Apple planning to ditch Intel chips in Macs for its own custom silicon in 2020

Posted:
in Future Apple Hardware edited April 2
Apple is expected to ditch Intel's x86 architecture using its own chips in the Mac as soon as 2020, with the fruits of project "Kalamata" similar to a move that it has pulled twice before.




According to Bloomberg, "Kalamata" is in the early stages, and is part of a larger strategy to make Apple devices work with more integration than they currently do. At first glance, the project appears to be similar to the "fat binary" approach Apple has taken in the past, both with the shift from 68K processors to PowerPC, and then with PowerPC to x86 with Rosetta.

The transition is said to be "multi-step" -- but few details are known at present.

Prior to a shift in hardware, Apple plans to lay groundwork with software as part of its Project Marzipan initiative. Aside from a new code name, today's report offers little in the way of new information.

The prediction isn't revelatory, given Apple's history. Apple internally started the shift to PowerPC chips in 1991, with the transition happening for consumers a bit more than two years later. Additionally, the company had Mac OS X builds for Intel chips since nearly the launch of the operating system, with that shift happening about five years later.

Rumors that Apple plans to switch away from x86 to its own custom silicon have been circulating for a decade. Rumblings have been heard since at least since 2008, when the tech giant purchased chip designer PA Semi for a reported $278 million.

After specifying several iterations of ARM-based system-on-chip packages for use in early iPhone models, Apple purchased chipmaker Intrinsity in 2010 and released its own mobile processor design in the A4.

Industry scuttlebutt concerning inevitable integration in Mac followed in 2011, when a report claimed Apple would deliver a desktop version of its 64-bit ARM platform within one to two years, gossip seemingly backed up by a buy-in into a chip fabrication plant in 2013. That hearsay bore no fruit, but Apple's work toward a first-party solution continued.

In late 2010, Apple began a concerted effort to build out its CPU design group, a years-long project involving rounds of poaching, including former Texas Instruments engineers, and new acquisitions like efficient chipset maker Passif.

Other key moves include the purchase of a chip fab once owned by Maxim and the establishment of SoC-related research and development facilities in Israel and beyond.

Rumors of an ARM-based Mac cropped up again last September, when a report claimed the company was looking to cut back on its reliance on Intel.

Apple's A11 Bionic processor has a single-core processor speed of 4205, with a multi-core speed of 10122. The results are very similar in performance to the 2016 and 2017 i7 MacBook Pro for single-core performance, and the multicore performance of the original 15-inchMacBook Pro with Retina Display.
«13456789

Comments

  • Reply 1 of 176
    kpomkpom Posts: 601member
    Interesting. I didn’t expect that so soon. Windows on ARM emulates 32-bit x86 but not x64. With Apple’s push on 64-bit I wonder if they plan to emulate x64. 
    edited April 2 watto_cobracalebbenbekke
  • Reply 2 of 176
    rezwitsrezwits Posts: 571member
    Yeah and this is going to be a BREEZE for Apple because of experience, versus um... COUGH COUGH
    anton zuykov
  • Reply 3 of 176
    Yes!!

    bloggerblogwatto_cobra
  • Reply 4 of 176
    macxpressmacxpress Posts: 4,091member
    I saw this coming a mile away! You can see the writing on the wall and its not with invisible ink either. Apple is making very powerful mobile chips, custom controllers and security chips, etc. Its only a matter of time before it happens....and I still think a Mac mini would be a perfect test bed for this. Maybe even a new MacBook Air (called something different). 
    cornchipwatto_cobra
  • Reply 5 of 176
    djames4242djames4242 Posts: 470member
    I've been able to run Macs for work because of the ability to virtualize Windows. Not sure this bodes well for the future of Macs in certain business segments unless emulation performance under these new chips will be acceptable.
    xzurazorpitbonobobasdasdviclauyycelijahglostkiwidysamoriawozwozwatto_cobra
  • Reply 6 of 176
    Writing has been on the wall for the past 2-3 years.  Wouldn't be surprised if it is discussed at WWDC next month.
    watto_cobra
  • Reply 7 of 176
    mdriftmeyermdriftmeyer Posts: 7,100member
    As a NeXT/Apple alum you folks are blatantly ignorant of the meaning of Fat Binary. Fat binaries were the binaries of NeXTSTEP/Openstep that were built binaries of the OS to run natively on different hardware architectures instruction sets.

     Apple continues working on shoring up the custom ARM based CPUs of its own design and still licenses the IP in order to produce them has nothing to do with leaving macOS to fend for itself on ARM based only instruction sets.

    More importantly, the effort to create OS X even with decades of x86/PPC/Moto/SPARC expertise took 5 years to get a limped version out the door, and that was already with a platform native on x86. The Rosetta was a compatibility layer on top of it.

     The logical solution moving forward is for Apple to license IP from AMD to have them build custom ASIC designs of SoC APUs and use their discrete CPUs/GPUs with the upcoming Thunderbolt licensing [now royalty free] to have a custom Thunderbolt controller designed by Apple on their boards, that are compatible with AMD's x86 chipsets, thus freeing Apple from relying solely on Intel.
    edited April 2 kpomchiaJonInAtldewmerepressthisfastasleepbrakkenTomElostkiwih2p
  • Reply 8 of 176
    curtis hannahcurtis hannah Posts: 1,617member
    This has been rumored for awhile. While I can believe this in two years. Interestingly that would be the likely year of another MacBook Pro redesign, so will they use the same chips as iOS devices or something else created by them.

    I also feel like by the vagueness of this rumor, it’s more likely an iPad Pro that runs Mac OS then a MacBook necessarily switching off intel, but I feel they would switch the regular MacBook and not the MacBook Pro first, which askews the update cycle.
  • Reply 9 of 176
    This means that ARM has almost caught x86. We can expect the A11X to be significantly greater in performance over the A11. I would guess and say at least 40-50% improved. In which this sets the stage for the A12, A12X, then to the A13/A13X (2020), in which Apple will rename the processors differently. My guess by 2020 we should expect at least 6500-7500 single core with a multiscore of 45,000-50,000. This is bound to happen...
    fastasleepwatto_cobra
  • Reply 10 of 176
    1- end of hackintoshes.
    2- should give Apple some extra room to improve security.
    3- how much you want to bet we will see fat binaries again?


    viclauyyc
  • Reply 11 of 176
    macxpressmacxpress Posts: 4,091member
    I've been able to run Macs for work because of the ability to virtualize Windows. Not sure this bodes well for the future of Macs in certain business segments unless emulation performance under these new chips will be acceptable.
    There is a technology called x86 on ARM...perhaps this is your answer. Microsoft looks to be supporting this as well. 
    razorpitmercelwatto_cobra
  • Reply 12 of 176
    SoliSoli Posts: 6,739member
    This push to eliminate x86_32 so that only x86_64 exists does open things up to have fat binaries for macOS with the addition of AArch64. We already know that even a power efficient A-series chips designed for the iPad and iPhone are besting several of the expensive Macs. I can't even imagine what we may see out of the gate in terms of how the UI "feels" if Apple made an SoC designed for the Mac that is close to 3 GiHz, has 8+ GiB of RAM, and all the other inclusions that a traditional PC needs. 

    kpom said:
    Interesting. I didn’t expect that so soon. Windows on ARM emulates 32-bit x86 but not x64. With Apple’s push on 64-bit I wonder if they plan to emulate x64. 
    They certainly could, but I don't think they have to. With the advent of the Mac App Store and all the advancements they've been making to make coding easier to developers to support multiple architectures I'm guessing that it'll be a fairly simple process for developers to compile for an Intel and ARM Mac, and I assume that it's not some all-or-nothing event that will wipe out Intel Macs. I expect we'll see ARM-based Macs for the low-end with the high-end being Intel Macs.
    edited April 2 watto_cobra
  • Reply 13 of 176
    tipootipoo Posts: 829member
    Hopefully they can decouple themselves from Intels blunders and delays and really distinguish themselves that way. Look at everyone pooh poohing the 16GB limit because Intel doesn't support LPDDR4, while A series chips have for a few years, yet Apple gets the blame for it. And the fully Apple GPU will be just as interesting.

    Question is then if they can avoid falling behind with such blunders on their own, but their A series execution has been excellent year over year. 
    SolirazorpitrepressthisStrangeDaysfastasleepbrian greenwatto_cobra
  • Reply 14 of 176
    Mike WuertheleMike Wuerthele Posts: 2,349administrator
    As a NeXT/Apple alum you folks are blatantly ignorant of the meaning of Fat Binary. Fat binaries were the binaries of NeXTSTEP/Openstep that were built binaries of the OS to run natively on different hardware architectures instruction sets.

     Apple continues working on shoring up the custom ARM based CPUs of its own design and still licenses the IP in order to produce them has nothing to do with leaving macOS to fend for itself on ARM based only instruction sets.

    More importantly, the effort to create OS X even with decades of x86/PPC/Moto/SPARC expertise took 5 years to get a limped version out the door, and that was already with a platform native on x86. The Rosetta was a compatibility layer on top of it.

     The logical solution moving forward is for Apple to license IP from AMD to have them build custom ASIC designs of SoC APUs and use their discrete CPUs/GPUs with the upcoming Thunderbolt licensing [now royalty free] to have a custom Thunderbolt controller designed by Apple on their boards, that are compatible with AMD's x86 chipsets, thus freeing Apple from relying solely on Intel.
    Before you throw stones, you'd better look at your history, and see what else the term was also applied to.

    I'm aware of your definition, and the usage you cite. However, there are more.
    lordjohnwhorfinmaltzking editor the grateDeelronrepressthisosmartormenajrtokyojimufastasleepbshankronn
  • Reply 15 of 176
    eightzeroeightzero Posts: 1,921member
    2020? I thought Apple would be long gone before then. I heard they were doomed. 
    SpamSandwichandrewj5790mwhiteviclauyycDeelronrepressthisrevenantStrangeDaysdocno42cornchip
  • Reply 16 of 176
    bloggerblogbloggerblog Posts: 1,724member
    This is great news, and it's been expected. However, news like this causes harsh friction between Apple and Intel. Remember when Apple decided to use ARM in the early days of the iPhone and Intel freaked out calling any company would be "stupid" to use RISC, and started flashing around odd shaped devices resembling the iPhone on stage?
    watto_cobra
  • Reply 17 of 176
    bkkcanuckbkkcanuck Posts: 705member
    kpom said:
    Interesting. I didn’t expect that so soon. Windows on ARM emulates 32-bit x86 but not x64. With Apple’s push on 64-bit I wonder if they plan to emulate x64. 
    This is not what they are talking about.  Windows on ARM is a poorly planned hack to have the processor emulate x86 (and currently only 32 bit) and not actually running on the processor.  The use of LLVM, the app store supporting applications that are independent of the processor architecture and the move to a rumoured fat binary as part of a unified app store are likely all pieces of supporting both ARM and Intel.  Apple may be likely to use ARM on laptops - but they likely will still use other (likely Intel) on the higher end computers like the Mac Pro for longer than 2020.
    razorpitasdasdrepressthisfastasleep
  • Reply 18 of 176
    tipootipoo Posts: 829member
    creemail said:
    This means that ARM has almost caught x86. We can expect the A11X to be significantly greater in performance over the A11. I would guess and say at least 40-50% improved. In which this sets the stage for the A12, A12X, then to the A13/A13X (2020), in which Apple will rename the processors differently. My guess by 2020 we should expect at least 6500-7500 single core with a multiscore of 45,000-50,000. This is bound to happen...

    5x in just two years sounds pretty optimistic, even with how impressive they've been. Say they manage another 80% MP boost every year, we're more like at ~3x at 30K multicore if it manages very good scaling, but we've more often seen one year with a big boost, one year with a minor boost, alternatively. 

    Now, this is all without active cooling so far, so if we're talking about a clamshell or desktop with active cooling on top of their already impressive uArch designs maybe. I wonder what a Xeon W equivalent A series would look like...
    edited April 2 repressthis
  • Reply 19 of 176
    SpamSandwichSpamSandwich Posts: 28,818member
    Presumably they’ll unify iOS and macOS to a degree that the chipsets powering the entire product line are largely the same?
  • Reply 20 of 176
    SoliSoli Posts: 6,739member
    Presumably they’ll unify iOS and macOS to a degree that the chipsets powering the entire product line are largely the same?
    I would expect the chips to be distinct, and that iOS and macOS will never be unified. There will be crossover as we've seen from the start with macOS being stripped down and then rebuilt back up into iOS and efficiencies from iOS being brought into macOS, but I think that sharing—not unification—will continue to be at the core of these discrete OSes.
    repressthisStrangeDaysdocno42brian green
Sign In or Register to comment.