Five barriers that might hold Apple back from moving Intel Macs to custom ARM chips

1246

Comments

  • Reply 61 of 102
    Dan_DilgerDan_Dilger Posts: 1,583member
    Quote:



    Originally Posted by cseeman View Post



    I can't take any article seriously that talks about leaving Intel without also commenting on Intel's Thunderbolt. I don't think Mac market would survive a rapid apandonment of Thunderbolt.

     

    Thunderbolt is PCIe + DisplayPort, managed by a proprietary Intel chip. It originally aimed at offering high end data connectivity outpacing similar external buses ranging from high end FibreChannel and InfiniBand to the more prosumer FireWire, as well as eSATA and SAS. 



    USB 3.1 now cheaply achieves the speed of original TB. Today's TB 2 is 2x as fast, and TB 3 should again double speeds, with the option to use fibre optic cabling. TB is the high end, pricier, all around best interconnect/data bus for virtually everything. It's essentially a port that acts like a PCIe slot.



    If Apple were to embark upon efforts to migrate its existing Macs to ARM, TB would be difficult to port over because it's designed to be a proprietary toll booth that sticks users to Intel's architecture. So it's a factor in preventing Apple from easily moving to AMD x86 chips. 



    That said, it wouldn't be that hard to develop a new interconnect to replace today's TB with something that's faster and even better. TB was developed before Apple had started selling 70M iPads per year. If anyone can develop a superior replacement for TB, it would be Apple. 



    The entire issue of TB could be included in barrier #5: one of the many disruptions a move to ARM would cause for 3rd party developers. There are lots of less visible issues as well, including, as others have mentioned, software that directly targets other Intel technologies such as SSE.  

  • Reply 62 of 102
    Apple is not worried so much of the cost of Intel Chips. After all, Apple will pay what it costs for what it wants.

    Apple wants performance.

    And certainly in the area of mobile, ARM chips have completely swamped Intel chips.

    But on the desktop, no ARM chip can compete against an Intel chip.

    Every Apple transition has always been with higher performance. Apple will not go to slower chips and slow performance in any desktop transition or laptop transition for that matter.
  • Reply 63 of 102
    wizard69wizard69 Posts: 13,377member
    [B][/B]

    DED gets it wrong again, the primary reason to go to ARM would be control over the entire silicon die. This to me is fundamental to Apples long term success. Of course they could start building custom chips with Intel as many other customers are doing.
  • Reply 64 of 102
    wizard69 wrote: »


    DED gets it wrong again, the primary reason to go to ARM would be control over the entire silicon die. This to me is fundamental to Apples long term success. Of course they could start building custom chips with Intel as many other customers are doing.
    Fundemental to Apple's success? So they have been a failure so far?
  • Reply 65 of 102
    tyler82tyler82 Posts: 1,105member
    Looks like I'll be hanging on to my 2008 MacBook for at least a couple more years :/
  • Reply 66 of 102
    wizard69wizard69 Posts: 13,377member
    appex wrote: »
    True full x86 compatibility is a must. And that means Intel. Full stop.

    That is an extremely small concern from an even smaller minority of Apples customer base. Apple could easily placate these people by keeping the "Pro" hardware i86. The proof is in the IPad which has been a runaway success with ARM inside.

    Hey don't misunderstand me, I bought a MBP in 2008 specifically because it could run Windows natively. I ended up never running Windows on that laptop and as time has gone on I've abandoned all interest in running Windows on the box. Times change and needs change, i86 just isn't a check off item anymore.

    Rather I look at it as what can the box do for me capability wise that is as good or better than a current MBA. Along these lines it would need to support Java, XCode, all of the basic Apple apps, eclipse (via Java) and GCC. Most of my other required apps would be a compile away.

    In other words i86 means nothing these days. Full stop????????.
  • Reply 67 of 102
    wizard69wizard69 Posts: 13,377member
    Lol no.
    Im with you there.
    Excel for OS X will be feature compatible with the Windows version in Office 2015. You're also missing the fact that MS ported a 99% feature complete version of full-blown Office to ARM for the RT devices. With a few tweaks, that version becomes 100% feature complete.
    Im just not seeing a huge demand for excel on Mac anymore. It maybe a problem for some no doubt there, but it isn't a big factor for most of Apples customers anymore.
    AMD is a joke anymore; people say they want x86 power, then AMD is not your option. They don't have the talent nor the cash to compete with Intel anymore.

    Actually AMD isn't as bad as many make it out to be. Their next processor coming out in a few months is in many ways a better choice than Intels solutions if you are not wrapped up in the i86 mantra. The fact is they have taken cues from Apple and built in a lot of custom hardware to make it a low power high performance chip for specific tasks.
  • Reply 68 of 102
    Dan_DilgerDan_Dilger Posts: 1,583member
    Quote:

    Originally Posted by wizard69 View Post




    DED gets it wrong again, the primary reason to go to ARM would be control over the entire silicon die. This to me is fundamental to Apples long term success. Of course they could start building custom chips with Intel as many other customers are doing.



    What, specifically, are you saying was wrong? Because the article outlined "Apple might want to own the core silicon technology in Macs" as #3 of 3 reasons why Apple might consider leaving x86.

     

    Is it possible for you to offer any comment, particularly one in agreement with or taken right from the article, without prefacing it with a generic personal attack on the author?  

  • Reply 69 of 102
    sandorsandor Posts: 659member
    Quote:

    Originally Posted by Misa View Post



    But as I said in the last thread on this topic. Apple would need to triple the clock speed of the A8 to 4.5Ghz to match the single-thread capacity of Intel's best offering

     

    The MHz myth is still alive and well.

     

    Clock speed is not a comparative metric of computing performance across chip types.

     

    Remember the days of 266 MHz G3 PPC processors outperforming 400 MHz PII's?

    I do.

  • Reply 70 of 102
    pfisherpfisher Posts: 758member
    Quote:

    Originally Posted by gerry g View Post

     

    this is a dumb article and a dumb hypothesis, high end users be they Apple or Microsoft customers have been loosing out of late due to Intel shifting focus to the low end of the market where the volume and the money is, Apple replacing risc with cisc will only be of benefit to the low end market these chips will never replace Oct core sever processors and we at the high end will be left hanging. Worse, if  the OS transitions to the new architecture and all the high end pro apps are still on Intel I predict a huge mass exodus of high end professional users migrating to Windows just as fast as they can.


    It wouldn't matter if high end users migrated: They would have two machines, and they would keep their Macs for home use and Windows for work.\

     

    If the high end market is only 5% of the sales, why let that hold you back? Just playing devil's advocate here. What if...

     

    Just have two platforms. Mac OS X and iOS laptops and so on. They you'd keep your Mac OS X people and wean most of the Windows users off that platform. Most iOS users use Windows.

  • Reply 71 of 102

    Apple doesn't have to move Macs to ARM. Don't waste the time, money and man-power on it. Tablets and phones are moving to take over popular computing devices. Macs and PCs will be a regulated to small niches; such as pro and enterprise uses. That is why Intel is trying so hard to push Atom into tablets.

     

    At the rate Apple is pushing the A-chips, it will be more powerful than any  x86 in a few years. You can do all the pro and enterprise in iPads. 

  • Reply 72 of 102
    Great feature! Thank you!

    Apple had to start somewhere. Power consumption versus Processing horsepower made the decision for them. Intel was obviously the only way forward in 2003-2004, and they made the transition beautifully in 2006-2007. A-Series chips were well past the cocktail napkin stage by then. Long-term, I see no loyalty or rules. The x86 paradigm, like the ARM paradigm, are not the only options.

    If you wanted to optimize all factors, then you'd invent a new paradigm, which means a new architecture that serves all mobile and desktop needs. I can't help but think Swift is a foundational clue of what is to come. Apple has the money to design, manufacture and implement an architecture that supersedes x86 and ARM. Imagine if you will a leapfrog architecture, leaving Intel, AMD, etc in the dust. Imagine Windows being ported to that architecture out of shear necessity. That would be a bold move, in keeping with Apple's legacy.

    Why not? Anybody?
  • Reply 73 of 102

    What, specifically, are you saying was wrong? Because the article outlined "Apple might want to own the core silicon technology in Macs" as #3 of 3 reasons why Apple might consider leaving x86.

    Is it possible for you to offer any comment, particularly one in agreement with or taken right from the article, without prefacing it with a generic personal attack on the author?  

    That's kind of you to come to the author's defense, but it is not necessary. Let DED speak for himself, if he chooses to join the forum community.
  • Reply 74 of 102
    Misa, how do you know that a triple-clocked A8 processor would match the single-thread capacity of Intel's best offering? That sounds like a fishy comparison to me. Intel CPUs are built on an entirely different architecture, with different instruction pipelines, processor subunits, etc.
  • Reply 75 of 102
    This will not happen any day soon if ever. In addition to the risks the author mentioned, some other issues suggest Apple would not bother, including the "Tyranny of Numbers" problem fabricating processors of higher complexity brings, which is something Intel has mastered to a degree no other chip maker has with the same success.

    Another factor is Intel's deep technology R+D, patent portfolio, and design library, which is part of what customers get for the price premium and something not duplicated at lower cost - the alternative is to do without.

    Apple may talk about ATM as an alternative to make Intel work harder and as a negotiating tactic, but if spinning out designs of high performance ARM processors competitive with x86 at volume were an easy task it would already be done.

    More likely would be closer collaboration of Apple and Intel on future processors: Intel customizes chips for companies like Google (server chips) and has been touted as one possible fab to replace Samsung - not crazy since Intel brings lots of IP and even better fab process to the picture.

    I don't buy Apple taking on the hardship for what are low end notebooks as long as they generate good profits. Tim Cook is smarter than that.

    But they will let the rumor mill churn as long as the press spells thier name correctly.
  • Reply 76 of 102

    I don't think they'll transition the Mac, at least not yet. The mere rumour and capability of transitioning will keep Intel providing cheaper chips to Apple too.

     

    On the other hand, I do think this can be looked at from the other angle, as an iPad with a keyboard. Call it the 'iBook'.

     

    Touch screen, 64GB SSD, iOS, Retina Display, bundled with Pages, Numbers. Many apps - Email, Safari, Facebook, MS Office available. A 13" widescreen is the same height as a 9.7" iPad screen... just wider.

  • Reply 77 of 102

    There are actually other options Apple could pursue with A8 chips etc.

     

    * my comment above - an iPad with keyboard could look like a thinner MBA.

     

    * Take an AppleTV and combine it with a retina iMac... and you get a 4K smart TV. With iOS derived apps of some type?

     

    * Terminals... I've always wanted to buy ONE Mac, and have 4 terminals. The terminals could use ARM chips.

     

    None removes OSX. 

  • Reply 78 of 102
    MacProMacPro Posts: 19,773member
    Thunderbolt is PCIe + DisplayPort, managed by a proprietary Intel chip. It originally aimed at offering high end data connectivity outpacing similar external buses ranging from high end FibreChannel and InfiniBand to the more prosumer FireWire, as well as eSATA and SAS. 


    USB 3.1 now cheaply achieves the speed of original TB. Today's TB 2 is 2x as fast, and TB 3 should again double speeds, with the option to use fibre optic cabling. TB is the high end, pricier, all around best interconnect/data bus for virtually everything. It's essentially a port that acts like a PCIe slot.


    If Apple were to embark upon efforts to migrate its existing Macs to ARM, TB would be difficult to port over because it's designed to be a proprietary toll booth that sticks users to Intel's architecture. So it's a factor in preventing Apple from easily moving to AMD x86 chips. 


    That said, it wouldn't be that hard to develop a new interconnect to replace today's TB with something that's faster and even better. TB was developed before Apple had started selling 70M iPads per year. If anyone can develop a superior replacement for TB, it would be Apple. 


    The entire issue of TB could be included in barrier #5: one of the many disruptions a move to ARM would cause for 3rd party developers. There are lots of less visible issues as well, including, as others have mentioned, software that directly targets other Intel technologies such as SSE.  

    Agreed. Plus the original copper-based version of the Light Peak concept was co-developed by Apple and Intel. So it is not as if Apple is not already well versed in the technology. Apple seem to have done a lot to help Intel of the years if memory serves, only for the PC world to benefit eventually. If Apple ever severed the Intel ties it would be a pretty damaging blow to both Intel and the already dying PC world I would think.
  • Reply 79 of 102
    One of the funniest %u201CApple fanboy%u201D arguments I%u2019ve read in a long time:

    %u201CHistorically, while Apple doesn't waste money frivolously it does not appear to have a problem in paying a premium to gain access to superior technology. Apple buys the best display panels it can source, it paid a premium to acquire AuthenTec, and it licensed the real Helvetica. Microsoft and Google are notorious for accepting low quality displays, both passed on the idea of expensive fingerprint scanners and both sourced second-rate knockoff versions of Helvetica: Arial for Windows and Roboto for Android.%u201D
    %u2014 http://appleinsider.com/articles/15/01/16/five-barriers-that-might-hold-apple-back-from-moving-intel-macs-to-custom-arm-chips

    I really wish to know how a typeface design that is 50 years old, has been diluted over decades in numerous %u201Cconversions%u201D across different media and has seen virtually no comprehensive design supervision since the 1983 %u201CNeue%u201D release can be considered %u201Csuperior technology%u201D.

    While the 2007 %u201CHelvetica World%u201D effort by John Hudson, the 2010 %u201CNeue Haas Grotesk%u201D effort by Christian Schwartz and 2011 %u201CNeue Helvetica eText%u201D effort by Akira Kobayashi are all commendable, they are neither used by Apple, nor do they %u201Cclean up%u201D the mess Helvetica is today. Each of them (and also the %u201CNeue%u201D series) has some good bits and some poor bits. It requires a lot of effort to pick the good fruits from the Helvetica ecosystem, and Apple hasn%u2019t even done that. A large number of typographically-aware designers and type experts have criticized Apple for %u201Cgoing Helvetica all the way%u201D, and the fact that several apps were made that replace the Yosemite UI-default Helvetica Neue with something else (for example the old Lucida Grande, but also other fonts) are a strong sign that citing Apple%u2019s choice as a great one is very naive.

    The most recent %u201CSan Francisco%u201D effort by Apple (the fonts used in the Apple Watch) are a step in a good direction, but it%u2019s actually much easier to point out that San Francisco is an %u201Cexpanded and improved clone%u201D of Christian Robertson%u2019s Roboto than to claim that Roboto is a knockoff version of Helvetica.
  • Reply 80 of 102
    Quote:
    Originally Posted by rcfa View Post



    People don't get it: NOBODY ports anything to a CPU except the OS and compiler engineers.

    The OS (Darwin) and the compilers for ARM are already finished otherwise there were no iOS software.

    Apple's MACH/BSD kernel in Darwin always is and always has remained multi-architecture. It doesn't matter if it's PPC, x86,

    x86-64, ARM-32 or ARM-64.



    In essence: OS X for ARM is but a recompilation for Apple, plus a few device drivers, if they run it on hardware using different/additional components than the iPads. Heck, I'd be surprised if that had not been running in Apple's labs for quite some time.



    Transition thus, for OS X native apps is recompilation and some QA, nothing more, except for an absolutely tiny fraction of the market that still might use assembly code for some small aspects of their code, but that's negligible.



    A move to ARM wouldn't be about cost, but about performance: for many years now (GCD, OpenCL) Apple has been telling developers to increase performance with multithreaded code, and to stop relying on increased clock speed and similar tricks to increase single threaded performance, which simply starts hitting physical limits, particularly in the performance-per-Watt measure.[...]




    While an hypothetical A9 or A10 with a clock speed not limited by iPad/iPhone heat dissipation limits and battery limits might have a lower single-core performance than an equivalent Intel Core i3 CPU, Apple could have a 12+ core ARM chip in the power envelope of a dual-core or quad-core Intel chip; so we're talking 12+ cores vs. 2 to 4 cores (4 to 8 threads with hyperthreading) and that could be able to match performance while scaling power usage by turning off anywhere from 0 to 11 cores, while still being able to do low CPU tasks like web browsing or even video streaming if decoding is done in the GPU.



    So not only would Apple be able to save costs, it would be able to offer battery life like no other in the industry.



    [...]

    2 Yes, even a MacPro could go ARM: one could have 64 cores where there are now 12, and the GPUs could be just as powerful as what there is now, as these are not directly depending on the CPU choice.



    3 So really, the only real obstacle is x86 emulation for running legacy Windows apps.

    With more and more corporate computing being cloud based, and with a more OS agnostic M$ (Ballmer is gone, as is evidenced by M$ Office for iOS), this may be less of an issue than ever; and possibly, the streamlined ARM-64 instruction set might be particularly well suited for CPU emulation, or for Macs, Apple could ditch the ARM-32 legacy instruction set and use the freed silicon for

    some special instructions needed for efficient x86 emulation.

    your first point is wrong.  Every developer would have to recompile their apps into fat binaries, to a) avoid the emulation performance tax, and b) any app worth their snuff will likely run into 'issues' running in emulation mode [been there done that].

     

    your second point power/performance/heat...  brings up the next point of recompilation.  a single threaded app on a quad core will perform slower on a octocore that is 1/2 the clock speed.   and an ARM chip clocked up will be hotter and more power hungry.   There is no free lunch here.  It may be slightly more efficient, but the at best, we need a better compiler to decompose your app into parallelism, and some thoughts by the developer to better thread their apps.   And then you'll have a hot/fast power hungry set of chips.

     

    And...  I partially disagree on your 3rd point. we had decent emulators (slow) on PowerPC/68K of the x86 line.  Slow, but functional.  But things like BootCamp and dual booting would be right out, and in essence, the mantle  of 'the best Windows Laptop is a Mac' would have to be passed.  And a majority of 'power users' (one who would pay for such a beast), live in a world where 80% of their core work is Windows based.

     

    I think DED's point of maintaining a dual path OS supporting fat binaries as a fork of resources that Apple doesn't want to pay on the Mac line is an valid and probably the most compelling argument, but I still think it's a necessary one to pay if you take a 'long view' of technology and controlling your destiny.

Sign In or Register to comment.