Looking for some technology companies with mouthwatering returns but without big names, stadium sponsorships, or dots or coms?

Here's one way to find them. Go to a mutual-fund screening Web site. Look up the biggest and best tech funds. Study some of the stocks to which they flock, and buy the ones you like--the stocks, not the mutual funds.

Cheating? I suppose.

I mean, I would never confess as I mutter "Xilinx" under my breath to impress a hot-tip seeker at a dinner party that I had found this stock on some mutual-fund Web site. But I can't believe fund managers don't look at one another's sites, too.

In fact, far from the route to "hot tips," peeking at successful funds' portfolios is a way into a world 90 percent to 95 percent occupied by institutional investors. These are folks who spend a lot of time and money finding companies that aren't household names, many of them companies that do business only with businesses, which is why they have little identity with consumers. In the technology world, these companies are an important part of the infrastructure that makes the sector so dynamic.

Have you heard of Xilinx Inc.? It's share price has appreciated 283 percent during the past 12 months. That's better than Cisco Systems Inc. (131 percent) or Lucent Technologies Inc. (83.3 percent) and roughly 90 percent of all other stocks for sale.

How about Altera Corp.? Ring a bell? Fifty-two-week low: $14.75; 52-week high: $53.81. Net income since December 1994 has gone from about $10 million to nearly $150 million. (Amazon.com has gone from zero to minus $100 million since December 1995.)

Xilinx and Altera--about which more later--were among the most frequently listed lesser-knowns among the top holdings of a dozen big technology mutual funds I checked. (They are, of course, well known in their industries and to the professionals. But they are lesser-known or even unknown to us and our friends, unless you've got some special friends.)

A few samples of the other lesser-knowns that showed up most frequently: Tellabs Inc., Solectron Corp. and Sanmina Corp.

I do not recommend these or any other stocks. They are meant to illustrate what's behind the scenes in big industries providing critical materials or services. While they're not famous, what they do is help make famous companies famous, and sometimes make the products famous companies sell under their own names.

Solectron and Sanmina, for example, produce some of the products sold under the nameplates of household-name companies. But as investments, they might be cheaper than the big names, depending on your timing. Cisco has a price-to-earnings ratio of about 113. Solectron's is 65; Sanmina's, 56. (As you can see, none of these is a bargain stock.)

Let's look at Xilinx. It's 15 years old and based in San Jose. It has had significant earnings-per-share growth every year since 1990 (from 6 cents to 84 cents over that period) and revenue last year of $662 million. Among other things, it makes programmable logic devices.

Who needs programmable logic devices? Most of the big telecommunications and router companies in the world.

When companies such as Lucent or Cisco want to develop, say, a new switch, they have two choices. They can order a chip custom-made, complete with the exact circuitry necessary to make the switch operate.

Or they can purchase the equivalent of a generic chip from Xilinx or Altera and program it themselves, using appropriate software from the same companies.

It's the difference between buying a cassette tape with the music already on it and buying a blank tape that you put the music on yourself, explained Ann Duft, public relations manager at Xilinx.

In the case of chips, a custom-made one costs less but takes longer. The programmable chip costs more, but the company gets it faster.

What makes companies such as Altera and Xilinx so important, said Eric Rothdeutsch, a Merrill Lynch & Co. technology analyst, is that the programmable device gets that new switch to market "much more quickly."

"And if you can get to the market three months ahead of the competition, that's where you make the nice profits. . . . It's the first-move advantage," Rothdeutsch said.

Plus, he noted, big companies continue to want to focus on what they do best--contracting out other activities to companies that do it better so that companies such as Xilinx, Sanmina and Solectron have a very good chance of continued strong growth.

Xilinx has roughly 30 percent of the market share for programmable logic devices; Altera has another 30 percent.

Sanmina and Solectron benefit from the same outsourcing craze. They are "contract manufacturing companies" that make things sold under the names of other companies, the way Toyota once made cars sold under the Chevy name. Sanmina, also in San Jose, had total sales over the past 12 months of about $1 billion; Solectron, $8.3 billion.

The methodology for finding these stocks--looking at the holdings of mutual funds--is notnew. It's just much, much easi-er than it used to be. Go to www.valueline.com or to www.morningstar.com, which lead you to funds, to the top 10 holdings of the funds, and to hyperlinks to those companies' home pages all in remarkably short order.

Now, as I said, you could buy the funds or the stocks. Each has certain advantages. If you were to keep $20,000 in some of these mutual funds, you'd pay a fee, annually, of $300 and on up, plus take a tax hit every time the fund had a capital gain (unless, of course, the mutual funds were held in a tax-deferred retirement account).

If you bought the five from an online broker, you'd pay between $35 and $100 one time. The timing of a sale--and a capital gain or loss--would be your business, not some fund's.

On the other hand, the fund offers diversity. The stock that would be 20 percent of your portfolio would be, maybe, 2 percent or 3 percent of the fund's. If something went wrong, the fund could take the shock better than you.

Buying stocks heavily owned by institutions also has its pros and cons. The pros are that "the institutional investor is a more knowledgeable investor, doing a lot more homework and really deep analytic research," Rothdeutsch said. "Their goal is to make money for clients. They go visit the companies and are much closer to the source of the data."

If they keep buying, which many do, it helps keep the stock trekking up.

On the other hand, institutional investors, always worried about quarterly bottom lines, are often the first to bolt at negative news. And have no doubt, they will have that news before you do and bolt before you can, leaving you with a depreciated stock. You have to be prepared to hold on if necessary, feeling lousy about your net worth, hoping for the recovery.

Yes, I got these names from mutual funds. To the charge of benefiting from someone else's homework, I plead guilty. We should all do our own research on stocks, of course.

But you have to have a stock to research before you can research a stock. And this is a good way to find one.

Fred Barbash (barbashf@ washpost.com) is The Post's business editor. Of the stocks mentioned in this story, he is a long-term holder of Cisco and Lucent.