Paul Otellini on Intel and the PC's Evolutionby
Since taking over as CEO in 2005, Otellini has focused on pushing Intel beyond the PC. He talks with Ashlee Vance about the high-stakes chip-making game and the company’s push into mobile devices.
How much does it cost to build a chip-manufacturing plant these days?
North of $5 billion. And in order to run those efficiently and pay for them, you need to have $8 billion to $10 billion of annual revenue out of that factory at a reasonable margin to pay for it. There are fewer of them being built, and fewer people can build them. You’re dealing with levels of purity of chemicals and air and precision and in temperature and in rinse and in everything that’s involved in the manufacturing process that have never been invented before. If you think about the output, the things that we’re building are the most complicated machines that human beings have ever built. You kind of get what you pay for. If you look at the last generation of technology that we just released, we reinvented the transistor, the first major reinvention in 60 years or something. That fundamental invention took us 10 years of R&D to do, and it’s not going to be reverse-engineered anytime soon. So it’s not just a matter of having a large bank account.
Your company keeps building plants in this country. Microsoft and Google are getting into the hardware game. Tesla Motors is assembling its electric sedans in Fremont, Calif. Are we seeing a hardware renaissance, possibly even a manufacturing resurgence in the U.S.?
I could draw a scenario where that happens. There are a lot of places on earth where people are uncomfortable deploying their capital, so the United States is actually a pretty safe haven. The risk of expropriation is essentially zero. The currency is stable. The infrastructure is good. The electricity costs are relatively low. The workforce is exceptional. The work ethic here is second to none. So you look at those things, and all things being equal, we could be a haven for capital and for manufacturing if we did some things right in terms of permitting the factories and competitive global tax rates.
What did you make of Microsoft deciding to make its own tablets?
It’s one more customer for us.
How do you react when people say we’re in a post-PC era?
We’re moving from the era of the personal computer to the era of personal computing, and that means that there are going to be computers all around us and in different form factors. The PC is not going to go away anytime soon, if ever. It’s going to continue to evolve. Right now, it’s the most powerful tool you can have, but it doesn’t mean that there won’t be tablets or phones or even connected cars. The beauty of all these things is, if you get it right, the sum total of them has more value than the individual devices. That’s the model that we’re working toward.
Has the pace of technological change sped up?
I’ve seen a lot of changes in my 40 years in the industry. I saw the mainframe get eclipsed by the minicomputer, which got eclipsed by the PC. I saw Intel-based servers take on supercomputing characteristics. We saw the PC move from a brick to an ultrathin notebook. You saw phones get smart. This is evolution. I’m not sure there is anything revolutionary here. I think every generation tends to discount the stuff before them. And that reminds me: It was that famous Ronald Reagan story that when somebody came up to him and said, “You know, you guys have it so easy today. You’ve got all these phones and faxes and things.” He said, “You forget one thing, my son. We had to invent all those.”
Had we not done the basic work in microprocessors, for example, none of this would be possible. While there are some really interesting moments where you see great leaps forward—and I would put the iPhone in that category—it’s not like it was the first thing. I mean, you see this being argued out in the IP courts today of who had a phone that was similar to that. Who had swipe gesturing 10 years ago, those kinds of things.
You guys have been churning out record revenue pretty regularly. But people aren’t paying as much attention to your strengths, like high-end chips that go into data centers. Critics are more fixated on, “Gosh, Intel missed this whole smartphone surge.” Is that fair?
We’ve had a very good last few years. Part of it is that the analyst community has their own agenda. They may be long on something and short on something else. We tend not to worry about that. We just tell our story to them. There have been believers all along. They have made a lot of money, and the people that invested in us at the trough at whatever it was, $12, a few years ago have more than doubled their money, and they’re happy campers.
Now that you do have smartphone chips coming out, is Intel finally going to come roaring into this market?
I have characterized this as a marathon, not a sprint. The most important thing for us is to continue to grow our presence in computing, in personal computing, and in the data center, and then over time get stronger and stronger in devices.
Google and Microsoft and Amazon and Apple are all going in very similar directions, developing both Web services and devices. Have you ever seen a head-to-head battle on such a giant scale?
Go back to the mainframe. There were six giants. Then go back to the minicomputer days, and go back to the early days of the PC. There were four or five big companies. It eventually became Dell, Compaq, IBM, and now you’ve got a different top 10 list. The ones that have the ability to create lasting success do so through scale. That scale gives you all kinds of benefits that allow you to perpetuate that growth and that position. I think you’ve seen that now with the first-mover in search, the first-mover in social networking, to some extent the first-mover in devices or consumer devices like Apple. They have achieved sufficient scale, all, to be able to now participate more broadly than their initial offering. And I would put Intel in that same boat in terms of our ability to leverage the investments and scale we have into new areas.
If you could go back to when you were first coming on as CEO and give yourself advice, what would you say?
Oh God. I don’t know. I wish I had been more prescient about the economic downturn, but that was not when I first started.
What will computing look like in five years?
I think it’s likely to be an extrapolation of what we have now, which is that these devices get smarter. You’re going to have a multitude of devices. They’re all going to be interconnected. They’re going to be more speech-enabled. They’re going to be gesture-enabled, and they’re going to have commonalities between them such that they work for you as opposed to you having to deal with each of them as independent entities. I mean, it’s closer to Star Trek than we are today.