Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
"When Will Apple Release an Arm-Based Mac?"

I think there's an argument to be made that an iPadOS-based laptop is closer to reality than an ARM-based Mac.
 
Well, to be fair, the PS3 was a really complicated machine. The Cell processor was essentially a general-purpose PowerPC RISC core combined with six 'Power Processing Element' coprocessors, and in the original version, also had the entire chipset of a PS2 onboard for the backwards compatibility feature. It was really hard to program, even though on paper it was incredibly powerful. It was really only towards the end of the PS3's run that studios learned to take advantage of it.

To be fair, it's not the PowerPC per se. The Xbox 360, GameCube, Wii and Wii U were all based on PowerPC variants and we didn't hear about programmers having issues with them at the time... it's just that the PS3 was over-ambitious and over-complicated in its system design.

Well very true, but a lot of games in the past before console became the dominate gaming system they ran on PC systems so programmers could easily move a game to a x86 platform. The more the development time and cost, the less likely it would show up on a PS3 or the Mac game market in the day.
 
They can't, Intel owns x86, and they have only licensed it out to AMD. They show no interest in licensing it out further, and the license to AMD becomes void if someone buys AMD as well.

Sounds like the writing is on the wall for X86 being on the way out the door in the next 5 years

I wonder if intel/AMD is working on their own ARM chip... they definitely have the means to do so
 
Looking at the Mac Pro, the most expensive Mac (ever?), it’s clear that it’s geared towards high end users of Apple’s applications. Sure you can virtualize on them, but it wouldn’t matter to Apple if you could or couldn’t (they don’t even mention that on their product pages). Their goal is that there be no better way to run FCPX at incredible speeds.

Now, if they can produce an ARM processor AND a custom build of FCPX that runs circles around the highest spec’d Intel processor... for less than what an Intel machine would cost, you can BET those higher tiers are going to flip. ESPECIALLY if Apple makes it easy for plug-in developers to recompile for ARM.

Plus, it’s not like releasing an ARM Mac Pro in the future would kill the system anyone’s using. All your software, plug-ins, everything still just go on working. At some point in the future, they’ll drop Intel like they dropped PowerPC before it.

I like your thinking, but don’t forget Logic Pro X 😉
 
I run bootcamp for gaming myself but honestly, I don’t think Apple cares for that segment and would happily drop support for Windows in a heartbeat if the conditions are right.
 
  • Like
Reactions: MandiMac
only a small amount of first party software will run at an acceptable speed
The fact that ANYTHING is running at an acceptable speed is a pretty big accomplishment, especially if it’s fully compatible (many would switch out their work laptop for one RIGHT NOW because it’s light, lasts all day and can be online without tethering).

At this point, there are many companies in the world that would love ANYWHERE mobility, running the Windows and Office that provides amazing compatibility with their current infrastructure. In fact, they likely see it as a bonus that most stuff you don’t WANT folks to put on your network won’t run. :)
[automerge]1572998597[/automerge]
I run bootcamp for gaming myself but honestly, I don’t think Apple cares for that segment and would happily drop support for Windows in a heartbeat if the conditions are right.
I believe this statement to be correct.
 
When Apple switched to Intel, there was a massive increase in software from mainstream developers as they were comfortable with Intel. I fear a change to Arm will will cause the opposite to happen ...
How many developers do you think are developing for the Mac in assembly language these days? If the compilers function properly (and once you get a C compiler and the LLVM backend working on machine, all the other compilers fall into place), and the machines are fast, I don't see the instruction set being any sort of stumbling block for developers.

Another thing to keep in mind - when Apple switched from the 68k family to the PowerPC family, they went from one off-the-shelf part to another. When they switched from the PowerPC family to the x86 family, they again went from one off-the-shelf part to another. In both cases, for a given chip family, they had to pick and choose amongst the available models for ones that were the best (or least worst) fit for Apple's needs. But none of them were precise fits (witness Apple using "laptop" CPUs in some desktop systems, to meet their power/cooling requirements - and witness Apple having to act excited about years of tiny speed-bumps on the PowerPC family as IBM lost interest, with the faithful waiting in vain for the "Powerbook G5" that never appeared).

When Apple switches from Intel to Arm for Macs (no, it won't all happen at once), their chip supplier will be Apple. And if the machine designers or software developers really want/need the CPU to do a certain thing, hit a certain spec, all they have to do is convince upper management to tell the chip guys to do that - that's radically different from the previous generations where Apple could at best "suggest" that gee, it'd be really nice if IBM, or Intel, were to make a CPU with particular speed / instruction-set / thermal or power characteristics.

The iPhone designers have had this luxury for a few years (and it has made for some astoundingly capable phones). The Mac designers will get it before long.

I suspect Arm Macs will be pretty nice to develop on, and develop for, once the designers get to spec the Arm CPUs they really want.
 
Who is asking for thinner MacBooks?

All the many thousands of customers who purchased the discontinued MacBook because the Air and Pro were too heavy for their personal preference or use case.

To me, the best candidate for an ARM-based Mac would be a new iteration of the 12" MacBook.
I don't see any hardware issues here. The challenge is to remake macOS for ARM chips which, imho, is probably a bigger challenge than transitioning from PowerPC to Intel chips.

macOS and iOS use pretty much the same OS kernel. The rest of the upper OS layers were made multi-platform ISA-agnostic for the PPC to Intel switch long ago. So there is no big challenge. Unix(non-TM) on ARM is trivial, zillions of ARM CPU Raspberry Pi's have been sold and run Linux and Unix apps.

The only thing I’ll need to know if this comes to a future MacBook Pro, is can I still virtualise Windows? Non-negotiatiable, mission critical requirement.

Probably not the Pro at first. But Apple knows exactly what percentage of their customers run VMs on MacBooks, and the number is probably too small and shrinking for them to care about. Chase them off to Dell. Keep the higher profit margin Mac customers who just want to run basic Apple and App Store apps (including Catalyst ones), and soon all the newer 64-bit apps that can be easily ported (almost none require x86 assembly language code).

All the applications I use (including heavy-duty Fortran numeric stuff) can be recompiled to run on an ARM ISA. And an ARM-based 11" or 12" MacBook would be faster then one with a Core i3 or low-power i5, have better battery life, and either be cheaper or increase Apple's margins.

But the big win might be that Apple's Neural engine would be included with the Arm SOC, and thus be a lot faster for ML apps than the Intel GPU.
 
Give me like a 15" MacBook Air with ARM and passive cooling - that's the dream.

Not likely, but it's my dream nonetheless. 13" is just too small.
[automerge]1572999078[/automerge]
 
Sounds like the writing is on the wall for X86 being on the way out the door in the next 5 years
This is highly unlikely for two reasons: the enormous amount of software that runs on x86 (better: IA), and the fact that the ARM market is far more fragmented than the IA market. It's not only about CPU cores and instruction sets, but also about the infrastructure around them. Intel supplies an entire ecosystem around their CPUs including chipsets, standardized interfaces and buses, firmware SDKs and other things. This makes it easier to build IA-based systems, and is also the reason why you can easily build a single OS that runs on all computers based on this ecosystem (which e.g. enables things like Base Camp on Macs). For ARM, each OEM has to build their own platform around the CPU (which is e.g. why there is no single version of Android that runs on all Android devices).
I wonder if intel/AMD is working on their own ARM chip... they definitely have the means to do so
Intel has been and is using ARM architecture in various products (e.g. the Agilex FPGA platform).
 
It would be the end of the mac in the corporate world. Not being able to virtualize Windows is a deal-breaker for most companies. Corporate users are a very large and growing portion of the market.

And no, emulation isn't good enough - far slower, vastly more complex, and notoriously unreliable.

it'd also limit gaming on the mac to iPad games, and pretty much eliminate any chance (whatever's left since Apple moved to Metal) of porting games from Windows.

But then again, this would be a typical Apple move - all decided inside the bubble, without regard to the real

If they had two lines, the 'mac only' ARM, and the 'real computer' Intel, that would suck as a cross-platform developer...essentially doubling the QA work for a major application (think photoshop).

[automerge]1572999394[/automerge]
The only thing I’ll need to know if this comes to a future MacBook Pro, is can I still virtualise Windows? Non-negotiatiable, mission critical requirement.

By definition no. Virtualization, well, virtualizes the CPU. It doesn't emulate it (like Rosetta did). The latter is vastly harder to pull off with any kind of reliability or performance.
 
  • Like
Reactions: ssgbryan
It's not about if, but when. The unification of the app ecosystem is an obvious sign.
Imo the hardware is ready, at least for typical consumer use. But obviously the software, especially apps, have to be there as well. So it's just a matter of time. Once there are enough developers going with the unified Catalyst, I'm sure Apple would start releasing an Ax powered "Mac".
[automerge]1572999856[/automerge]
If they stop supporting a 2 year old machine for the transition they’re finished as far as I’m concerned. My wife’s 10! Year old Toshiba laptop runs the latest version of Windows 10 X64 absolutely fine. Apple have no excuse at all.
I have to concur. I can criticize Windows as much as I want, but I cannot deny its flexibility to run on old PCs. Even old Core 2 Duo PCs can run Windows 10.

Apple's vintage policy on Macs need to be revised. With SSD, even old macs run great. I have a 2012 MacBook Air that still performs fine for its use, and it would suck to see it being dropped from the next macOS just because of its age.
 
Last edited:
One of the biggest reasons I own a Mac is because I can use a virtual machine to run the other OSes I need. Mostly that is Linux but also Windows. If Apple moved to ARM this would no longer work. OK maybe I could run an ARM version of Linux but the older versions of Windows would not.

I'd have to abandon the Mac. Then I'd have no good reason for sticking with the iPhone. I'd be better off with Android.

Mac sales took off when they moved to Intel.
 
Why can't developers just check an ARM box, click the Build button, and release the ARM version of their product? After all, it shouldn't be harder than exporting a video to Youtube, right?

Ideally, that should be the case, but the real world is not so simple.

For starters, the byte order is different, so every time your app is loading binary data, the bytes will be in the reverse order, which may be a big deal. This needs to be addressed before we can recompile our apps. Chances are that those of us who develop for both macOS and iOS already have this sorted, but not everyone is free from this issue.

Another problem is OS features and functions. A big application doesn't stand on its own. For example, we have to deal with GUI, fonts, rendering, and so on, and many other system functions, which are often not portable. When a new OS comes out, it's not just the CPU and the architecture that changes, but the OS and its functions as well.

And finally, there are the 3rd party dependencies. If you rely on 3rd party software, and that is not ready for the new platform, you are not ready, either.

Then there's the small issue of QA, testing, packaging (digitally speaking, not actual cardboard). For example, x86 Windows requires completely different installation mechanism than ARM Windows. And no one knows anything about macOS ARM, because it doesn't exist, but you can bet it's not a simple check box to check.

Small devs with small apps will be able to move quicker. Large companies have more resources, but usually more complex software, and very slow decision making by conservative management. Even Adobe doesn't have infinite time and resources. The "lazy developer" comment isn't always appropriate. Enthusiast developers will move on very quickly. Tired old corporations aren't slow because of lazy developers. We work day in and day out, but we report to our managers. When you have a team of 100 developers, it's not like you can move overnight. 2 months of testing is routine as well.
 
  • Like
Reactions: Shirasaki
I have no doubt an arm-based Mac will allow Apple to provide a better experience overall, but I have two concerns: The first is running Windows/Parallels, how will that work, if at all? And second, will this mean that Apple will move to a controlled-app ecosystem like iOS?
Those are my two main concerns as well. Aside from the fact that a 8 year old Intel CPU powered computer is more capable today than a 6 year old Apple SoC device.

So maybe this new ARM powered Macs will be much more efficient, with a better integration and a better experience overall, but without the ability to install Windows, with a fraction of apps that we have today on Mac, and this computers will turn old after 5-6 years.

However I’m sure Apple will solve the app problem and let external apps to be programmed for Apple SoC, otherwise it makes no sense that Apple pushed security measures like the App Notarization, if they were to ban all the existing software for Mac outside the App Store.

I’m eager to see the next MacBook Air, and the first Macs with ARM SoC.
 
One of the biggest reasons I own a Mac is because I can use a virtual machine to run the other OSes I need. Mostly that is Linux but also Windows. If Apple moved to ARM this would no longer work. OK maybe I could run an ARM version of Linux but the older versions of Windows would not.

I'd have to abandon the Mac. Then I'd have no good reason for sticking with the iPhone. I'd be better off with Android.

Mac sales took off when they moved to Intel.
And I don't think Apple is that stupid either. I'm sure the current Mac lineup will continue.

My guess is Apple would create a new lineup, probably geared towards lay consumers first. They might not even call it a Mac.

It seems that Microsoft also wants to escape from the Intel world (see Surface Pro X). So Apple is not alone.
 
  • Like
Reactions: cppguy
No more Bootcamp? If Apple moves from X86 then we have to go back to the days of using Virtual PC ( a PC emulator) to run Windows apps. But the product no longer exists.
[automerge]1573001229[/automerge]
I'm excited about the transition as ARM chips get more powerful. Truth be told, a lot of computational heavy tasks these days seem to be be offloaded onto GP-GPUs. 68000 to PowerPC to Intel to ARM.... if anyone can pull off a transition of this magnitude it's Apple.

That being said, if the year was 1998-2010, i'd be more confident in Apple. Today, I worry that they might throw an ARM CPU into Macs and just to market an ability to run iOS Apps or the Mac with touchscreens or something along those lines.

You missed the CPU's that ran the Apple II apps and series of computers that was called ProDOS. That came before the 68K because early 68K macs had ProDOS chips in them just to run Apple II apps.
 
This article it's just blatant click bait, for things such this to happen, first has to have economical sense, the Mac is far away in sales from iOS devices, the savings providing it an advanced IPC focused arm Chip doesn't pay the development costs neither the software transition costs, buy leaves apple depending on a single Chip for a very diverse family of computers, it will be insanely expensive having 4 to 6 different desktop A+x chips, Buy further the risc for underpower configurations it's inaceptable leaves apple fighting against Intel and AMD, catalyst it's more about to bring new apps to the macOS ecosystem than to migrate old x86 apps to iOS
 
Not being able to virtualize Windows is a deal-breaker for most companies.
MOST companies? A lot, maybe, but I would wager that MODT of the macOS devices in use in companies never see much software outside of office and/or outlook. The fact is and always has been, that most people don’t come close to utilizing their system to their fullest potential.
 
  • Like
Reactions: Icaras
The only thing I’ll need to know if this comes to a future MacBook Pro, is can I still virtualise Windows? Non-negotiatiable, mission critical requirement.

At best you could virtualize the ARM version of Windows. Certainly not x86. And that's a big if. While x86 is very well defined, ARM is just a loose collection of designs. I'm not entirely sure if Apple's version of ARM would be 100% compile with what Windows ARM runs on. It could be, but it doesn't have to be. Windows is not Linux that you can just hack and build on any CPU. In fact, Apple has the ability to make huge changes to the design, so much that Windows no longer runs on it.

Contrary to popular belief, even Intel and AMD are not compatible. On the application level, yes. But on the operating system and virtualization level, no. Microsoft and VMware have written code such that it's compatible with both Intel and AMD. But is Microsoft going to modify Windows ARM to be compatible with Apple's CPU? That's very hard to tell. It's all speculation.

Let's suppose Windows ARM can run on Apple's CPU. Apple takes the time to write all the necessary BootCamp drivers. Even then VMware has to be completely redone from the ground up. Also VMware requires special CPU-level instructions that help achieving virtualization. If the CPU doesn't have those special instructions, getting VMware to work is a physical impossibility. We don't know if Apple's hypothetical ARM chip would have virtualization support or not. And we have no idea if VMware would be interested about recreating their software, and how long that would be. It also depends on the success of the new platform. So it's a chicken and egg problem. There's no software without the hardware. But it's not worth selling the hardware with no software on it. A transitional time is inevitable.
 
  • Like
Reactions: Shirasaki
MacPad with Smart Keyboard is coming!

yes. this.

i am currently using a MacBook Air 2018.
its a brilliant machine. quick, light. love the keyboard.

but my next machine i am targeting to buy in 2021 is definitely an Arm based iPad (Pro?) that has a track pad incorporated into the Smart Keyboard.

running Pages, Numbers, Keynote, and all the great apple written iOS apps.
when i look at my MacBook Air dock, the only non-apple app that i have is google chrome.
google chrome is needed because apple Maps sucks so bad.

waiting on that trackpad total integration and still further improvements in iPad OS.
by 2021 it will be there.
 
I have no doubt an arm-based Mac will allow Apple to provide a better experience overall, but I have two concerns: The first is running Windows/Parallels, how will that work, if at all? And second, will this mean that Apple will move to a controlled-app ecosystem like iOS?

Controlled apps would wreck it for me.
[automerge]1573002181[/automerge]
If they stop supporting a 2 year old machine for the transition they’re finished as far as I’m concerned. My wife’s 10! Year old Toshiba laptop runs the latest version of Windows 10 X64 absolutely fine. Apple have no excuse at all.

I can imagine that happening. Legacy Intel support for a few years and then be done.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.