There’s no doubt about it, Intel has some of the most impressive technology on the planet. Its processes and its fabs are beyond compare. When Mark Bohr spoke at the Intel Developer Forum (IDF) earlier this week, he was confident that the steps to five nanometer technology might be a little uneven, but he was sure that his company could do it. And details both disclosed and undisclosed about Haswell show that the firm has some cool stuff waiting in the wings.
But there is a problem and that is highlighted by Microsoft Surface, a $199 tablet based on a Tegra CPU from Nvidia and able to support Office applications. Other tablets from other manufacturers such as Android based machines and even the Apple iPad underline the difficulties that the chip giant will face in the future. The rumour at IDF this week is that Microsoft is only paying $10 for the CPU.
That is a price that Intel is totally unable to match and this poses many questions over its very future. The behemoth has always relied on high gross margins – far higher than other semiconductor companies can demand. These high margins are the reason why it is able to fund the fabs used to build the microprocessors – and fabs are getting increasingly expensive as process technology marches on.
And it is difficult to see how Intel will surmount this problem. Its complex network of partners are not going to build tablets that won’t sell. The curious thing is that Chipzilla had the ARM technology several years ago. At a previous IDF, we even asked the company why it did not just use StrongARM to build a very inexpensive, responsive and low power reference design. The reasons why they failed to adopt this possibility come down to the old business model – joined like Siamese twins to Microsoft, they essentially owned the market and were able to offer faster CPUs and upgrades to the Windows operating system in tandem.
That model has changed. Individuals don’t, generally speaking, care what’s under the hood of their computers anymore. They want to be able to read their email, tweet away, post their pictures to Facebook and to browse the web. As Microsoft produced its Windows and Office upgrades, PCs equipped with older CPUs slowed right down and prompted people to want to upgrade to a shinier machine. Sure, Intel and Microsoft produced extra features, but many of these were unnecessary or unused by the vast majority of people.
You can’t underestimate Intel, ever – but price is a very knotty problem and we’re not sure quite how it can apply the Gordian knot answer to its dilemma. The business model is changing and it’s unclear just how Intel can re-invent itself given the high capital costs it inevitably has.
As for Intel so too for AMD, of course.
If they fail to get into new form factors they might get pushed into server only and eventually out of that too as they have less and less share and funds for fabs and R&D.
They do have the fabs and they could try to become a foundry too but that's somewhat problematic,making chips for the competition would be unrealistic.
AMD already started to change,not very sure where they are going but they do seem to have a plan and in recent weeks the new team started to take shape,after some more arrivals.Maybe they are going more ARM like and design all kind of compute units that go in all kind of chips. Guess we'll see about thta in a few years
Anyone remember those not so long ago days when AMD brought us 32/64 bit capability, twin cores, integrated cache memory and many other innovations that made their chips the better buy?
Intels response....Manipulate the chip buyers rather than innovate. Everybody knows the story. AMD finally got a $3 billion settlement but by then their company was effectively destroyed as a major competitor.
Paying your customers not to buy your competitor's product is a tactic that will fail in the long term. Your staff, paid to innovate, becomes fat and complacent.
Maybe they can try Apple's current tactic. Litigate rather than innovate.
Only if tablets replaced most PCs and notebooks out there, then maybe they would get a problem, but I fail to see this as well. Tablets are a hype, which will peak at some point not too far away and then fall again just to level off somewhere. For some people and applications, a tablet makes sense as it is actually better than a netbook (e.g. if you have to work while standing and you have only one hand for input because you need the other to hold the device). But for most people, tablets are just useless toys, they are better of with a smartphone.
That said, I also fail to see Intel or AMD being concerned with the phone market, for the same reasons as above.
Where be your chimes now? your
gambols? your songs? your flashes of merriment,
that were wont to set the Tablet on a roar? Not one
now, to mock your own binning? quite chap-fallen?
Bons mots most sweet to the sheets.
http://www.motherjones.com/politics/2011/02/income-inequality-in-america-chart-graph
Heehee!