If the WiMAX solution is a success, Intel's situation in Israel will be pretty good, to put it mildly" David (Dedi) Perlmutter, senior VP and general manager of Intel Corporation's Mobility Group, and the most senior of the company's Israeli executives, talks to "Globes" about the mistakes the giant has made in recent years and his hopes for the new developments. Shmulik Shelah 27 Jun 07 16:13 It?s been a successful year for David (Dedi) Perlmutter, senior VP and general manager of Intel Corporation's (Nasdaq: INTC) mobility group, and the most senior of all the Israeli executives at the global chip giant. It has been exactly a year since the company launched Woodcrest, the code name for its dual core processor for the server market. The launch was a fairly significant event, since it represented Intel's transition from the Pentium 4 product family, its most recent product line for PCs and servers, to the Pentium M family. "The company is happier with the new solutions and I don't think that's a big secret," says Perlmutter. "The new product line changed our balance sheet in terms of products, business, image, and market segment, chiefly in servers. I think that right now all the pressure is on our competitors, as to when will they launch their own products." Perlmutter has a reason for feeling good about the sense of satisfaction at Intel. The Pentium M architecture, which was conceived and developed for laptops under his leadership at the company's center in Haifa at the beginning of the decade, has also become Intel's operating strategy for its product lines for the server and desk top market. This change could, perhaps, have been responsible for Perlmutter's assumption of responsibility for Intel's entire architecture development activity - not operations-wise, but primarily responsibility for decision making.
Globes: What was it that actually caused Intel to lose such a large market share?
Perlmutter: "A large part of the blame for what happened can be attributed to what Intel didn't do. Between 2000, when Intel launched the Pentium 4, to 2006, when we launched the new architecture (Core 2 Duo), the company did not launch a single leading product. We did launch all kinds of more advanced technology, but there was some sort of wall. In retrospect it turned out that Pentium is a difficult architecture to develop. It had an inherent complexity because it had a high output, and a complex structure that made it difficult to develop the next generations on it. In the final analysis, the true test is not in a specific product itself but in the entire product line, and Intel had difficulty with this one. We reached a kind of wall that was difficult to climb."
"Intel entered the cellular field fairly late and we even had to compete with companies such as Texas Instruments, Qualcomm, Nokia and others over copyright, as well as the entire issue of customer relations. It was an excellent schooling but perhaps too expensive, and we learned a lot of lessons that we're now implementing in all matters relating to WiMAX and mobile Internet solutions. But when you arrive late, you need to have something unique in order to overtake the competitors and we didn't have it. Our situation was "we have it too." You find it hard to compete in a situation like this. We didn't differentiate ourselves in the field in terms of technology or costs. Marvell could have an excellent business from this."
All the commentaries have referred to Intel's insistence on operating in smaller markets, which is why it failed. What will happen in the WiMAX market?
"Because cellular was a field that was far from us, I think that it took us a long time to realize who was against who, and what sort of game they were playing. It was a good learning experience, but getting in there was not a trivial matter. I think that we have an advantage in WiMAX, because we have a lead over the competitors. We have a chance because we're the ones that started the process. We have a very strong group in this field and we have the ability to deliver processors that can run Internet far better than the others. In buying DSPC when we did, we wanted to enter a market that we were unfamiliar with at that time."
What role do you see for the Israeli development center in Intel's future plans?
"The center in Israel is currently in an excellent position. The Core 2 Duo was developed in its entirety here, and the staff here are working on development right through to 2010. The wireless and WiMAX technologies were developed in Israel as well. We have more than 600 people working here on WiMAX and WiFi and I think that this is the largest center in the world in this field. If the WiMAX solutions are a success on the same scale as our processors were, Intel's situation in Israel will be pretty good, to put it mildly."
Published by Globes [online], Israel business news - www.globes.co.il - on June 26, 2007
? Copyright of Globes Publisher Itonut (1983) Ltd. 2007
Looking back, Perlmutter says that Intel changed the timing of the release of new products. "We're currently working on a model called "Tick-Tock." We realized as far back as 2004 that we had to make significant changes to the product almost every year. So it works out that we base our progress on new architecture every two years, and follow that a year later with a change to our production capacity to something more advanced. Next year, for example, we'll be moving the Core 2 Duo to 45 nanometers."
"The coming years are critical for WiMAX"
In recent years, Intel has invested considerable resources in its entry into the broadband wireless telecommunications world. It chose to make its entry using the WiMAX 802.16e technology standard, which it plans to install in every processor it supplies. WiMAX provides solutions which, Intel's management feels, are more successful, both technologically and performance-wise, than other available technology solutions on the market. Perlmutter notes that WiMAX also solves the problem with intellectual property, which has caused fierce battles in the wireless market with Qualcomm Inc. (Nasdaq: QCOM), for example.
There is a still a serious debate as to whether there's any justification at all for WiMAX.
"One of the things that we did differently is that WiMAX is an open solution and you don't have to pay royalties to a certain company. If you compare the pace of the development of the open standard to that of the closed standard, the difference is huge. This is also true of WiFi. Today, every country has different frequencies for cellular handsets. I carry a few telephones with me in my briefcase, but that's not right. If I want to put something in my laptop, it is supposed to work everywhere. I can't say for certain that WiMAX will be a success."
I find it hard to believe that you did not bet on a winning horse in this field.
"Unfortunately, Intel has invested in a few things that were not successful. I am not saying this as an apology, if you don't try you'll never succeed. Either way, I find the increasing opposition quite encouraging. Two years ago, you wouldn't hear a bad word said about WiMAX, simply because no one could care less. Now it's beginning to take shape, and there's a chance that it will happen. The next two years will be critical for WiMAX." Intel plans to implement its investment in WiMAX in the first half of 2008 with the launch of combined WiFI and WiMAX communications cards that will be integrated in laptops.
Talking of failed investments, one good example is the investment Intel made in cellular processors (the Xscale product family). "Intel saw after a few years that it wasn't going to be number one or two in this field, and it has a strategy of not continuing to invest if it can't manage to reach these targets," Perlmutter says by way of explanation of the company's acquisitions in 1999-2000, most notable of which was DSPC, which it acquired for $1.6 billion. The entry did not turn out that well for Intel, and a year ago the DSPC unit was sold to Marvell Technology Group (Nasdaq: MRVL) for $600 million.