Going Separate Ways?

Add Your Comments

The New York Times had an article about Paul Otellini, who is expected to succeed Craig Barrett as head of Intel in about nine months.

What is most interesting (well, at least for us) is that Mr. Otellini is the Intel exec most responsible for the deemphasis on speed at Intel:

. . . [H]e championed what Intel insiders refer to as the “right-hand turn” – breaks with the cherished belief that nothing matters more than ever faster, more powerful computer chips.

Mr. Barrett agreed, acknowledging that Mr. Otellini, more than any other individual, deserves credit for convincing people inside the company that producing a chip that could process data at, say, 3.6 megahertz rather than 3.4, was not nearly as important to their success as making chips with built-in WiFi, thereby saving consumers from having to add hardware to their PC’s. This was the “right-hand turn” started in 2001.

Through most of Intel’s history, every new product followed a simple pattern: the engineers figured out what was possible and then told the marketing department what to sell. The company understood the importance of consumer focus groups, and employed ethnographers to study how people use computers, but their influence was minimal before Mr. Otellini took charge of the chip-making division.

MR. OTELLINI’S user studies led him to make a heretical pronouncement in 2000 at a strategy meeting with Mr. Grove, Mr. Barrett and other top executives. Engineering’s focus on incrementally faster and more powerful computer chips, he said, was no longer the kind of innovation users wanted most.

“The history of the industry was the better-mousetrap syndrome: You build a faster thing and the world will beat a path to your doorstep,” Mr. Otellini said in an interview. “But as the industry matured, that no longer became the best way to look at the problem.” People want built-in security features, wireless connectivity to the Internet and better graphics and audio, he told them.

One need not believe that Intel’s deemphasis on speed is solely due to Mr. Otellini’s persuasive brilliance. Prescott’s heat problems no doubt was a big ally in persuading others at Intel that the old way wasn’t going to work anymore.

But that was then, this is now. This is the guy who’s going to be running the company, so for him, it’s not just a matter of making lemonade out of lemons. He believes this is the right way to go.

And he’s going to run the show.

Whither AMD?

If you’re the typical person reading this article, you don’t want to hear this. You don’t want to hear, “The Age of Speed is over.”

Of course, you do have someone else to look towards. You have AMD, and while they’ve talked some of the talk, too, they don’t seem to be converts.

For the last twenty years, AMD has essentially been a clone of Intel, in all essentials, doing the same kind of things Intel has been doing, building faster and faster chips.

Now what do they do? Do they say “Me too” to Intel’s course change? Or do they stick to the old way and old emphasis and try to become the speed king by default?

Again, I don’t doubt what most of you are thinking. But how many of you are out there compared to all the Joe Sixpacks out there?

For those of you who vote for speed, well, what happens if AMD turns into something that looks like Alpha used to be: high-performance chips at high prices?

I see some hands going down. 🙂

And if AMD isn’t around to temper Intel’s pricing (and it’s not a matter of just price, this could turn into standard vs. optional features) what happens to Intel’s prices?

This is not something that’s going to cause a dramatic change in the next year or so, but beginning after that, we could see the two major CPU companies take much different paths.

To analogize, if these were car companies, Intel is going to put its main efforts into giving people a smoother ride rather than a bigger engine.

AMD will have to decide to either follow Intel, or set out on its own and keep building bigger engines.

The first approach is safer, but it will continue leave AMD as second banana. The second approach is far riskier, with potentially much bigger rewards, but also much nastier, even fatal consequences.

What does AMD do? What happens to us if one company drops out of the speed race?

What happens to us if both of them do?

Ed

Leave a Reply

Your email address will not be published. Required fields are marked *