Intel says it wants to change

Intel says it wants to change

Summary: But why?

TOPICS: Data Centers, Intel

The Intel Developer Forum this week in San Francisco, is Intel's most important event of the year because it's a chance for its top executives to talk about future strategy.

Unfortunately, it's also the time of the year when Apple usually announces new products, and there's multiple conferences in town,  so it's always tough getting attention. But it's especially important for Intel this year because  it has a new CEO in Brian Krzanich, and a new president, Renee James.

It's a coup for Intel's communications team to get some attention from the New York Times in a very news-heavy week. Quentin Hardy spoke with Ms. James about: Intel's Extensive Makeover

The changes include not just products, she said, but a transformation of Intel itself, from a client-server era company to a cloud era company. It is a work in progress, she said, that has involved consultation with Google, Facebook and other cloud companies about how they work. Everything, including how employees design products and compensation, is being rethought, she said.

“This is bigger than any device,” Ms. James said. “It’s a cultural transformation. We’re trying to be thoughtful about it.”

The company's product line is changing too:

Intel’s chief product announcement to the developers was a new small, low power line of sophisticated chips aimed at wearable computers and sensors connected to the Internet. In other words, a product for a world where Internet-based computing intelligence is deployed anywhere and everywhere.

Foremski's Take: I've covered Intel as a reporter for more years than I have fingers and toes, and it still feels the same, and has many of the same people.

Cultural change is slow and it's even slower if you don't take a chance at changing things when the chances are there for the taking.

Intel's board decided to appoint insiders Brian Krzanich and Renee James when CEO Paul Otellini resigned in May, yet it could have chosen outsiders. The excuse being that Intel's culture is so solidly set that only insiders would know how to deal with it!

I'm sure that Intel would like to change its culture but why? It's not a good idea to mess with something that has been so successful for such a long time.

Intel won the microprocessor war — the most lucrative chip category. It has the lowest manufacturing costs in the chip industry for the highest premium cutting-edge chips — you can't ask for more than that.

Armless profits…

Yes, Intel has been unable to wrestle the tablet and phone market from UK-based chip design firm ARM, but that market would do little to add to its fortunes. 

Intel says it wants to be in every segment of the market but it doesn't need to be because its most profitable chips are its server chips.

The more edge devices, the more sensors, the more Google goggles are sold— the more need for IT data centers. If the data center changes without Intel inside, then Intel would need to change but there's no sign of that.

Last week, Diane Bryant, senior vice president and GM of Intel's Datacenter business group announced a large number of new chips aimed at running cloud applications, data storage, and network equipment, and cementing its place in IT installations even further.

"Intel Inside" the data center is what it needs to defend and "Intel Outside" in wearable computing, etc, is nice to have but will do little for Intel. Its competitor ARM has profits that are rounding errors on Intel's financial results.

The best thing for Intel is to evangelize products such as smartphones, tablets, and wearable computing because it'll drive sales of its lucrative server chips.

But Intel's culture is highly aggressive and it cannot stand to lose in any market no matter how tiny —  and I don't see that changing anytime soon.

Topics: Data Centers, Intel

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • Re: Intel won the microprocessor war

    You realize it's only number 3? ARM is way ahead at number 1, outshipping x86 10:1. Even the number 2 architecture ships 50% more processors than x86.

    x86 is an overly complex and expensive dead end. The only thing keeping it alive is the need for Microsoft Windows binary compatibility to run legacy apps on traditional PCs. The new mobile markets don't care about that.
    • ARM is way ahead!

      You do know ARM doesn't actually make any processors, right?
      Michael Alan Goff
  • The problem is the architecture.

    The Intel architecture is bloated, inefficient, and inherently slow - causing a huge amount of overhead to implement, and that translates into a lot of power.

    The underlying RISC processor is pretty good though. Unfortunately, to support the poor architecture requires boatloads of translation hardware, reordering hardware, and clumsy memory access barriers (which causes even more bloat).

    ARM has a cleaner architecture, easier to extend, simpler to speed up.

    Intel knew this when they got shafted by MS with the Itanium. MS was supposed to support the new architecture, but pulled out before anything was really ready.

    Only Linux was usable when the chip was released. Was the first chip a success? only as a proof of implementation. It took several iterations before it was actually faster.

    ARM has been through 9 design iterations to reach a point that it is now challenging Intels hold on performance. It will eventually surpass it for speed AND power. I think what it needs next is a faster shared memory bus.
    • Ermmmm ... no.

      Microsoft has supported Itanium from XP/Server2003 through Server2008 R2 in 2010. Longer, in fact, than RedHat supported Itanium until RHEL6 in 2009.

      Itanium wasn't broadly accepted in the market due to it's price, performance, power and complexity, but it's true nemesis was, in all reality, AMD's 64bit addessing extensions that were later adopted by Intel and which now power almost all AMD/Intel 64bit processors.
      • Re: Longer, in fact, than RedHat

        1) Linux is not (just) Red Hat
        2) Linux remains fully-functional on Itanium, while Windows is just a crippled subset
        3) As jessepollard pointed out, Linux was first on Itanium anyway.
  • Long live AMD

    Great article but its nuts paying premium for an Inside Intel sticker on the case. Hope AMD get their act together in 2013-14 and bring a little competition to the market place. I still think their hidden weapon is in the low powered low end processor.
    Darko Gavrilovic
  • Reimagining Intel.

    Like Microsoft. See the results.