Intel Expecting Apple to Transition to Custom ARM-Based Chips Starting in 2020

[Piggie]

Is this when reality hits hard?

So many saying, Apple rules with it's custom arm chips.
Forgetting it's only running iOS and apps designed specifically for this.

So we shall see just how storming fast ARM is when you load a full desktop MacOS onto it, and a full heavyweight app.

We'll then see if Apple's ARM design blasts past Intels CPU's then.

 

[Howard2k]

I don’t run Windows on my MBPro. I’m not super keen on the idea of an ARM transition, and after the T2 debacle, I’m sceptical, but from a technical perspective, this should work for me, if the pricing is sensible.

 

[ROGmaster]

Another one that does not understand what I wrote.

Well good for you then.

 

[gavroche]

Even if 100% of mac owners dual boot (they don’t), that is still MUCH fewer people than own iPhones. There are simply many more iPhones around than macs. Apple has reported the mac installed base as 80 million. Apple has reported the iPhone installed base as 900 million. That means at least 820 million iPhone owners don’t own macs (more or less, ignoring that some people have multiple phones and/or multiple macs). So 820 million is bigger than 80 million. So, again, the number of iPhone owners that do not own macs HAS TO BE LARGER than the number of mac users who use boot camp, because 820 million is more than 10 times bigger than 80 million.

It’s just math.

I think you are wasting your time arguing with someone that thinks "everyone" dual-boots. Or that the number of people who dual-boot could in any way surpass the number of iPhone users...
[doublepost=1550823841][/doublepost]

I didn't bother reading the 21 pages of Comment. But two things.

1. ARM has plans to enter high performance CPU design with a very strong roadmap. ( https://www.anandtech.com/show/13959/arm-announces-neoverse-n1-platform )

2. What will happen to this year's Mac Pro? Assuming it is still coming this year.

Unfortunately, this whole thread/topic is one in which a number of people reading it could probably enlighten us quite a bit... but are bound by NDAs and can't comment. So we are left with endless debates about not being able to dual-boot.

 

[alFR]

Have all the people moaning that "ARM will never be an i7" actually seen the benchmarks on the latest iPad Pros? They are almost as fast as the 6-core i7 based MacBook Pros, and that's in a thermally-constrained chassis. Also, back in the old days of ARM (when it stood for Acorn RISC Machine, google it), ARM processors in desktop machines handily beat the performance of their Intel / Motorola competitors. Anyone who thinks that it's not possible for an ARM core-based processor to perform as well as an Intel one with equivalent TDP must be smoking crack.

 

[Smartass]

Well, it'll be good enough for Facebook and Youtube, foe which people buy Macbooks in 99%.

 

[alFR]

The transition to ARM based chip will run a derivative of iOS natively

My bet is it won't: they'll port OS X to ARM. They probably already have, in fact.

 

[Reindeer_Games]

I think you are wasting your time arguing with someone that thinks "everyone" dual-boots. Or that the number of people who dual-boot could in any way surpass the number of iPhone users...
[doublepost=1550823841][/doublepost]

Unfortunately, this whole thread/topic is one in which a number of people reading it could probably enlighten us quite a bit... but are bound by NDAs and can't comment. So we are left with endless debates about not being asked to dual-boot.

Just everyone I know that has spent the money on a Mac dual boots (except one thats never used Windows on their machine)-maybe not in your circle. That's my 10+ years xp with Macs-assumptions of iOS upgrades/conversion's that haven't occurred yet aren't quite factual. When it happens-then it is factual.

Many times I have said-build it. I'll take a look at it, but that doesn't mean I'll buy it.

 

[WatchFromAfar]

My only concern with this is will this change still allow us to run Boot Camp and Parallels Desktop on ARM based Macs?

No; Windows is compiled for an x86 chip not an A-series chip. Even Windows on Arm won't run on an A-series chip as it is compiled for certain Snap-Dragon chips.

 

[Ries]

Nope! Not gonna happen. Here are a few reasons why:

1) No more virtualization of Windows on Mac.
2) No more directly running Windows on Mac (aka Boot Camp).
3) No way Intel is going to license their CISC proprietary technology to Apple to run on A-Series RISC CPUs.
4) Most Mac software would be dead-ended and would have to be substantially rewritten to run on RISC CPUs.
5) No more macOS = UNIX without a major rewrite of UNIX for RISC CPUs.
6) There is at this time no realistic speed advantage for moving from Intel CISC chips to Apple A-Series RISC chips.
7) No more GPUs for Mac without again a substantial driver rewrite, if that's even possible with the loss of proprietary Intel CISC technology. Apple has screwed things up badly for Mac GPUs already, locking out anything new by Nvidia, aka stupid move.

And there's more. But I find the above is enough to freak out most Mac users.

And please, those who don't understand the difference between CISC based Intel CPUs and A-Series RISC based CPUs, please read about the difference before making statements that prove your ignorance. Thank you. Oh and no, Apple's Marzipan project is entirely unrelated to this stupid rumor.

The real Question: Why does this stupid rumor come up year after year? IMHO it's re-perpetrated by those who have no idea of the technology shift involved or the damage that would be done to the Mac market if it happened.

1-3) Snap dragon is arm and runs windows 10 x86. Apple could add virtualization support, this not not imposible.
4) All modern CPUs are RISC, even intel. The CISC instructions are broken into RISC internally. This just requires a recompile and fixing of the odd pointer problem here and there. Normal desktop app don't use assembly, they are coded in C/C++/... and compiled to a architecture. It's only the OS and highly optimised code that needs to hjave parts rewritten (OSX already runs on ARM, it came from ARM). What would kill mac software would be low usage numbers (not worth the hassle for making it for max osx).
5) Already runs on ARM.
6) Correct, by it allows Apple to have control of the CPU/CPU and not being depended on Intel/AMD's releases/features. They'll be free to add features they deem "innovation/usefull". Imagine LTE support in the SOC and all the features your phone has build in.
7) Mac OSX got rewritten to metal same as IOS, apple has GPU drivers for their GPU, they'll work just fine powering Mac OSX UI. You won't get nvidia/cuda, but opencl still works just fine.

 

[laz232]

1-3)
4) All modern CPUs are RISC, even intel. The CISC instructions are broken into RISC internally. This just requires a recompile and fixing of the odd pointer problem here and there. Normal desktop app don't use assembly, they are coded in C/C++/... and compiled to a architecture. It's only the OS and highly optimised code that needs to hjave parts rewritten (OSX already runs on ARM, it came from ARM). What would kill mac software would be low usage numbers (not worth the hassle for making it for max osx).

I do not believe it is that simple. There are often hidden snags w.r.t. register size, endedness, integer lengths etc.
Adobe AAM and Firefox haven't even been compiled to 64bit macOS (or only recently) - and that's "just" a change to 64bit, same x86 arch...

 

[femike]

Bye bye Intel. Apple should ditch all Intel chips in their products. Apple should of stuck with Qualcomm while they sorted out their own modem chips.

 

[dwaite]

Some more level-headed approach to the matter and a possible and sensible timeline:
- announcement of Apple designed ARM chips coming to Macs at WWDC together with documentation etc for developers to prepare their applications for it. Showing off compat. layer for legacy applications.
- 2020 will see some smaller scale introduction of the chip (maybe in a new MacBook - not pro)
- full transition in 2021 when devs hopefully have migrated to ARM.

Oooh, I'll take a pass:

• WWDC would be a good time to announce it, but who knows if that would be this year or next.
• If it is not previewed until next year, I'd expect a new T3 chip that absorbs more hardware and offloads more OS tasks - for instance, USB+thunderbolt support, and background tasks like checking for software updates/push notifications. Networking might be absorbed into or be run through the T3 chip as well.
• I'd expect that with a WWDC preview, they would target a March first OS/hardware release (as a minor release of snow Mojave or whatever). Targeting xmas would be aggressive and give them little wiggle room in their biggest quarter if customer perception is against the new internals.
• I'd expect focusing on the MacBook first, which all of a sudden would make sense in the line-up as a distinct model from the Air.
• This also allows them to wow an initial "apple to apple" comparison based on the restricted power/thermals in the existing MacBook. It's the Mac that has the most room for improvement.
• I'd expect MacBook Air to follow, then iMac, MacBook Pro.
• Mac mini is a bit of a wildcard still - for all we know, they could target an early mini launch and sell the mini as a development preview machine after WWDC.
• I suspect the Mini, MacBook Pro, and other pro machines will sell both new arm models and previous intel models for a while.
• iMac Pro and Mac Pro are likely a few generations away. It will be hard to match Xeon and high-end discrete GPU performance without specialty designs, and pro users probably prefer to be late adopters (to give tools time to catch up.)

From a programming Point-of-View, I don't know how difficult it would be to compile native X86 applications to ARM (given, they don't use frameworks that do the job for them) but if it "just" making sure to move the intrinsics and inline asm stuff to ARM, it might be doable. But this is a very arm-chairy take on it.

You can already cross-compile apps between all platforms (macOS, iOS, tvOS, watchOS) - but do need to adapt to differing SDKs, system services, and user experience. The UI frameworks are quite different on purpose because the user interaction and human-interface guidelines are quite different - there will likely always be some barrier to an iOS app 'just working' on the Mac, without making any UI customizations whatsoever. But a lot of the SDK and system services differences at this point are Apple Tech Debt.

But that said, I think Apple knows that the process of making apps across all the hardware is too difficult, and they have a varying app ecosystem on each platform as a result. The iOS App Store has App of the day, while on Mac it is the App of the Week. Technology to focus on that is very likely to be a big focus of this year's WWDC. That is also why I have my doubts that Apple will bring up any sort of ARM transition this year.

 

[ph001bi]

I use an MBP for my consulting business because I can bootcamp Windows into a partition, and run some Windows-only business applications that my clients utilize (MS Project, MS Visio, etc). If I can't run them, I'll have to switch to a Wintel laptop.

May be good for the wallet as I won't have to pay the ever-increasing prices on mbp's. I'll just switch for an imac (have one that is 8 years old and still ticking) and iphone for personal use.

I have a Citrix VDI on my MacBook Pro so that I can run Windows apps. Mainly SAP in PC version, Internet explorer for these never updated company apps and Access. It works incredibly well and will continue to work on an ARM powered MacBook Pro.

 

[laz232]

I hope Apple moves to ARM soon. I believe Intel is yesterday's news. The ARM processors are getting very fast and the graphics capabilities are better than what we see in Intel-based thin and light laptops with Intel Iris Pro or Intel HD graphics. My MacBook doesn't have the muscle my iPad Pro has both in CPU and GPU performance. It's time to ditch Intel and move to ARM. Bring it on!!

to paraphrase: "it's the applications, stupid" (PS not an insult - paraphrase from https://en.wikipedia.org/wiki/It's_the_economy,_stupid)

I need to run programs that I use for work and pleasure (engineering SW, LibreOffice, Python, Win10 in VM) - and I don't think that the ARM can do that without a lot of hassle for the next few years as stuff breaks.

an iPad can never be the platform on which I design Radar equipment, satellites etc...

 

[Proji]

I think many people are missing something here:

What effect will this have on running Windows natively on the Mac (i.e. through Boot Camp)? I know not everyone does this, but many people do and I don't think this is going to be possible on an ARM based system is it?

I know there will be emulators that will pop up, there always are, but they consume overhead.

Windows has run on ARM in the past, Microsoft did it with the original Surface (not pro). The problem they had with the ARM surface was that developers didn’t port their apps across for it, so it ultimately languished and was finally removed from sale.

That was a few years ago now though, so things might be different this time around. Also unlike the windows App Store, the iOS App Store is loaded with applications and active developers so initially an ARM Mac would have lots of apps to work with, the real question is whether the desktop class apps get ported.

It’s an interesting move for sure and I totally get the logic behind it, I’m just not sure how I feel about it

 

[audiophilosophy]

If this happens, then let’s hope that in the process Apple doesn’t try to lock down 3rd party app installs through the Mac App Store only. If Apple decides to do something as boneheaded and arrogant as not allow its users to install apps outside of the App Store, then massive numbers of people will switch to windows. I could see Apple’s currently leadership deciding to go ahead with something like that. They’ve pushed their users into accepting sacrifices a lot of times, especially their Mac users in recent years. The Mac is certainly not as well off as it was about 10 years ago.

 

[Royksöpp]

Great! I can’t wait for the 2020 MacBook Pros to become obselete in 2 years.

 

[EntropyQ3]

Just everyone I know that has spent the money on a Mac dual boots (except one thats never used Windows on their machine)-maybe not in your circle. That's my 10+ years xp with Macs-assumptions of iOS upgrades/conversion's that haven't occurred yet aren't quite factual. When it happens-then it is factual.

Many times I have said-build it. I'll take a look at it, but that doesn't mean I'll buy it.

The only ones who really have data on the prevalence of dual booting is Apple (and possibly Microsoft).
I assume they wouldn’t make a transition if usage patterns showed major impact. We’ll see. This is just a rumor after all. However if they transition to ARM, they can simply emulate x86, and it will perform pretty well. I can’t believe there are a lot of people around who both dual-boot and whose Windows x86 apps are absolutely performance critical. Both conditions have to apply for there to be an issue.

 

[0924487]

My bet is it won't: they'll port OS X to ARM. They probably already have, in fact.

No it’s easier to rewrite OS X than to port it to ARM. They know it, OS X is a horrible piece of code and they have gave up development on it since Marvicks that’s OS X 10.8. After that to the current 10.14, they are all maintenance updates with no significant OS X specific updates. Yes, the Kernal is updated, the file system is updated, but those are not a part of macOS per se, they are shared code for all Apple OS.

All they need to do is to rewrite AppKit really. Also the Finder app. But the iOS would also need a Finder app, not just that useless Files app. Since APFS, a lot of finder functions have been rewritten from ground up, and they are ARM compatible.

You know what’s better? Since the conception of OS X, Steve Jobs demanded that the core function of OS X must be processor architecture independent. That means only AppKit need the update. Drivers, of course, but what do you expect?

 

[xyz01]

Now I am definitely delaying my purchase until later 2021. Interesting times ahead though. I hope Apple is working with Microsoft to build a port of Windows 10 for A Series with full desktop Office apps. Apple should get some developers on board to do the same for their apps: Adobe, AutoDesk, Intuit etc.

Office is not going to be a problem, you can even run it native for Macs. The problem is going to be for all of us who run VMware/Parallels to run special applications (often legacy or semi-custom enterprise applications) on Windows, and all of those apps that aren't updated anymore for Mac as well.

If this happens for all Macs, I'll have switch back to Windows/Linux on Intel. For those who just use email and a web browser, with the occasional word processing, ARM will be just fine - but it would fragment an already small user base.
[doublepost=1550829546][/doublepost]

Yes, that’s the way life goes. You buy a machine for the software that it can run at the time you buy it. At some point in the future, you may not be able to get updates or support for that software, unless you have a firm maintenance contract with a fixed term. What about poor me? I can’t get updates for my copy of Microsoft Multiplan for my TI-99/4A anymore.

It's more complex than that. If Apple says this WWDC that "next year, will migrate all our Macs to ARM - just look at these performance numbers", buying an x86 Mac will be a bad choice.

• For those who would be fine with the migration, they could just wait and get a better system.
• For those of us who can't do that move, it would be a bad choice to buy a Mac which Apple soon would stop caring about. The life expectancy on updates - both meaningful and security - would be much lower. The application situation could detoriate in the medium term. Thus, buying a Windows/Linux system instead would be a better choice.

Even having the uncertainty that this could happen makes the latter choice more appealing - "this rumour site says Apple would switch next year. Buying something very expensive that possible wouldn't last very long might be a bad idea".

 

[Coyote2006]

This is a good choice, as Intel can't deliver the needed chips in time. I wonder if we'll already see a MacPro with these chips. Or at least a MacPro that can be upgraded to them later. Modular and more ...

 

[scsjason]

I think many people are missing something here:

What effect will this have on running Windows natively on the Mac (i.e. through Boot Camp)? I know not everyone does this, but many people do and I don't think this is going to be possible on an ARM based system is it?

I know there will be emulators that will pop up, there always are, but they consume overhead.

You can get windows 10 for arm.

https://docs.microsoft.com/en-us/windows/arm/

 

[EntropyQ3]

Office is not going to be a problem, you can even run it native for Macs. The problem is going to be for all of us who run VMware/Parallels to run special applications (often legacy or semi-custom enterprise applications) on Windows, and all of those apps that aren't updated anymore for Mac as well.

These are exactly the kind of apps that emulation works brilliantly for. If older x86 processors could run them, there will be no performance issues running them emulated on the faster processors of tomorrow.

 

[Janichsan]

My only concern with this is will this change still allow us to run Boot Camp and Parallels Desktop on ARM based Macs?

Nope.

A completely worthless comparison without details about the export quality in the iPad version.