Why Intel should be scared of the iPad Pro’s A9X chip


The iPad Pro isn't ready to compete with Intel Macs... yet.
The iPad Pro isn't ready to compete with Intel Macs... yet.
Photo: Leander Kahney/Cult of Mac

Although we’ve been skeptical of the idea of ARM-based Macs in the past, rumors have been heating up that Apple will switch to ARM processors for all Macs this year due to massive performance gains. But as great as Apple’s A-series chips are, are they really ready to power Macs?

The answer isn’t entirely clear-cut, but a new series of benchmarks suggest that Apple’s ARM chips are starting to compete with Intel’s Core M chips, which power the 12-inch Retina MacBook. If this pace keeps up, iPads might be faster than Macs in just a few generations.

AnandTech’s in-depth review of the iPad Pro puts the A9X SoC that powers Apple’s 12-inch tablet through its paces with a series of benchmarks. The testing compares the iPad Pro to CoreM devices like the Retina MacBook, ASUS Transformer Book T300 Chi and ASUS ZenBook UX305CA.

The results? The A9X is competitive in many benchmarks with the MacBook and ZenBook — it won half the benchmarks, but lags in overall performance — but not the Transformer Book, which beat the iPad Pro in every category.

Even so, AnandTech’s conclusion is that Apple’s ARM chips could go head-to-head with Intel in just a few generations.

A9X can compete with both Broadwell and Skylake Core M processors, and that’s something Apple couldn’t claim even a generation ago. That it’s only against the likes of Core m3 means that Apple still has a way to go, particularly as A9X still loses by more than it wins, but it’s significant progress in a short period of time and I’ll wager that it’s closer than Intel would like to be, especially if Apple puts A9X into a cheaper iPad Air in the future.

So it doesn’t look like we can expect ARM-based Macs in 2016, despite what ex-Apple execs may predict.

But by 2018? That’s another story. Intel should be very scared of that, given the fact that Macs are the only class of desktop or laptop computers still gaining market share.

Source: AnandTech

  • TJ

    What people seem to overlook is that Apple could and might develop a separate range of A-series chips for its Mac lineup. I can see the MacBook maybe running on the same A chip that the iPad runs on, but when you come to the MacBook Pro you need something more powerful. Probably something with more cores, higher clock speeds, etc. especially when it comes to the higher spec models. And with the Mac mini and iMac you don’t need to worry about battery life so you can run a more powerful chip aimed at desktop processing.

    I think it’s only a matter of time before Apple does make the transition to the A-series chip in Macs. But I find it unlikely that they’d use the exact same mobile processor from something like the iPhone in all of their Macs due to the different demands of each machine.

    • CaedenV

      Nobody is saying that this exact same mobile chip is going to find it’s way into the MBP line, or the desktop line… but the tech behind it could potentially be scaled up to compete with Intel; First in smaller devices, and slowly but surely up the food chain.
      Keeping in mind that the largest issue with ARM is that it is horribly inefficient, especially for higher performance devices. But as tech keeps progressing those efficiency differences are slowly shrinking, and it may well be a matter of time.

      Still, I work with a ton of Mac Books (I run tech for a school district, and some non-profits), and every one of them needs to run Windows at one point or another. The idea of moving away form x86 and loosing Windows on Mac would be a pretty big blow to laptop sales. But then again ‘Windows’ is also gaining more and more compatibility with ARM… so maybe in a few years you will be able to get an ARM MBP and still run full Windows on it… or maybe just Windows Mobile with Continuum… but that might just be enough.

      • HowmaNoid

        That’d have to have some kind of Rosetta-like emulation layer again. Transitioning the app base to ARM would take years.

      • CaedenV

        would it? Remember that Apple is the great follower. They have a HUGE arm based app platform that could very easily transition to windowed desktop apps, and when they do it the whole process would be ‘innovative’ and ‘new’ (never mind that this is what MS has been trying to do for the last few years).
        You and I may not be satisfied with it, but I know a whole lot of iPhone and iPad users who are constantly annoyed when they get their first Apple PC and it is nothing like the mobile offerings. They would be very happy to see a large glorified iDevice running mobile apps. And they would already have most of the porductivity tools they would need (office, rdp, etc.)… again, you and I would not easily stand for it, but a lot of people would welcome the change.

      • 5723alex .

        The next SurfacePhone (Lumia is dead) will run Windows 10 on ARM.

  • Geoffrey Spencer

    As long as VMWare and Parallel continue to work with these “Mac” chips and definitely not as emulators, I am fine with it but again it must match the power of the MacBook Pro level. FYI, I work in software for a consulting firm.

  • David Malcolm Puranen

    I think it’s important to remember that Apple doesn’t even use the same processor in the 6S as it does in the iPad Pro. So you won’t see an A9X in a MacBook, you’ll probably see something beefed up. Also they’ll probably handle it similar to how they did the PPC to Intel transition. Move the low end over, and then keep the high end available for those who need legacy support.

    The tough part is gauging when to kill x86 processors. There were some upsides like the ability to virtualize Windows, that’s going to go away with the switch to ARM. (So once they start the transition, if you need that buy whatever the last MacBook Pro is, because after that you’re going to be using a remote desktop protocol to log into a virtual windows instance on a server or something!)

    My guess is that they’ll do the MacBook first, then a year later do the MacBook Pro. Though I think they’re waiting for their own in house GPU instead of using the PowerVR GPUs they currently use. Once they’ve got something that they can happily run the 12 inch MacBook with, I think you’re going to see a compressing of the product line. The MacBook will probably start to out perform the MacBook Air, at which point the Air is phased out, and once they’re able to get performance in the MacBook Pro that rivals whatever Intel will put out at the same time, they’ll move over to the Pro, probably with a thinner design, or leave the thinness and say it gets 24 hour battery life. (It’s already pretty frickin thin, though I say that about my old Sandy Bridge 13 inch Pro.)

  • Rastervision

    You can buy 4 Core M laptops, and upgrade each to 4gb, for the price of the iPad Pro. Not to mention the Core M processor is really only good for light daily tasks.

  • bdkennedy

    Look back to the transition from PPC to Intel. That was going on for years in secrecy. We don’t know what Apple’s A chips are capable of because they are likely under-clocked so they don’t need fans to cool them. Think of the impact of under-clocking their processors and then springing over-clocked CPU’s onto the world in an iMac. THIS will be Tim Cook’s shining star. Just look back to the first Intel iMac 11 years ago.

  • calden74

    Why should Intel be scared, the A9x is only used in a single product line and Apple has no interest in selling their SOC to other manufacturers. So why again should Intel be scared, this is nothing but a fluff article. Even if Apple started using the A9x or next iteration, the A10X, in other product lines, say the MacBook 12, it will be just for that line and not say the MacBook Pro 15. Intel is doing just fine with their Atom SOC and new M CPUs, the new z8300, z8500 and z8700 are no slouches and will continue get faster. I don’t know if any of you have noticed but the Atom has almost tripled in speed in the last year alone. My little Xiaomi MiPad 2, which is just a fantastic little tablet, I bought both the Android 6.0 and Wimdows 10 versions, anyway, the Atom Z8500 they use is a very peppy SOC, at least the apps that I’m using on it don’t lag in the slightest. I can stream 4K without a single frame drop and play Halo for PC or Doom 3 at 60FPS ,yes, their very old games but still look good and are tons of fun.

    Every time Apple comes out with a new SOC articles like this always popup. Yes, the A9X is a fast SOC, no, it won’t affect Intel in the slightest, at least not until Apple starts selling their SOC’s to other manufacturers but I highly doubt it. Also, we won’t be seeing ARM used in any of Apples computers until they can make that move for the entire line. As there are no OSX applications compiled using ARM yet, just switching, say the MacBook 12, would spell doom for that line, I mean who in their right mind would buy such a product. No, like the PowerPC to Intel move before, Apple will do it all together, however that isn’t going to happen for a very long time, if at all.

  • Matt

    What is the benefit to the user if Apple were to Switch from Intel to ARM? Would iPhone and iPad apps and games become compatible with Macs?

  • CelestialTerrestrial

    Why would Apple go through the expense and disruption to change from CISC to RISC for OS X? It makes no sense. Plus, it wouldn’t be able to run Windows unless you had the RISC version and I don’t think people that run Windows are going to want to do that. I think it’s just ridiculous to even conceive of that. It’s not going to happen. Would people just stop thinking that Apple isn’t going to do that. All they are going to do is to make their iDevices more powerful and eventually have more features to help make it better, and iOS apps will just become closer to a OS X/WIndows version only for a mobile device.

  • P0l0nium

    Another way of looking at it is this :-)

    In order to make an SOC that’s nearly as good as last year’s lowest power Broadwell SOC you need a revenue stream the size of Apple’s to support the CPU team, you also have to subsidise the foundries’ process development and you end up with a part that’s 50% bigger than the Intel part *.

    And then you can only sell it profitably in a $1000 slate.

    * And you can’t reuse that CPU core in a server (for example) because it won’t clock over 2.5GHz without melting its path to the center of the Earth !!

  • Gal5

    Sure, because we know history, and technological history, just keeps progressing linearly for everyone and everywhere.