Intel's fab flexing may pay off big amid capacity shortages

Intel's fab flexing may pay off big amid capacity shortages

Summary: Qualcomm and Nvidia need more supply from their manufacturing partners. Intel's bet on integrated manufacturing and design could become a huge advantage.


There's a semiconductor manufacturing capacity problem and Intel appears to be one of the few players that can control its own destiny.

For years, Intel has preached that its manufacturing prowess---it is one of the few chip makers that has its own fabs---is a competitor differentiator. Indeed, Intel and its tick-tock cadence produces smaller more powerful chips at a steady rate.

Intel's bet on manufacturing appears to be paying off now. Why? A bevy of competitors are wrestling with their contract manufacturers to increase capacity.

For instance, Nvidia can't get enough 28-nanometer Kepler GPUs to meet demand. Nvidia and its fabrication partner TSMC failed to plan correctly.

On a conference call with analysts, Chris Evenden, director of investor relations, was asked repeatedly about Nvidia's ability to meet GPU demand.

Evenden's reply:

There's just not enough capacity and the planning process when we think about these new nodes just simply needs to be better and I don't think there's any way around that. In combination between us and TSMC, we under-planned for the supply of 28-nanometer and we need to fix that in the future.

Sound familiar? It should. Nvidia sounds a lot like Qualcomm.

Paul Jacobs, CEO of Qualcomm, said in April that the company was also faced with a 28-nanometer chip capacity problem.

We are seeing very strong demand for our industry-leading MSM8960 and other 28 nanometer products. Although the manufacturing yields are progressing per expectation, there is a shortage of 28 nanometer capacity, and at this stage we cannot secure enough supply to meet the increasing demand we are experiencing. We are working closely with our partners to bring additional capacity online; however, the constraints on 28 nanometer supply are limiting our potential revenue upside this fiscal year.

The common thread: These technology companies in most cases only have a few manufacturers to run to---TSMC, Samsung and Globalfoundries, which is AMD's old manufacturing arm.

Complicating the capacity issue even more is Apple. Matt Cleary, an analyst with Primasia, recently outlined Apple's foundry options. To date, Apple has largely relied on Samsung to make its semiconductors. However, Apple is looking to diversify away from Samsung, said Cleary. Samsung and Apple have repeatedly dueled in court.

The problem for Apple is that it only has so many partners to work with. Intel, TSMC and Samsung are the only ones capable of making Apple's next-gen chips in volume. According to Cleary, Apple would have to rewrite some of its iOS to use Intel's manufacturing process since the chip giant is a no-ARM zone.

That reality means that TSMC is the likely Apple partner going forward. TSMC will then have to build enough capacity to accommodate Qualcomm, Nvidia and Apple. Guess who wins in that pecking order?

Cleary said in a research note:

We believe that only Intel, TSMC and Samsung have a viable shot at being selected as Apple's next foundry partners. UMC clearly lacks the scale and leading-edge process level in order to compete in this league, while GlobalFoundries probably still has to prove itself following the public divorce from AMD. TSMC is pulling out all the stops in order to win Apple's business, including hiring teams of IC design and system level engineers, accelerating its 20nm process development....

It seems most likely that TSMC will have the inside track to become Apple's next foundry partner—once TSMC gets 20nm up and running. The broader question will be whether Apple will prove lucrative enough as a customer to justify all of these efforts.

Add it up and you may have the bulk of the mobile industry moving through one manufacturing partner---TSMC---in 2014.

Intel's advantage here---even though it is a bit player in mobile today---is that it controls its own manufacturing and therefore its destiny. Architecture squabbles with ARM aside, Intel could gain traction in mobile simply because it can deliver the manufacturing goods.

The chip giant did a lot of talking about its manufacturing cadence during its analyst meeting this week. Evercore Partners analyst Patrick Wang said that Intel was "fab flexing."

Indeed, Intel CEO Paul Otellini spent a lot of time talking about manufacturing. Otellini argued that the semiconductor industry was at an inflection point. The "fabless" model---companies that design chips and then outsource manufacturing was faltering. Otellini's main point: Design and manufacturing should be one.

Speaking at Intel's analyst meeting May 10, Otellini said:

When design and manufacturing are deeply integrated you get a spiral of advantage. We can decide where we want to optimize (to lower costs). If you're a foundry and fabless company there's an economic tension there.

You've seen a consistent cost per transistor improvement from Intel. This curve from the foundries is difficult as invention gets harder. My contention is we'll increase our capabilities and advantage going forward.

Otellini said a new state-of-the-art fab will run companies $10 billion. That investment will raise costs for foundries and prices for fabless chip makers.

Not all analysts were completely sold. Piper Jaffray analyst Auguste Gus Richard said:

Intel highlighted its manufacturing prowess as its key differentiator at its analyst meeting. Management spent time emphasizing its fab capabilities in an era where it believes the foundry model is faltering. However, we believe that the law of diminishing returns is beginning to overtake Moore’s Law. Moreover, Intel is expected to shoulder more of the development costs as one of the few remaining practitioners.

Richard makes a solid point on Intel's development and capital costs. However, if the foundry model does falter, it is very easy to pick the winner---Intel. Perhaps, Intel's manufacturing spend becomes an issue, but given the shortages faced by Nvidia and Qualcomm it looks like the chip giant's bet could work out nicely.

Topics: Processors, Apple, Hardware, Intel, Samsung

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
  • Artificially cause shortages, ramp up prices and pocket the dough

    It's hardly a new tactic...
  • Very interesting

  • And - - - -

    Possibility of "outsourcing" looms again???
  • shortages and cycles

    Shortage today (actual or intentional) will deflate prices later, as they'll have missed selling the chips *today*. Once everyone makes new commits, plans, and executes, then you're left holding outdated chips.

    The real winners are the ones who can alleviate the shortages FIRST.
    They'll be saving other companies, and giving them strength to succeed, whereas those who cannot (or hold out for higher prices) will be left at the roadside.

    If Intel offers to help Apple divest/branch from ARM (one product at a time, and generate tools to migrate the rest of the dies later) and Apple makes a new line with a different chipset, Apple will have even more price bargaining power with Samsung and TSMC, and Intel will get yet another mobile fab line running full tilt.

    Also, tell me again, who makes the Apple iMac CPUs - right now...?
  • But it leads to overconfidence

    Yes, it's true that controlling it's own dedicated fabs gives Intel an advantage. But it also leads to overconfidence and product mistakes.

    For years Intel told us that it's "next" generation Atom processor would overtake the ARM. Even though the x86 architecture is bloated, power hungry, and just not suited for mobile devices, Intel always assumed that having the latest generation fabs would allow it to produce devices that were just as fast, cheap, and energy efficient as ARM. So far, that hasn't happened over several mobile processor generations. But instead of throwing in the towel on the x86 as a mobile processor and designing something different, Intel just keep doing the same thing over and over because they thought their fab technology gave them such a huge advantage.

    Now Intel is about to release its latest generation of mobile devices, but even if they got it right this time, they still have probably lost the war. Meanwhile, there was a lot of collateral damage in the industry. Microsoft, committed to Wintel, wound up with a mobile operating system that nobody would use because the Intel processors required were too power hungry and expensive. They've had to retrench, expending massive resources on Windows RT, but they've probably lost the market window. And despite switching to Intel for its laptops and PCs, Apple saw the mobile problem early on and spent hundreds of millions for its own ARM design house. Apple is now so big they can spend whatever it takes to get themselves first in line at any fab producing ARM processors.
    • The war is never over..

      unless you give up or die.

      Intel is not dead yet and does not seem to have given up.
      • Don Quixote never gave up either

        And look how far he tilted.
    • Do you know what you are talking about?

      Which mobile OS from Microsoft are you referring to? If your definition of mobile devices is smartphone, WP7/7.5 do not run on Intel processors for smartphones like the Z6xx and the Z25xx. Only Android and Moblin/Tezen can run. If your definition of mobile devices is tablet PC, Win7 is only good for powerful computing device which has a lot of on-board resources (DRAM, HDD). If your definition of mobile devices is notebook PC, Win7 is very successful there. Media tablets based on iOS and Android are capable of only applications which does not require much local resources on-board.

      If you pay enough attention to hardware design, or remember enough of the lectures on computing electronics given in the college days, you should know well the following two facts: 1) nothing come for free, 2) comparison not conducted at the same context/level often gives deceiving results. Today, Cortex-A15 provides far less absolute computational power than the latest incarnation of mobile-x86. For the relatively simple tasks which most consumers expect the media tablets and smartphones to handle, Cortex-A15 is probably excessive. Once ARM architecture grows up to match the mobile-x86 at the time to tackle the more demanding tasks, DON't expect the ARM core to consume much less power than the mobile x86! This is simply the law of Physics. In fact, if Intel is able to maintain the process superiority over TSMC/Samsung/GF, ARM core could even consume more power than the mobile-x86!
  • What does it mean to America?

    Re: "The fabless model - companies that design chips and then outsource manufacturing - was faltering. Otellinis main point: Design and manufacturing should be one."

    I've long said that a corporation that breaks the feedback loop between design/engineering and production was trading a fast buck now for obsolescence tomorrow.

    But the question from America's perspective is will this approach of closely-coupling design and production accelerate (as opposed to "begin") the offshoring of design/engineering, and so finally render the great lie of inequitable free trade - that America would become a "service" economy that generated the technical and engineering "bright ideas" for manufacturing elsewhere - visible to the American public?

    Will the American people finally understand that everything that made America a leader in the 20th century is being - if not has been - offshored to make a few a fast buck now and the safety and security of the United States of America and the 'general Welfare' of the American people be damned?
    • The high-tech economy will not stay in America - now or in future!

      Macro-economics as taught and researched in American universities sees the free-trade principle as the basis of all economic activities across the world. But free trade as seen through American eyes is different from free trade through the Korean or Chinese or Indian eyes. It is also impossible to make them convert to the American free trade due to local socio-economic-political considerations in each economy. So that leaves America and partly Europe as the free trade practitioners. And actually only America.

      Average standard of living in America will go down because high-tech companies need to have an Asian strategy just so that they can use the massive low-cost production and manufacturing and now design power of those workgroups. Apple is an example of a American design/Asian manufacturing systems company (indirectly through Samsung and Foxconn). IBM can come close to it through some of its relationships in China. And so do HP and Dell but these other 2 companies are not in new technology businesses or markets. Intel has the capability but it is no systems company. No matter how much fabrication + chip design prowess they may have, they will not match Apple's or IBM's systems capabilties in certain markets. Microsoft is a pure software company (excluding Xbox) and all of Google, Facebook, Amazon and Paypal-Ebay are Internet based which are commodity systems companies. That last model is different from the first ones. Point being - Intel may look better than Nvidia but more importantly one of these ARM CPU companies has to be acquired by Microsoft or Google for them to make optimized systems like Apple does. And Intel is too big for everyone which also leaves it as a standalone company that serves the x86 server cpu market only. Other than that, it is hard to see a future for Intel not winthstanding their words.
    • American Capacity?

      Should have watched Kul Chown (?) on PBS last night explaining how American manufacturing is actually diversifying and expanding but not in the Ford Way of the 1920s. There are actually companies being formed to make hitech stuff which is for the near future use - as yet unthought of. Maybe it was a feel good show but it certainly reduced my pessimism about the US and I'm not even living there. Look into what you say before you write it. Just because all the latest news (naturally negative cause no one reads positive news) about Apple is about overseas "slavery", it aint necessarily true for anyone else. Darn, why is it always about Apple?
    • Almost 50% of Intel's Design Teams are already outside the US...

      Design and debig, validation, development is already around 50% US and 50% offshore.
      The percentage of domestin fab capacity oscilates with the process node.
  • Damned if you do, damned if you don't

    At the end of the day, neither path is particularly fun. Keeping a captive foundry up to date is a mind-boggling huge investment and actually fairly risky -- at every node there is a chance for screwing up the whole generation -- and can be a management distraction (i.e. you have to manage essentially an entire TSMC in addition to all your other business). Being fabless avoids that but then makes you beg for capacity, putting the "power" in the hands of the foundries.

    The real problem with fabless isn't the model itself, but rather the fact that it is for practical purposes single-sourced. There are only a couple viable options for any node, and even then you usually have to do major re-design to move between fabs.

    The statement that design and foundry are more closely coupled with a captive fab is mostly a red herring -- all the major foundry customers work extremely closely with TSMC or whoever.

    I would say that the main thing needed for the semi industry is more fabless options, or the ability to move between foundries more easily. Not sure how that will actually get solved though ... easier said than done.
  • Analysts have missed

    These analysts have missed everyone of Intel's 7-9 record breaking quarters over the last couple of years (I've lost count). All the while, Intel has been accurately forecasting their results. Nothing has changed. The analysts seem about as accurate at forecasting the technology market as weatherman forecasting the weather by looking out the window to see if it's raining before predicting rain.
  • Intel is ahead and eventually Atom will compete with ARM in the perf. per W

    IBM couldn't supply Apple so Apple switched to Intel. Intel is already at 22nm and expect to be at 14nm by 2014. They have already shipped a cell phone ready Atom at 1.6 Ghz

    Note Intel could always do ARM as well. They did before.
    • Intel Already Shipping Phones

      Intel has already designed and has started shipping phones with Atom Inside.
      Can't add URLs, so Google it...
  • "fab flexing"

    Where is there an explanation of this term, seems to be mentioned in this article without explanation, no reference to it on the WWW either apart from this article!
    • I don't know either

      I think It's a new corporate fitness thing.

      I didn't know Intel sold gym equipment though.

      I've been insourced, outsourced, upsized, downsized, homogenized, rehomogenized, and now I'm off to "Flex my Fab" (or in my case flab).

      I'll be really fit after all this.

      these "business speak" Euphemisms can be very annoying

      the only other explanation I can think of is that Intel can pick up entire production facilities and bend them to and fro like a giant "Wobbleboard" without affecting production. (apart from the odd employee rattling loose and falling out)
  • Intel sales pitch

    Intel's comments may be well founded but it sounds more like a sales pitch to me. Yes, they excel at manufacturing but they do not have to run the wide range of products that a Foundry has to so it's comparing apples and oranges. The capacity shortages are a short term issue that will go away. All companies want more capacity when the business is the but it's a fine balance for the Foundries to get the most out of their investment without having excess capacity.

    Until Intel starts to use their fans for significant Foundry use then they will operate in very different spaces. And then they will face very different challenges to their current manufacturing operations (few products in large volumes currently v many products in smaller volumes).

    I am still not convinced that Intel's mobile strategy is going to succeed as mentioned they haven't moved from their x86 model for the mobile arena. Power consumption is still too high for most users of smartphones and this is still an area that requires significant improvements. How many users still have to recharge after one day of use?
  • ARM is just that good. CPU chips are on shortage because it sells.

    If ARM architecture was not very good, would they have a shortage to begin with?

    Intel can make it's own processors but will they beable to deliver on intel powered Windows 8 ultra books? Will users flock to these new age laptops as they do to smartphones? Because if sales are low, then supply will not be an issue.

    Intel can't make smart phone CPUs because x86 is too old and requires more power and cooling. Even ultra book CPUs are too big and power hungery for tablet and much less smart phones.

    People by smart phones over new PC. Ask someone if they got 500 dollars free, would they get a new smartphone or laptop. The smart phone will win out majority of the time.