Will the 2019 Mac Pro be powered by an Apple ARM-based chip?

“Apple’s new Mac Pro won’t arrive until 2019, the company now confirms,” Carl Velasco writes for Tech Times.

“It’s possible the new Mac Pro could be the first Apple computer to come with an ARM-based processor,” Velasco writes, “Such a decision will confirm the company’s transition away from Intel chips for its Macs, as reported recently.”

“It’s pretty unlikely, though. Apple is trying to repair the Mac Pro’s reputation to its target audience, most of which have been disappointed at Apple’s lackluster support for serious Pro users,” Velasco writes. “Putting an ARM-based processor on what’s supposed to be a huge comeback is a risk, and Apple is too calculating and careful to take that, especially when the stakes are high.”

Read more in the full article here.

MacDailyNews Take: Putting an ARM-based processor in a Mac Pro isn’t as big of a risk as not bothering to release a new Mac Pro since 2013.

How many Mac Pro customers will be left by 2019 (December 31, 2019, most likely, this being Tim Cook’s Apple) and what gives the remaining Mac Pro customers any hope that Apple won’t abandon whatever they release for another five years or longer?

SEE ALSO:
Apple: No new Mac Pro until 2019 – April 5, 2018
Apple reiterates they’re working on an all-new modular, upgradeable Mac Pro and a high-end pro display – December 14, 2017
Why Apple’s promise of a new ‘modular’ Mac Pro matters so much – April 6, 2017
Apple’s cheese grater Mac Pro was flexible, expandable, and powerful – imagine that – April 6, 2017
More about Apple’s Mac Pro – April 6, 2017
Apple’s desperate Mac Pro damage control message hints at a confused, divided company – April 6, 2017
Who has taken over at Apple? – April 5, 2017
Apple’s embarrassing Mac Pro mea culpa – April 4, 2017
Who’s going to buy a Mac Pro now? – April 4, 2017
Mac Pro: Why did it take Apple so long to wake up? – April 4, 2017
Apple sorry for what happened with the Mac Pro over the last 3+ years – namely, nothing – April 4, 2017
Apple to unveil ‘iMac Pro’ later this year; rethought, modular Mac Pro and Apple pro displays in the pipeline – April 4, 2017
Apple’s apparent antipathy towards the Mac prompts calls for macOS licensing – March 27, 2017
Why Apple’s new Mac Pro might never arrive – March 10, 2017
Dare we hold out hope for the Mac Pro? – March 1, 2017
Apple CEO Cook pledges support to pro users, says ‘we don’t like politics’ at Apple’s annual shareholders meeting – February 28, 2017
Yes, I just bought a ‘new’ Mac Pro (released on December 19, 2013 and never updated) – January 4, 2017
Attention, Tim Cook! Apple isn’t firing on all cylinders and you need to fix it – January 4, 2017
No, Apple, do not simplify, get better – December 23, 2016
Rare video shows Steve Jobs warning Apple to focus less on profits and more on great products – December 23, 2016
Marco Arment: Apple’s Mac Pro is ‘very likely dead’ – December 20, 2016
How Tim Cook’s Apple alienated Mac loyalists – December 20, 2016
Apple’s not very good, really quite poor 2016 – December 19, 2016
Apple’s software has been anything but ‘magical’ lately – December 19, 2016
Lazy Apple. It’s not hard to imagine Steve Jobs asking, ‘What have you been doing for the last four years?’ – December 9, 2016
Rush Limbaugh: Is Apple losing their edge? – December 9, 2016
AirPods: MIA for the holidays; delayed product damages Apple’s credibility, stokes customer frustration – December 9, 2016
Apple may have finally gotten too big for its unusual corporate structure – November 28, 2016
Apple has no idea what they’re doing in the TV space, and it’s embarrassing – November 3, 2016
Apple’s disgracefully outdated, utterly mismanaged Mac lineup is killing sales – October 13, 2016
Apple takes its eye off the ball: Why users are complaining about Apple’s software – February 9, 2016
Open letter to Tim Cook: Apple needs to do better – January 5, 2015

37 Comments

    1. Putting an ARM instruction set based chip in the next Mac Pro (assuming the next Mac Pro ever shows up) would the the second stupidest thing Apple could do. (Just killing the Mac Pro all together would be *THE* stupidest thing.) — Remember Apple’s A-series chips are based upon the ARM instruction set that they license from ARM. Apple does not license the ARM chip designs from ARM like several other ARM chip manufacturers do.

      IF (an absolutely HUGE IF) software developers that generate software for true professionals were to write all new software based around the ARM instruction set it would take an absolute minimum of a year (and in many cases two or more years) after they had actual hardware upon which to test their early software builds before they shipped actual, robust applications to end users. Thus, IF Apple shipped an ARM based Mac Pro in mid to late 2019, Professionals likely would not have usable professional software until mid 2020 or later. What’s a Mac based professional to do in the meantime? Keep using that 2012 Mac Pro? (And realize that by 2020 Apple’s latest macOS at that time will very likely have dropped support for all Mac Pros prior to the 2013 trash can model and maybe even that model.)

      1. First of all, if you think the debut of an ARM based Mac Pro would be a surprise to Apple’s main 3rd party developers, you don’t follow Apple much. Apple would notify 3rd party software houses long before the switch is ever made final AND they would offer those developers the tools to start the migration.

        Second, developers do not base software around a specific ISA unless they’re writing that software in assembly language, which just does not happen much anymore.

        And finally, moving from one architecture to another would NOT take a year to do. Remember developers used to be able to target both PowerPC and Intel at the same time with no change to their code. In fact this happens now with iOS apps, they are compiled for Intel when run in the iOS emulator and compiled for ARM when run “on device”.

        Yes, you will not immediately be able to copy an application from an Intel based Mac to and ARM based Mac. The developer will have to recompile their application to target both ISA’s, before that will be possible.

        1. Just because the software developers “aren’t surprised” by something like ARM – – doesn’t mean that they’re going to drop everything else that they’re doing and RUSH to go update their Mac software.

          Case in point: 2008 update of Adobe Photoshop: CS4 was 64 bit for Windows, but not for Mac. Remember? And it isn’t like Adobe is a small developer.

          Rewrites take money. Follow the money. No money = no rewrite.

          The historical reality is that for all of the pieces of an advanced workflow to be rewritten … and debugged … expect for it to take (at least) the better part of two (2) years before it is solid and resulting in positive value-added…

          …and for some workflows, the reality is “NEVER”, because some important cornerstone piece is no longer being actively updated by its developer, so the time frame until ARM conversion is “Never”. That’s why we sometimes will still see Macs running OS 9 (yes, pre-OS X) on a workfloor – – personally, I’m aware of one vendor of mine who, as of my last visit two years ago, still had it plugging away on an old G5 PowerMac. Can it be replaced? Not with OS X or Windows, and since it is the controller/interface hooked up to a piece of specialty equipment, every day that it keeps on running is another week that they don’t need to find $2M in their budget to replace that specialty equipment. Needless to say, they’ve bought up some old G5’s off of eBay/etc.

          1. You are talking about some VERY specific cases, that regardless of what happens, that software will remain as it is – stuck on the system it was designed to run on. Should Apple have stuck with PowerPC just so that one person could continue running his software on modern hardware? No. That’s ridiculous to expect and doesn’t have any point in this conversation.

            Also, Adobe had a long history (and probably still do) of not using Apple’s developers tools, so they’ve always had problems migrating from one system to another – even from one version of the OS to another.

            Steve Jobs hated the fact that 3rd party software could keep Mac users from updating their systems. That was the main reason for originally pushing developers to move to Xcode.

            So, as I stated above… almost all Mac software written today will have absolutely no problem moving from Intel to ARM.

            1. “You are talking about some VERY specific cases,..”

              No, I’m merely relating one specific example anecdote: the point remains (as you point out) that software is a key factor that holds back updating of systems.

              Because no one buys a PC/Mac just to watch an OS run a screen saver.

              The value-added is the productivity enhancement from software applications working on data. And when Adobe (or any other”800lb Software Gorilla”) wants to do things their way, they do things their way.

              Finally, I do not dispute that “almost all” software will have no problem moving from Intel to ARM, if your definition of ‘problem’ is centered around the technology. However, that doesn’t pay for the conversion work, so the software which actually does get rewritten is a subset of that which *can* get rewritten, and the software that didn’t change anything at all (including all data formats) is a subset of that, and the software which then gets bought by customers is yet another subset of that, and so on.

              The net result is a lot of non-converting customers running on older legacy systems, resulting in customer market fragmentation risks, many of which will avert to Windows on their own schedule, rather than the ones being imposed on them by Apple & a plethora of 3rd Party software updating vendors.

              Overall, this is why a major conversion like this has to be carefully considered, with the primary objective being the question of what makes this change COMPELLING for the _customer_ to want it?

              And heed well that I said customer: changes like this need to be a “customer pull”, not a “Supplier push” (same as a “technology push”).

              To this end, just what is _compelling_ about an ARM conversion to a Mac Pro customer? Is it going to slash the cost of (hardware+software) by 33% per seat? No, it literally can’t. Is it going to result in 25% more productivity? No, not that either.

              Until there’s a good answer that, the customer has no reason to buy.

  1. I will never understand, why they did not simply take the old cheese grater Mac Pro case with new motherboard, processors, graphic cards. I am still running half a dozen of these fine machines from 2010 in my company, with updated graphic card, more RAM, and SSD. Still one of the best Macs ever, if not the best.

    1. Me too. I refused the trash can version and knew the design was a dead end. I am a pro user using a 2012 Cheese Grater Mac Pro upgraded to the max. The perfect design for a pro computer at its inception, it is still a viable design for now. It is slower than I need, but with PCI disks and SSD’s, the cheese grater performs admirably… especially for music.

  2. Apple needs to kill this rumour NOW, if they are in fact remaining with Intel for the next Mac Pro. After 5 years in the dark, that’s the *absolute least* they can do for the pros they claim to care about, yet still deny them new hardware for close to *2 years* after announcing they’d get a “modular” Mac Pro (announced early April 2017, earliest release Jan 2019 would be 20 months).

    Otherwise, I wouldn’t blame pros one bit for moving away from Apple. They deserve a lot better than the years of contempt and now bare scraps of info that are being tossed their way.

  3. That’s a point I also brought up. If we are in a transition period why would anyone spend upwards of $10,000 on a Mac Pro only to find themselves shortly thereafter left out in the cold by an architecture change and all that implies?

    It keeps coming back to a Windows/Linux solution that we know will be stable for some time to come. Pros love stable! Apple is really putting their pro users in a choice/expense quandary and it’s only going to be getting stickier.

  4. Dead on arrival if they do this. OBVIOUSLY they would start with the consumer products in the transition and leave the Pro apps and hardware for last. Pros are already scared of Apple commitment to them, the last thing they would want is a move to unproven chips and the wait and issues with the pro apps recompiling to a non-intel architecture. Not happening.

    1. I agree, Tom, that a Mac Pro based on ARM processors would be a big risk for Apple.

      The scenario that you outlined of introducing A-series SOCs on a lower-end consumer Mac as a pathfinder – perhaps the MacBook or the Mac mini – is consistent with Apple’s general approach on new technology introductions. I have been promoting that scenario for years, and I believe the the A11/A11X is the ideal starting point for that process. Based on the success of that effort combined with advancements in A-series performance in subsequent generations, Apple would then expand its ARM-based offerings upwards through the Mac lineup. The processing and graphics performance of Apple’s A-series processors has improved rapidly since 2007, and the A11 is the performance ballpark of mainstream Intel processors used in consumer-grade computers.

      1. ARM chips are not equal to Intel’s x86 chips in total performance.

        When you talk about pro computing, chip energy efficiency takes a backseat to many other factors. A chips aren’t the answer even if the changeover OS costs were ignored.

    2. One of the questions that I have had some time is, “Is the A-series SOC designed to enable interconnection of an array of SOCs for massively parallel computing?”

      The A11 is a hex-core SOC – two high-performance cores and four high-efficiency cores. But can you place multiple A11 SOCs on a motherboard and have them work together to form a more powerful parallel computing platform? I ask this because my recollection is that the old PPC G4 processor was designed for that type of application, while the PPC G3 was not. So, the question arises – can you gang together an array of A11 SOCs, or will Apple need to evolve its A-series SOC architecture to enable a massively parallel configuration. The performance expected in a Mac Pro would require an array of A11 SOCs.

  5. Lets look at some known factors –

    1) Apple now has a bunch of top quality professionals ( musicians, video editors and artists ) working in their Pro Workflows Group.

    2) The Pro Workflow Group are liaising closely with Apple’s hardware and software designers.

    3) Most users of FCP or Logic use it on pretty high powered Macs, such as the Mac Pro.

    4) The next Mac Pro is known to be at least a year away.

    My hunch is that Apple is planning to come up with a tightly integrated solution which is so compelling that it will leave everything else in the dust. If they decide that these systems are principally to be used with software written for Macs, then the argument for using Apple designed ARM processors becomes much stronger – especially if they adopt a blade approach where users can plug in as many boards as they feel are appropriate.

    Not much has been written about the Pro Workflow Group so far, but I’d love to know if any PixelMator developers are involved. A seriously beefed up PixelMator Pro, designed solely for ARM Macs could shake things up even more than doing it for Logic and FCP.

    I’ve yet to see any hint about graphics development, but it’s no secret that Apple is not impressed with the way that Adobe currently operates and generally speaking, when Apple doesn’t like something, they come up with an alternative.

    Apple is clearly taking time to produce the next Mac Pro. I hope that they are going to make it worth the wait and there are some hints suggesting that something amazing could be on the way.

    1. I fear you are right. In Apple’s close-knit world the only “professional” work they envision is producing graphics for the next album cover or post-production of the latest rap video.

      Apple seems content to ignore the other 98% of professional applications.

    2. An interesting take….

      But … isn’t a “tightly integrated solution” pretty much the definition of the 2013 Trash Can?

      After all, it was reported to have been great for FCPX … but where that all fell apart was that it wasn’t necessarily all that great for any other Pro’s workflow/needs.

      Similarly, while its good to hear that they’ve pulled in Video & Music – – there are other Pro venues that are going unmentioned.

  6. If they are going this direction there better be an announcement at WWDC in June. Otherwise, every professional app, except maybe Apple’s own, will not be able to run natively on this machine. You would be taking huge performance losses running a Rosetta style emulation layer on an ARM Mac Pro until everything is written to run natively. Apple better have really good developer tools ready to hand out if they want the machine to be a success.

    1. Because Apple knows that the new MacPro better be perfect on all levels, I strongly suspect that it brought on its development board the CEOs of all of the pro software so that the MacPro, when it’s released, would already have the updated software that real pros want and need. A boondoggle follows should it not have this in development already.

      1. Apple’s recent history says it will deliver a less than perfect product (assuming it will actually deliver).
        Mark my words: If and when Apple actually delivers a new Mac Pro, it will use a processor that has been available in Dell and HP machines for nine months. Apple is so far behind the ball, it isn’t even funny.

  7. If they are going this direction there better be an announcement at WWDC in June. Otherwise, every professional app, except maybe Apple’s own, will not be able to run natively on this machine. You would be taking huge performance losses running a Rosetta style emulation layer on an ARM Mac Pro until everything is written to run natively. Apple better have really good developer tools ready to hand out if they want the machine to be a success.

  8. I still think they should make it the most upgradeable thing on the market. Make a beautiful box (even a good sized cube (in honor of Steve Jobs), that not only will hold expansion cards, but also allow for the swapping out of the motherboard with processor, etc. Then, it won’t matter if there is no new box design for a while, and the Pros will be able to upgrade to their heart’s content. I am sure this would require some engineering prowess, but nothing beyond Apple’s reach.

  9. Why take this long? Box design? No. The next Intel chip? No. Some new I/O? No. Modularity? No. Storage? No. An Ivy-specified, red aluminum chassis? No. The engineers and designers in the Mac Pro division are busy smoking too much weed? No. (well, maybe a little.)

    What is the hold-up?

    It’s the new A13/A14 chip + OSXc. And how will it be deployed? It will be deployed in the form of small blade servers which the new Mac Pro will allow you to slide in as many fit in the box, while daisy chaining as many boxes as you like. In other words, each box will be a mini cluster super computer that is easily extensible.

    If all they end up doing is using the same hardware and software then they should be criticized for not having had a new solution for users several years ago. As everyone has argued, why not just jack up the cheese grater? That was a cool machine with lots of potential for new hardware upgrades… which could have been introduced in 2017.

    No, I say they are creating a cluster super computer using their new A chips. It seems to me the only logical choice.

    What?

    1. You might have something there. Taking 2 years to design a desktop is ludicrous, unless you are designing chips for it. It might be true they are switching to Axx.

    2. Even if Apple switches the Mac to ARM, it will not be in the form of an Ax chip. That’s an SoC designed for smaller form factor / mobile systems. Any Mac that’s ARM based would have a custom standalone ARM CPU that’s different than what’s offered on their SoC’s.

      I believe Apple took a step back to see where things were headed and decided since the last Mac Pro wasn’t going any where, it was time to push the Mac platform into a new direction. They scrapped several projects and products and started to work on something new.

      I also believe these new Macs will have not have Intel CPU’s. However, there are three possibilities as to what they could use…

      1. Custom Apple ARM CPU’s
      2. Custom Apple x64 CPU’s
      3. Custom Apple CPU’s w/new Apple ISA

      Those are ordered by the amount time it would take Apple to develop each.

      Personally, I think Apple is moving away from needing to license ANYTHING from anyone other than what’s considered, “industry standard”. Apple developing and designing their own CPU architecture seems like their next logical step.

    3. With Apple attempting to design the new Mac Pro using iPads and Watches, it will take years to come together.

      Jony is going to take years to choose the color of the synthetic microsuede faux leather handles.

      Schiller is busy trying to figure out how high he can push up the price. Gotta be a way to install a gimmick bar on the back.

      Angela is working on new hidden cubicles to be installed in the basements of Apple Stores so nobody has to see a Mac when they walk in the front door.

      Tim is using his social network to choose a manufacturing site that his bosses at Foxconn will accept.

      The incompetent cast of so-called hardware managers is busy tearing apart Dells to see how they design workstations.

      It all takes a lot of time away from Golden State games.

      Take it easy on these poor overworked self-rewarding multimillionaires. Don’t forget they have 3rd and 4th mansions to decorate.

  10. Well again, you’s guys commenting get it right. Sure know more about Mac Pro needs than me. I don’t even need a Mac Pro now, even though I used to. I’d like to see an easily RAM and SSD changeable macbook.

  11. Do you note how they raise a question a preposterous one at that and then for the very same preposterous reasons dismiss it. Must be a quiet day in that office. If Apple were really trying to forget about the Pro Mac it would be precisely because they were want to transition quite quickly to Arm based Macs i.e. knowing it would have no appeal they would have to keep it going on Intel for quite some time. So making it the first Mac to do so and within 18 months or less is just a ridiculous notion. It doesn’t mean it won’t have ARM based chips in it they could be used for all sorts of clever tricks including my own speculation about seamless use between Macs and iPads for graphic pro users but it really is fanciful in the extreme to imagine it will use them as its main processor … unless Apple have done the near impossible and make an Arm based chip(s) of immense power by that time that can equally as good as natively run Intel based applications. Not gonna happen.

    1. I agree that making a new ARM CPU which could beat or even match the current top end Intel CPUs is unlikely at the moment, but Apple is pretty good at designing operating systems, software and hardware which works well with multiple processors.

      My guess is that Apple will opt for a blade or modular design with multiple processors working together as one. Apple’s ARM chips would be especially desirable in that role as they are particularly efficient and therefore heat management becomes simpler when a lot of them are squeezed into a small space. I’m assuming that they would need to be closely spaced in order to provide ultra high speed signal paths, but there are other ways to achieve that such as optical interconnects.

      If you look at how Intel has improved their processors over the last five years and compare it to how Apple has improved theirs, it’s hard to make an argument that Apple’s best way forward is to stay with Intel. At some point, Apple is going to have to adopt a CPU which better suits it’s needs and I think that time is fast approaching.

      An incremental upgrade of the Mac Pro could be released fairly quickly, but we now know that it won’t be happening. Whatever Mac Pro Apple is planning will still be in operational use until 2025-2030, therefore it’s going tho be planned for the future, not planned for compatibility with the past.

  12. “Putting [only!] an ARM-based processor…is a risk,”

    There. FTFY.
    I would not be surprised if the Mac Pro has BOTH processors from Intel AND Apple’s own processors, perhaps handling different functions.

Reader Feedback

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