poughkeepsie, New York. Soon he was trying—unsuccessfully—to clone mainframes for Trilogy Corporation in Silicon Valley. He had better luck with PCs. His Leading Edge Model D, designed for Daewoo of South Korea, was one of the best-selling IBM clones of the eighties and helped make IBM-compatible machines the choice of most computer users. Yet Kahng got only $200,000 for his invention; he had opted for a lump sum rather than royalties. “At the time, I was hoping Daewoo would sell a couple thousand,” he remembers. “I didn’t expect it to be so successful.”

Kahng got rich anyway. As a consultant to other major clonemakers, he was quickly pulling in $3 million a year. But just as quickly, he grew tired of working to make other people successful and wanted to realize his own dream—to build a computer company. To that end, in September 1993 a friend hooked him up with Elserino Piol, the vice chairman of Italy’s giant business equipment maker C. Olivetti and Company, which wanted to build a computer company based on the superfast PowerPC microprocessor. After a chat at a Silicon Valley eatery, Piol chose Kahng as his CEO. A $5 million ante from Olivetti and $4 million of Kahng’s own money provided the start-up capital for Power Computing in November 1993.

Given the crowded IBM clone market, Kahng—who had never even turned on a Macintosh—decided Power would do best to copy Apple’s fabled computer. The problem was, Apple had always refused to license its Macintosh operating system, or OS—the base layer of computer software. But with its share of the personal-computer market worldwide dipping below 8 percent, Apple was in danger of being deserted by software writers and overrun by Microsoft’s hot Windows package. Apple’s precarious situation forced it to take a calculated risk: that any loss of its sales to clonemakers would be offset by licensing fees and an increase in the sale of software and peripherals such as printers.

Kahng began negotiating with Apple in April 1994, but it was a tough sell. Apple was already eyeing bigger-name cloners like Compaq, Motorola, and Zenith. Nevertheless, with help from former Apple engineers, Kahng built a Mac prototype. By November he had designed a clone that was cheaper to manufacture and used off-the-shelf power supplies and monitors instead of Apple’s more expensive components. Both Piol and Apple executives were impressed. And on December 17, 1994, following Apple’s fruitless effort to attract better-known partners, Kahng’s tiny company became the first Mac OS licensee. “We took a risk and jumped in without any real guarantees from Apple, and it paid off,” says Kahng.

The dash to market began with the seven-day workweeks that Kahng had grown accustomed to during his IBM cloning days. A scant four months later Power, with a burgeoning staff of fifteen, introduced the Power 80, 100, and 110 models, which came pre-installed with $1,200 worth of free software (including popular programs like Quicken and ClarisWorks) and undercut Apple’s price by 25 percent. The clones received “very good” to “excellent” ratings from the Mac industry; PC Magazine called the Power 100 “the Macintosh’s identical twin in everything but price.” Then, last October, Power made the leap from cloner to innovator, introducing the PowerWave 604/150—in Kahng’s words, “the fastest desktop PC in the world.”

Now what? Power expects 1996 to be even more frantic than last year. Its cheapest clone yet, the new PowerCurve, with a sticker price of less than $2,000, is about to hit the market. Power plans to become a supplier of parts like circuit boards and possibly even complete computer systems to other computermakers who strike licensing deals with Apple; Kahng says he’d love to see 25 more companies like Power—they would all be potential customers. Later this year Kahng will carry Power’s brand-name recognition overseas to China, South Korea, and other markets that Apple has yet to exploit.

Closer to home, the company is getting unpacked after a big move. With clone production approaching 10,000 machines a month, Kahng relocated his troops in January north of Austin to the town of Round Rock, where he has leased plenty of office and manufacturing space along Interstate 35. This time, there’s carpet on the floors, windows in some offices, and enough room for Power to triple its work force, as expected, by year’s end. Just up the road is Power’s new manufacturing operation, a refurbished factory floor in a rust-colored motor plant where the enormous magnets for the supercollider were to be built. Power will be able to produce nearly three times the number of PCs a month it currently produces.

With such success, of course, come questions. Where will Power get the money it needs to finance its rapid growth? Last June a group of foreign investors, including Olivetti, South Korea’s LG Group (formerly Lucky Goldstar), and Japan’s ASCII, brought the company’s total financing up to $13 million. There is also the possibility of a stock offering. Kahng and Hirvela say they’re in no big hurry to go public, although they have met with investment bankers from several firms. “At this point, we’re more than sufficiently funded,” Hirvela says.

Could Apple abandon its licensing strategy and pull the plug on Power by refusing to grant it a license for Copland, the next generation of the Mac operating system? Not likely. While none of Apple’s licensees are guaranteed the rights to Copland, Apple’s vice president of technology licensing, Lamar Potts, says there is no reason why Apple wouldn’t license to Power again. “They’ve proved to be a good ally for us,” he says.

And what of the competition? None of the other Apple licensees (Radius and DayStar Digital in the U.S. and Pioneer Electronics in Japan) presently pose a threat to Power, but success breeds imitation—and a satisfied Apple now says it will license the Mac operating system almost to anyone who wants it. “You’ve got to believe bigger players will enter the market at some point,” says analyst Scott Miller of Dataquest, a leading Silicon Valley market research firm. “By then, Power will have to be big enough to survive.”

Kahng isn’t worried. He expects his company will be worth $1 billion in five years. “Dell,” he beams, “took ten years to hit $1 billion.”

Alexandra M. Biesada is a former Texas bureau chief for Financial World magazine.