Following up on yesterday's report from Bloomberg that has rekindled discussion of Apple potentially making a switch from Intel's processors to custom ARM-based chips for its Mac lines, AllThingsD takes a closer look at the landscape to analyze the benefits and challenges of such a move.
/article-new/2012/11/intel_arm_logos.jpg?lossy "intel_arm_logos")
The report contrasts Intel's focus on performance with ARM's focus on power efficiency and notes that there is no inherent reason why Apple couldn't push ARM technology more toward the performance end of the spectrum to make the chips more suitable for desktops and notebooks.
If a company decided it wanted to design an ARM chip that was, as [analyst Nathan] Brookwood put it, “hell-bent on performance,” it could be done. “You could get a pretty fast machine,” he says.
Trouble is, it would have to be not only be fast, but have a really excellent roadmap lasting well into the future that not only met but exceeded that of Intel. That’s a tall, tall order.
The report points to ARM's just-announced 64-bit processor designs as being key to any move into the Mac, and notes that Russian company Elbrus Technologies has developed a Rosetta-like emulation technology that could allow ARM chips to run software written for Intel processors.
On the flip side, Intel has a long track record of being at the forefront of processor performance and a strong history with Apple since the company's Mac lines transitioned from PowerPC chips seven years ago. As a result, any decision to switch from Intel to ARM obviously can not be made lightly, and Apple is clearly taking an extended timeframe to examine the possibility. But with Bob Mansfield now heading up a division of Apple dedicated new technologies in wireless and semiconductors, it seems that there may be a renewed focus on pushing the boundaries of the ARM platform even further.
Top Rated Comments
Biggest hurdle will be, that nobody will buy those products.
p.s. If they already have a roadmap for the switch, I hope they used "Google maps" at least.
Cause if there is a hurdle in the future they won't have the years to design a solution. They'll have to already spent the years making the solution. Ie better to have a solution you never have to use then get caught with your pants down and no immediate solution.