Apple's 64-bit iPhone fuels new speculation on ARM-based Mac

Apple's 64-bit iPhone fuels new speculation on ARM-based Mac

Summary: Is Apple moving its Mac desktop and laptop lines to ARM processors developed in-house? The company's powerful new 64-bit A7 processor found in the iPhone 5s is giving new life to the old rumors.

SHARE:
TOPICS: Apple, iOS, iPhone, Windows
34

More than a year ago, a crescendo of rumors talked up the idea that Apple would shift the Mac platform to its ARM-based A-series processors. This idea appeared to go into hibernation, but with last week's release of the iPhone 5s and its 64-bit A7 processor, the rumor has gained traction.

Many are impressed with the continuing results of Apple's microprocessor division following its buyout of P.A. Semiconductor in 2008. However, Apple has long produced a variety of ASICs and cores, including various bridge chips and during the PowerPC days in the 1990s, the floating-point instruction set called AltiVec, VMX, or the Velocity Engine. AltiVec was used in P.A. Semiconductor's PWRficient processors.

The A7 processor received a rave review from Anand Lal Shimpi at Andandtech. He compared the processor to the past generations of the A-series but also to Intel's Bay Trail FFRD and Qualcomm's MSM8974 Snapdragon 800 MDP/T.

Shimpi also addressed the topic of "larger machines."

Before I spent time with the A7 I assumed the only reason Apple would go 64-bit in mobile is to prepare for eventually deploying these chips into larger machines. A couple of years ago, when the Apple/Intel relationship was at its rockiest I would've definitely said that's what was going on. Today, I'm far less convinced.

Apple continues to build its own SoCs and invest in them because honestly, no one else seems up to the job. Only recently do we have GPUs competitive with what Apple has been shipping, and with the A7 Apple nearly equals Intel's performance with Bay Trail on the CPU side. As far as Macs go though, there's still a big gap between the A7 and where Intel is at with Haswell. The deficiency that Intel had in the ultra mobile space simply doesn't translate to its position with the big Core chips. I don't see Apple bridging that gap anytime soon. On top of that, the Apple/Intel relationship is very good at this point.

Although Apple could conceivably keep innovating to the point where an A-series chip ends up powering a Mac, I don't think that's in the cards today.

In his Monday Note blog, Jean-Louis Gassée offers some interesting analysis of the 64-bit transition for Apple and its competition. In the post titled "64 bits. It's nothing. You don't need it. And we'll have it in 6 months," he points out a number of problems for the Android infrastructure in the transition — not so much on the hardware side but for the operating system and app developers.

In addition, Gassée examines the chances for A7-based Macs.

Can we see a split in the Mac product line? The lower, more mobile end would use Apple’s processors, and the high-end, the no-holds-barred, always plugged to the wall desktop devices would still use x86 chips. With two code bases to maintain ß OS X applications to port? Probably not.

Apple could continue to cannibalize its (and others') PC business by producing "desktop-class" tablets. Such speculation throws us back to a well-known problem: How do you compose a complex document without a windowing system and a mouse or trackpad pointer?

We've seen the trouble with Microsoft’s hybrid PC/tablet, its dual Windows 8 UI which is considered to be "confusing and difficult to learn (especially when used with a keyboard and mouse instead of a touchscreen)."

There are many issues with input and performance with tablets. An iPad isn't a true substitute for a "real" computer such as a MacBook Pro or an iMac. Or the forthcoming Mac Pro workstation. Yes, an iPad is a fantastic piece of mobile technology and very capable of doing single tasks. I bring mine on short trips. But it can't really handle a complex content-creation workflow that requires the use of multiple applications and mass-quantities of data.

According to Gassée, the timing for ARM-based Macs is based on input, performance and workflow.

If Apple provides a real way to compose complex documents on a future iPad, a solution that normal humans will embrace, then it will capture desktop-class uses and users.

Until such time, Macs and iPads are likely to keep using different processors and different interaction models.

Now, I find the entire proposition of an ARM-based Mac is insanity, as I wrote a year ago.

First off, Apple has an existing, successful desktop strategy, part of which is based on a unique multiplatform value proposition: Macs are the only machines in the world that can natively run OS X, Windows and Linux. This capability is dependent on Intel logic.

The move to a proprietary ARM chip might return Apple back to its PowerPC days, where Macs required special "sole-source" justifications in enterprise and government for purchasing a machine that couldn't run Windows with adequate performance. With a modern Mac, users can run almost any commercially-available program in a number of ways: Mac native apps, Windows programs with a WINE wrapper, Windows in Boot Camp or in a virtualized environment, or native Linux. However, for all of this goodness to work, there needs to be an Intel processor.

What the performance and success of the 64-bit A7 brings is greater leverage with Intel when buying processors. I wrote last year that what Apple wants are price breaks from Intel.

So does Intel. The issue for Intel is likely all about margins — the company doesn't want to lower its gross margins. Intel could no doubt make its chips score better on power consumption, simply by lowering their performance. However, Intel wouldn't be able to charge as much for these chips (hint: performance is more the driver for price than power consumption).

Certainly, the best ARM processors next year and even through 2014 can't come close to providing the performance necessary for MacBook Air-class laptops. Still, perhaps in a few years, Apple might be able to have ARM-based processors good enough for a MacBook.

So, Apple's dance with its partner Intel continues. Apple can keep up the pressure in negotiations: make the lower-power and more importantly, lower-cost processors Apple wants, or Cupertino will phase out Intel. And now Apple doesn't just have to point Intel to stories about rumored future products; it can now point at the A7 and let Intel guess about the A-series roadmap. That won't take a clairvoyant.

Topics: Apple, iOS, iPhone, Windows

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Talkback

34 comments
Log in or register to join the discussion
  • Shhhhhh! It's a Samsung chip....

    The new A7 chip is a Samsung chip, not an Apple one - just used by Apple. Given the very public disputes in courtrooms round the globe, would Apple want to tie its desktop line to Samsung as well as its phones?
    Master668
    • No, it is an Apple chip

      made by Samsung.
      1,2,3
      • It's PWRficient Dual Core 64bit cores in a SoC!

        This is why it's so huge a footprint. Plus that's why they put an outside 3rd party co-processor in it too. The PA Semi 64bit core design was meant to compete with Intel from the start. It's basically a cross between an ARM RISC design and an IBM PowerPC RISC core. Which would be easy to map out for a genius like Jim Kelly with ARM Instruction Sets!

        Yeah..... although A4 was a Samsung Intrinsity design w/ the only difference being in Firmware Apple PA Semi design team was working on PWRficient core design for making it eventually being used to power their own . The following year for A5 Apple simply retained same design in iPhone 4s and iPhone 5 with A6 ARM A9 cortex cores designs off Samsung's ARM license shelf as a Merchant Foundry!

        It's quite obvious that the A7's CPU cores couldn't be designed from scratch in only 2.5 to 3yrs. This is why even ARM was shocked at A7 64bit announcement. ARM's own A50 series took more than 6yrs to develop. So these are not related to ARM cores. What Apple is doing is basically taking PWRficient core designs meant for desktops having Intrinsity apply FAST14 Logic designs for the A7 SoC and underclocking massive cores compared to ARM designs to get even more power efficient processors than the original PWRficient cores meant to kick Intel's rear end!
        KronJohn
        • PWRficient cores had hardware Virtualization!

          This too could easily be mapped out with ARM 64bit instruction sets on the dual core PWRficient processor shrunk down from 65nm to 28nm Samsung Gate First HKMG process. Which then underclocked could with FAST14 Logic be extremely power efficient. I don't believe M-7 is only used to calculate motion sense.

          This could be the core chip inside the SoC used for 64bit trusted zone computing with closed lockable vault core technology DSP? (like on Cell BE SPE)!

          This could easily translate into an introduction (finally) of a full ground up write of OS X to OS XI. Which could by clocking and using aggressive cooling techniques like liquid cooling, enable this low frequency A7 being able to clock all the way up to 4GHZ. Meaning it'd be a super computer in a SoC sized chip and with hardware virtualization now a part of ARM 64bit Instruction Sets, you could easily run Windows 8 or Linux on the ARM Hypervisor all at the same time!

          Apple was predicted to do what they've done since acquiring PA Semi.... being to use PA Semi's PWRficient core in desktops and notebooks as well as Apple TV, tablets and phones!
          http://www.tomshardware.com/news/Apple-PA-Semi,5325.html
          KronJohn
    • no...

      its an Apple designed chip. Its made in a Samsung factory...
      Just like.. an author of a book doesn't personally make all the copies... but the factory physically putting the book together didn't write the book.
      doh123
      • Not sure he would get that analogy.

        Books are clearly a foreign concept.
        matthew_maurice
      • Actually it's an ARM design chip

        ARMv8 was announced in 2011, Apple was the first to release a public available product, others will follow soon for sure.
        Processing power in mobile devices is growing like they do it in other kind of computing systems (it's a very old story), 64 bits is a natural step and the repetition of history - I really don't understand the fuss around it - it was obvious to me that this would happen.
        AleMartin
        • It's an ARM chip

          Apple licensed it and designed their own chip.
          Michael Alan Goff
    • Re: "The new A7 chip is a Samsung chip.."

      in the same sense that all the photos that I’ve taken and then printed on my HP printer are HP photographs.
      Slurry
    • A7 is designed by Apple.

      Its using a Samsung fab to create the silicon but that's because Samsung had the best price. (Apple was one of ARMs' founding partners.)
      MSBPodcast
  • Not at the moment

    An ARM based Mac is not likely at the moment as it doesn't have the necessary horsepower to run cpu hungry apps such as Final Cut Pro, Photoshop etc. And it can't emulate an X86 processor for Windows emulation satisfactory too

    However an 64 bit chips for an iPad like devices are great as it will push the limits of the apps
    ThinkFairer8
    • "it can't emulate an X86 processor for Windows emulation"

      Another couple of releases like Windows 8, and that won't be an issue.
      matthew_maurice
      • Another

        .. poor soul how never tried W8 and eats the media BS...
        brrunopt
        • He may not have tried W8, but I have

          and it is awful beyond words.
          1,2,3
          • How?

            Certain people here keep acting like it's horrible, yet everybody I've seen who uses it gets used to it within the span of a few days.
            Michael Alan Goff
        • Then Again

          Windows is now being ported to ARM.
          DannyO_0x98
  • What OS Would It Run?

    IOS isn't a proper computer OS, and OSX doesn't seem to port easily--look at the trouble they had going 64-bit.

    Maybe Android? Or some other Linux-based OS.
    ldo17
    • "Maybe Android? Or some other Linux-based OS."

      Nah, they'll want to use something that's actually popular on the desktop.
      matthew_maurice
    • Re: look at the trouble they had going 64-bit.

      OS X, by virtue of being an UNIX OS does not have any trouble to move to any architecture.

      In fact, OS X has been ported to 64bit long ago, during the PowerPC era. The PowerPC G5 was an 64bit CPU. OS X has also been fully ported to Intel's 64bit CPU years ago.

      iOS being a fork of OS X is trivial to port to 64bit, as has been evident from iOS 7 running on the iPhone 5S.

      Linux is no better in this regard. In fact, being a less unified OS than OS X (and BSD UNIX in general), Linux has more difficulties to be fully ported to other architectures. Fully, being the key word here. Let me be more precise here: 'fully' means that not only the kernel runs on the 64bit CPU and the compiler produces 64bit binaries, but that each and every single API has been ported (and well tested) to 64bit.

      Android as an runtime on the Linux kernel has other issues going 64bit too. It seems, nobody has been thinking about this for a while so it will take some more time -- but will ultimately go there.
      danbi
      • Re: OS X has been ported to 64bit long ago

        Just for comparison, Linux has been ported to an average of one new architecture per year during its two-decade existence. In fact, it went 64-bit at the same time it first went portable. Even Android now runs on 3 different architectures, compared to IOS or Windows PhRT.

        Now, what were you saying about OS X being "trivial to port"?
        ldo17