In many respects, Intel Corp. is unlike any other company in the electronic-components industry. Aside from its overwhelming dominance of the microprocessor sector, the company outranks its competitors in one other area: expenditures on research and development and capital equipment.

The numbers are staggering. By the end of 2007, Intel will have spent a whopping $80 billion or so on R&D and capex since only the beginning of this decade. The amount partly explains Intel's resilience over the years and the inability of rivals, including main microprocessor competitor Advanced Micro Devices Inc., to make a serious dent in the company's market leadership. With its main computing market rapidly maturing, though, Intel's management needs to accelerate the process of reinventing the company for new opportunities in the near future.

It will be at least several years--possibly even a decade--marked by disruptive shifts in end equipment demand, new technologies, rivalries and economic changes before Intel's vulnerability to the computing industry becomes a crippling problem. It's not certain, even then, whether the company's deep R&D activities and commitment to new process technologies will be sufficient to deflect the dangers posed by its PC-centric business.

How dependent upon microprocessor sales is Intel? Simply put, if the MPU market goes bust and disappears today, Intel--currently the largest fish in the semiconductor ocean--would become worm bait. Intel's largest business group is its digital enterprise unit, which in 2006 contributed 56 percent of the company's annual revenue of $35.4 billion. However, sales of microprocessors in the DEG division accounted for 41 percent of Intel's total revenue. Also, sales of microprocessors in the second largest group, the mobility division, represented 26 percent of total 2006 sales, compared with 22 percent in 2005.

Intel's commitment to R&D--even during periods when electronics revenue growth turns negative--has helped keep the company at the top since 1992, when research firm Dataquest Inc. first ranked it the No. 1 global chip manufacturer. That strategy also holds the key to its future.

For now, no other semiconductor company can match Intel's ability to spend whatever it deems fit on both capital equipment and product R&D. In 2001, when the entire industry was on its knees, humbled by its worst-ever downturn, Intel only slightly reduced its R&D expenses to $3.8 billion from $3.9 billion in the prior year, while jacking up capex to $7.3 billion from $6.7 billion.

Since then, the company has maintained both its R&D budget and capex spending at above 10 percent of annual revenue. Combined, Intel lays out about one-quarter of its annual revenue on the two.

No other company comes close in actual dollar amount. In fact, many semiconductor vendors, unable to fund the huge costs of new fabs and the attendant R&D expenses, have cut back on internal spending and hitched their manufacturing wagons to foundries. AMD, for example, has a joint process technology development arrangement with IBM Corp. for the 45-, 32- and 22-nanometer nodes. On the manufacturing end, AMD supplements its own internal production with a wafer-sourcing agreement with Chartered Semiconductor Manufacturing Co. Ltd.

AMD's annual R&D budget is substantial for a company of its size, but it is dwarfed by Intel's outlay. In 2006, AMD spent $1.2 billion on R&D, representing a hefty 21 percent of its $5.7 billion revenue for the year.

Nobody can charge Intel with not trying to diversify its revenue base. Microprocessor sales in the two major divisions have been dropping steadily as a percentage of total company revenue over the last three years as Intel ventured into new markets.

During that time, Intel has poured billions into new markets, many of which it has exited--often because it lacks the scale to compete with leaders in these segments, a strategy Intel isn't hesitant to point out.

"We may terminate research and/or product development before completion or decide not to manufacture and sell a developed product for a variety of reasons," the company said in its 2006 annual filing with the Securities and Exchange Commission. "For example, we may decide that a product might not be sufficiently competitive in the relevant market segment, or for technological or marketing reasons, we may decide to offer a different product instead."

The latest component sector Intel kicked to the curb was flash memory. In 2006, the company joined hands with DRAM supplier Micron Technology Inc. to form IM Flash Technologies LLC, a manufacturer of NAND flash products. Intel took a minority stake of 49 percent in the venture and poured in $1.3 billion to secure its interests.

More recently, the company reached out to STMicroelectronics and Francisco Partners, a venture capital outfit, to create a new flash memory supplier. Intel contributed its NOR flash memory assets and other undisclosed resources to the startup for a 45.1 percent stake and $432 million in cash.

Before it exited the flash memory market, Intel also got out of the wireless and digital telephony sectors. Last year it announced a wide-ranging restructuring that was expected to result in the loss of 10,500 jobs.

With all this behind it, what will Intel look like in the future? It's hard to tell. Due to intense competition from AMD, the company has refocused on the core computing business. However, Intel is unlikely to be a one-trick pony. After all, the company has reinvented itself before. Intel's original business was DRAM, but in 1985, the company dumped the memory business and moved on. For now, Intel might be content to squeeze the utmost from the computing market; but you can count on its strong R&D bench to help lead its next act.