Intel has been painfully slow to elbow its way into the mobile market and this year will only be releasing 32nm architecture Medfield Atom Z2460 processors, which should help to keep it in the running. Certainly, with the thumbs up from the likes of Orange, Medfield will help Intel claw its way into the market. But Otellini seemed to be hinting that this was just stage one of a cunning plan for domination of the mobile market.
According to Otellini, things will get a bit more interesting next year when it shrinks its chips down to a new 22nm process and then further to 14nm by 2014.
The shrinking of chips should mean that they take up a lot less space inside a smartphone and will use less energy. This will be more attractive to manufactures who will actually want Intel inside their gear.
Getting to 14nm is pushing Chipzilla close to the theoretical limit of silicon, which is around 9nm. However, the company has mentioned before that it expects to eventually jump to 8nm by 2015.
Intel can manage all this shrinkage by using FinFETs, which turn transistors into multi-gate devices. Since a transistor can do more it means that the number of electrodes and the size of chips can be cut.
Otellini hinted that Intel had plenty of manufacturing capacity to turn out the chips cheaply, too.
Such a strategy would appear to win in the long term. It does assume that ARM cannot come up with similar technology and that its partners will not be able to match Intel’s production.
While this is possible, it might prove more difficult for Intel’s rivals to pull off. Intel is clearly preparing its steamroller and if the rivals don’t manage to get theirs out of the shop in time it will roll right over the top of them.
For a while, many have assumed that Intel would not get its act together on mobile at all. Apple even went ahead and invested in its own ARM based chip development for its mobiles. But if Intel does manage to get these chips shrunk as fast as Otellini wants, then Apple is going to regret that particular move. Its ARM investment will not have paid off before it comes redundant. Apple is also desperately looking for space within its gadgets to shove in more toys, so it will need smaller chips quickly. The same will apply to other hardware makers.
Playing the longer game gives Intel some advantages, which, if it is canny, should lead it to dominate mobile in the same way that it does the PC market. If its rivals have any sense they will be thinking very hard about where they want to be in five years time. Otherwise, they will be looking at being the Linux desktop of the mobile world.