Benchmarks from Apple’s next-gen A10X processor show unmatched power

“Apple’s new iPhone 7 and iPhone 7 Plus are far and away the most powerful smartphones the world has seen thus far,” Zach Epstein reports for BGR. “In terms of both raw benchmark test scores on paper and real-world performance, Apple’s latest mobile devices outshine every other phone on the market. In fact, they’re even more powerful than Apple’s latest iPad Pro tablets, which is no small feat.”

“As powerful as they already are though, Apple is constantly working on next-generation products, looking to make them better and faster than their predecessors,” Epstein reports. “Now, it looks like benchmark tests from Apple’s next-gen [A10X] processor have leaked [ahead of it’s expected inclusion in iPads early next year], revealing huge performance gains that are somehow even more impressive than what we just saw from the iPhone 7 and iPhone 7 Plus.”

“As a refresher, the iPhone 7’s new A10 Fusion chipset manages impressive scores of 3490 in single-core Geekbench tests and 5580 in multi-core tests,” Epstein reports. “According to this leak, Apple’s upcoming A10X scored 4236 on Geekbench’s single-core test and a ridiculous 6588 on the multi-core test. ”

Read more in the full article here.

MacDailyNews Take: More than enough performance to begin powering Macs.

18 Comments

  1. I recall a review where the iPad air was use to edit video against a MacBook pro of the same time and the iPad was faster ( just a little bit but faster).
    4 generations ago the “A” series processor was faster than the intel one.. imagine now.

      1. Yes difficult to believe isn’t it Mike, as you said they will never be used in Macs.

        For me however the logical question is from a work perspective, not whether they CAN replace Intel (only a fool would think they can’t, very soon if not now) but do I actually want it, Bit conflicted on that one at the moment because on a non business perspective its probably a why not.

        1. I never said that A chips will never be used in Macs. I said that it would be stupid at this point for Apple to switch to ARM chips in Macs because:

          1) ARM chips are not fundamentally faster or more efficient than CISC chips at a system level in ways that matter to users. CISC has made just as much progress as RISC chips in the past decade – especially in other ways besides just core speed. Huge memory thoughput, PCI expansion, and Thunderbolt I/O immediately spring to mind. None of that has even been attempted on Apple’s A chips.

          2) ARM doesn’t support the most popular software that power users and businesses rely on. Macs must have that compatibility now and in the foreseeable future. This is why 10% market share mattters. Apple’s refusal to chase a bit more market share with desktop Macs means that it doesn’t have the developer base to follow them to any new chip architecture. That’s a significant part of why Apple had no choice but to abandon its G chips and adopt Intel in the first place. If the G5 chip was fundamentally better, then Apple could have produced its own chips or farmed out production to any other foundry when IBM lost interest in its money-losing venture. But Apple didn’t, they knew they had to go to Intel architecture to keep the Mac viable.

          3) Apple itself cannot support the software bloat that comes with forking its OSes or implementing dual-chip hardware. As it is now, Mac is CISC and iOS is RISC. That is the way it has to be, and it’s also the reason iOS will never be competitive in many computing fields. Merging the two is a fool’s errand, as Cook himself has proclaimed. See the MS Surface RT and tell us how offering mobile Apps in a laptop format makes any sense whatsoever. The full-featured Surface 4 is a hit, because it has the software compatibility and just enough hardware power & flexibility to be desireable to road warriors on and offline. iPad does not. An ARM Mac would not.

          4) Economy of scale — if Apple starts up a low-volume production run of special chips just for Macs, then how would that be cost competitive? Macs are already losing cost competitiveness because of low/stagnant production rates.

          1. Mike, I’ll again refer you to the intel transition when everyone was making these same arguments in reverse. It’s all about performance per watt, if the A series chips have a 2-3x delta over their intel equivalents, then a Rosetta solution handles the software issue, and if they can get them that fast? Llvm support is easy to implement. They wouldn’t “fork” the os, the code os already there. It would be a universal binary just like they did twice before in both the 68k – ppc transition, and the ppc-intel transition. The differences between cisc and risc aside, risc chips at a given clock can handle more information than a cisc chip. Lower clocked risc chips can easily out perform higher clocked cisc chips. That’s the one thing that hasn’t changed in the last 11 years. Arm can easily be brought up to speed to compete with or pass intel… with a single core of 4200+ and a multi core near 7000 on a dual core design for the A10X, that’s haswell and broadwell level performance, and even matches some dual core skylake chips like the U series. All while consuming significantly less energy. I’ll say it again, PA semi had a ppc variant risc chip that was 2x faster than the core 2 duo in 2008, there’s no reason they couldn’t have been advancing that design all this time.

            1. And to answer your economy of
              Scale question…. there are over 1billion devices using A series chips… if they’re all stamped at the same fabs that would easily control pricing for a desktop/Mac version of the chip.

            2. “unmatched power”, the headline reads. whatever. Everyone who has replaced their old MacBook Pro or iMacs with an iPad, please raise your hand.

              You would think that with as much vitriol as MDN slings at analysts, they would cast an equally critical eye on the meritless claims by another clickbaiting analyst who claims ARM processors have caught up with Intel CISC chips — all because Intel now offers low-performance netbook chips as well as high performance workhorse chipsets. I showed you on Geekbench that the very best A10X chip from Apple is only 1/5 of the performance of a desktop Intel chip. I don’t know what more data you need to understand that the performance gap remains wide and isn’t closing anytime soon.

              If you want to keep cheering for a new Rosetta era, go for it. There are many reasons that Apple has publicly stated they would not merge the iOS devices with the Macs. It seems like you’ve swallowed the iPad Pro marketing BS (which is identical to the Microsoft Surface sales pitch, by the way) hook line & sinker.

              So don’t take my word for it, go ahead and believe the internet rumors and vapid speculation. Apple CEO Cook can’t possibly have meant what he said…
              “We feel strongly that customers are not really looking for a converged Mac and iPad,” said Cook. “Because what that would wind up doing, or what we’re worried would happen, is that neither experience would be as good as the customer wants. So we want to make the best tablet in the world and the best Mac in the world. And putting those two together would not achieve either. You’d begin to compromise in different ways.”

              If you believe that Apple can wring more performance out of silicon than Intel can at a profit, then by all means, keep deluding yourself. Those of us who rely on Macs for scientific computing, professional audio, video, and graphics, or even just gaming would all strongly prefer that the Mac remain natively compatible with Intel chips. Performance — not energy efficiency — is what matters to the user.

              Challenge: if A-series chips have arrived, then PLEASE go ahead and write a Rosetta emulator that allows the fastest iPad to run a native Mac application. Go ahead. Go ahead and show us how that process is more efficient than just running macOS on an affordable, mainstream Intel CISC. Prove your theory instead of pushing it in everyone’s face as if it was destiny. There is no economical or performance reason that anyone does it today, so you can go ahead and be the first.

  2. Right now it’s way more important for apple to ship faster iPhones than to ship newer more power efficient laptops. I’m beginning to wonder if the hoped for a series Mac will end up being a maxed out iPad Pro instead. Within a year or two the chip speeds should be equivalent. And Mike should check out YouTube. There are render speed tests of the Mac Pro vs the MacBook. It’s close enough to be interesting.

  3. IS raw processing power enough to get market share?
    I had radio on this a.m. and there was a “tech correspondent” on that was touting the new google line-up. He seemed to be anti-Apple so I had cause for pause in absorbing his words.
    He said that google now has the best battery. One can charge in 15 minutes for 7hrs of use. He said they have the best camera per some camera rating service. And they have the best AI ecosystem because they’ve spent so much time developing self-driving cars etc. Also best ecosystem for APPs, search, navigation, and on and on. He suggested that the new phones would take a huge chunk of market share from Apple.

    What think ye mates?

    1. I had several collegues waiting and really wanting the ‘next Nexus’ or its replacement. They *ALL* ranted about the price this morning…

      They might get some of the Samsung clients that got ‘burnt’ by their latest purchase but for what I understood in short time the Android developer crowd was pretty vile and dissatisfied. They did liked the 3.5 mm jack though.

  4. What Google doesn’t mention is that their phones are less pretty physical copies of iPhones (ugly as sin half glass half metal back), are significantly thicker in order to do away with the camera bump (their camera cut out looks cheapo anyway; I like the iPhone 7 bump more than their flush implementation), doesn’t include 3D touch or linear haptic feedback, not water resistant, no stereo speakers, no OIS (how do you get the best DXmark score with no OIS?), no second camera for optical zoom or rangefinding, and it’s always snooping.

    1. Of the criticisms you have against the Google knockoffs, none of them matter to the price-conscious buyer.

      And the only one that I care about is the Google snooping.

      That being said, my MacBook Pro is constantly phoning home to Apple servers, so I’m not entirely convinced that Apple isn’t datamining everyone in order to teach Siri how to be useful. I seriously doubt that anyone can truly trust any company now with your private data. Whatever you backup to iCloud is theirs.

  5. What if the user or business needs to buy their licenses again and they have thousands invested in software? Think of many software business today where you need to buy a separate license per seat, per OS.

    It is a huge logistic and legal problem that will take years.

    But anyway Apple may be able to successfully create a “Mac/Arm” computer using proprietary chips inside, no problem there. For most people with common or simple needs it may be enough. And just as the app store grew up, a dedicated “Mac/arm store” could gain general desktop usage. It will be an all grew up desktop iOS-Mac ish device/computer. Why not?

Reader Feedback

This site uses Akismet to reduce spam. Learn how your comment data is processed.