Intel – Killing the Monkey

RFM AvatarSmall

 

 

 

 

 

ARM is looking less and less threatening to Intel.

  • Thanks to the blunder that is Windows RT (Windows 8 on ARM), ARM has made no real impact in the PC space and the outlook for it to do so is fading fast.
  • On top of this, Intel is far from standing still and the more time that goes by the closer Intel gets and the less that the gap matters.
  • Last week, Intel revealed its latest chip core aimed at the mobile space called Silvermont.
  • This is the first chip for mobile that makes use of the trigate technology which adds further improvements on top of Intel’s process advantage.
  • The problem that Intel has faced over the last 6 years has been the inherent design of the x86 processor which, all other things being equal, will always consume more power than an ARM processor.
  • This has been the monkey on Intel’s back as it is unable to change the nature of the processor as it is the foundation upon which 30 years of success is built.
  • As more and more of its market becomes battery powered, this represents an obvious threat for Intel, which was really brought home by Windows 8 being made available for the ARM processor.
  • Fortunately for Intel, Windows RT is so limited that the devices have not sold in volume and manufacturers are pulling support for the platform.
  • Furthermore, it is going to be some considerable time before the Microsoft Windows OS is properly harmonised to run ARM as well as Intel.
  • In the meantime Intel continues to innovate and the new cores will shortly appear in real devices.
  • The first chipset is Bay-Trail which will be in tablets later this year and handsets in 2014.
  • Intel has already done a reasonable job at closing the gap with ARM and this development will take that one stage further.
  • I have already seen Intel powered smartphones that are sleek looking, competitive on specification and price and can go a full day on a single charge.
  • For me, this turns the tables.
  • For years, I have been concerned that Intel stood to lose share in its core markets of PCs and servers, but now I am starting to think the opposite.
  • As handheld devices need to be increasingly powerful, there is scope for Intel to even take market share from ARM as it is at last beginning to compete on a level playing field.
  • Intel is not expensive and seems to be showing results in mitigating the greatest threat that it has ever faced.
  • I am inclined to have a careful look at this one.

 

RICHARD WINDSOR

Richard is founder, owner of research company, Radio Free Mobile. He has 16 years of experience working in sell side equity research. During his 11 year tenure at Nomura Securities, he focused on the equity coverage of the Global Technology sector.

Blog Comments

If you define PC as personal computing, rather than computers that run Windows, the situation has hardly improved for Intel. When Windows RT fizzled, did the potential customer stick with x86 based Windows devices or did they switch to ARM based Android and iOS devices? Slow Win8 adoption and falling traditional computer sales suggest the latter.

Whether Intel’s mobile CPUs now have low enough power consumption compared to the past is irrelevant, even if somebody somewhere can make an Intel based phone that can actually run all day. As long as it is still behind ARM CPU/GPUs in terms of power consumption or processing power, not many handset makers will switch to Intel. Even it closes the gap completely with a combination of architecture changes and continuing process technology leadership, it is not enough. Why should Samsung or Apple or Huawei switch to x86? They can customize their chips, so that they can have every feature they want and nothing they don’t need. The volumes are high enough to fund their R&D efforts and their margins are likely higher than they would if they were to buy them from suppliers, so Intel has to come up with something much better or much cheaper for them to give Intel meaningful volumes. Then, there is the matter of combining CPUs & GPUs with processors for cellular, GPS, WiFi etc. circuitry. Good luck to anybody trying to sell something like that at a lower price than Mediatek. 🙂

Maybe I got the gist of the article wrong. The view I meant to put accross was that ARM needs to dosometing it wants to have a shot in the Computing world.

In Mobile, I am not putting out the view that INtel will take share from ARM but merely the possibility. It stsill early days and as you sauy there is still a lot going against INtel in the mobile world. Importantly INtel is not sitting still but is developing better and better products. it is no longer out of the question in my view.

[…] Windows RT has confused and put users off the proposition. (see here) […]

How about ARM in servers? Data Center group is almost 30% of Intel revenue and margins are likely fat there. ARM in servers does not happen overnight, but there is plenty of money at the table in this segment and growth opportunities for hungry challenger willing to accept lower margins.

Calxeda the arm based server company went out of business in November…says it all really….