Intel's Longshot Run at the Texans on DSPs

Much hullabaloo over the joint announcement by Intel (INTC:Nasdaq - news - boards) and Analog Devices (ADI:NYSE - news - boards) in New York Tuesday that they're finally ready to talk about the architecture of their joint-venture design for DSPs, or digital signal processors. But investors in Intel and Analog Devices shouldn't look for big returns from this initiative anytime soon.

Texas Instruments (TXN:NYSE - news - boards) owns the market for DSPs, with greater than 50% market share. Lucent (LU:NYSE - news - boards), Motorola (MOT:NYSE - news - boards) and Conexant (CNXT:Nasdaq - news - boards) are serious competitors, too -- as well, of course, as Analog Devices. None is going to give up market share easily. Moreover, Texas Instruments' long-term commitment to DSPs and to market-development work for its chips, provides an exceptionally high barrier to entry for new players, no matter their other strengths.

Digital signal processors are already important, and are likely to be even more important in the years ahead. The name has always been confusing -- in a sense, all central processing units process things digitally. But think of DSPs as the monomaniacs of chips: They focus on one thing, and only one thing ... and do it verrrrry fast.

DSPs take analog input and convert it to digital signals. That's essential to all sorts of nondigital things, from the human voices "heard" by cell phones to the images digital cameras "see" as they take and save your photos.

Success at that narrow but critical focus grows out of both basic chip-architecture issues, and also from the programming tools used to set them up. In fact, of course, digital signal processors don't "know" anything special about the tasks to which they're assigned, in digital cameras or cell-phone handsets. Those application-specific smarts come through programming by the engineers who specify DSPs ... and Texas Instruments' programming tools, resulting from its relationships with DSP-specifiers in that active market-development work, are very strong.

An example of TIs' understanding of the needs of the engineers who choose its DSP products: its online DSP Developer's Village , a rich Web site full of information on choosing, sampling, programming and buying its digital signal processors.

Intel and Analog Devices are claiming several advantages for their new DSP architecture:

• Faster time to market, thanks to easier and quicker DSP programming. The new software tools they will soon deliver allow programmers to work in their familiar C and C++ languages -- the lingua franca of the tech world -- rather than having to know the specific programming tools of other makers' DSP products. This could save months in product-development time, and if the tools work as advertised, could be a major inducement to engineers to move to the Intel/ADI products.

• Lower power drain, thanks to "Dynamic Power Management" support in their new DSPs, allowing onboard software to monitor and adjust both voltage needs and operating speed, depending on what the DSP is doing at the time.

• An architecture designed to support next-generation handheld products running such applications as video, speech recognition, handwriting recognition and image capture and transmission.

This is right down DSP engineers' alleys, right off their Christmas wish lists.

But how soon can the Intel/ADI venture actually deliver? Tuesday's announcement comes a long -- maybe, too long -- 21 months after the joint venture was announced. And their new DSP chips are still in the pipeline, not in the mail. Look for emulators and programming tools soon, samples by midyear, and actual production chips "over the next 12 months," in the words of an Analog Devices spokesman.

And will engineers prefer to continue working with the familiar -- by one guess, more than 80% of the world's engineers using DSPs in their designs already know the TI DSP architecture and its programming tools -- or will they pay the price in training and frustration to get at the new chips' features?

This is a big play for Intel, a fact confirmed by its willingness to partner with Analog Devices. This could soon be a billion-unit market, and Intel clearly wants part of that. More, the new "Barrett inside" Intel has been growing increasingly frustrated watching the world move away from its core competencies and existing product lines, toward new worlds in which the company has not been a player. Partnering with Analog Devices was a fast path to market decision Intel hopes pays off big.

I think it was a good bet. By itself, this isn't a move which could transform Intel -- something Intel managers would argue long and loud isn't needed, anyway. But it could add another revenue stream, with considerable potential. Yes, Intel has to share that revenue with Analog Devices. But getting to market sooner, with a partner experienced in this area, may have been the key decision for Intel.

The other DSP players are hardly shaking in their boots. But for both Intel and Analog Devices, this is a nice mid- to long-term play. Investors in the silicon universe will have to factor time to payback for these new products into their analyses of both Intel and ADI, and are likely to see this as an important step for both companies.

---

It's hard to know what to say about Jim Cramer's decision to retire from Cramer Berkowitz. I wasn't exactly floored -- if you knew Jim, you could see this coming -- but I was surprised by the timing. Everyone wants to go out on top, but the counter-argument is that when you finish one great year, you really want to put one more notch in the gun.

I admire Jim generally, and I particularly like his attention to his family's interest in seeing Dad more. To the extent that this move contributes to a calmer existence for Jim, I'm all in favor ... but I still have trouble putting the words "calm" and "Cramer" in the same sentence.

Thank God we'll still have Jim's contributions here on RealMoney.com and TheStreet.com. I expect we'll see a lot more of him on TV, too, and now maybe Jim will let the public see more of that sly wit, which has always underlain his frenetic, often bombastic style. It's a heckuva combination.

I'm not shy about saying I love Jim. He's one of the great money managers, and probably more attuned to today's markets than any other. Will he be able to stay away? Beats me ... but I expect to see him front and center on the investing stage for a long time to come.

---

Jim Seymour is president of Seymour Group, an information-strategies consulting firm working with corporate clients in the U.S., Europe and Asia, and a longtime columnist for PC Magazine. Under no circumstances does the information in this column represent a recommendation to buy or sell stocks. At time of publication, neither Seymour nor Seymour Group held positions in any securities mentioned in this column, although holdings can change at any time. Seymour does not write about companies that are, or have been recently, consulting clients of Seymour Group. While Seymour cannot provide investment advice or recommendations, he invites you to send your feedback to Jim Seymour .