Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
ever tried to port a complex application across cpus? Especially from a cisc to risc platform?
Yes, numerous times. Across numerous different architectures, CISC and RISC. It certainly helps one find foolish assumptions one has made in the code. You learn to write more portably after a few of those.

Its rarely cost effective to do port from windows to Mac even on intel. Across cpu families, it's likely no cost effective At all.
Now you're talking about an entirely different subject which does not have any bearing on the subject at hand (I would also point out that juggling chainsaws is extremely hard, and that also has no bearing on porting between chip architectures) - moving from one OS to a completely unrelated OS is a giant pain - every system call has to be reconsidered and one has to work around the ones that are present on one system and missing on the other, to say nothing of rewriting all the UI code to some new API. But that's nothing like what a macOS-on-Intel to macOS-on-ARM port would require. Those are two entirely different tasks.

We aren't talking about moving to a different OS (which is what you're offering as a counter argument), we're talking about moving to a different chip architecture on the same OS. The only time moving from one architecture to another would require really massive rewrites is if you've written the whole application in assembly language (in which case I would question your sanity).
 
I'm tech savvy but have a limit..

Would someone explain to me what are the implications of this?

Here's some context with why I (average user) care about it: I've been wanting to buy a MBP for years now, waiting for that combo (redesign, hardware related issues control i.e keyboard, etc) just generally the "next era" MBP. With the rumored comeback of the magic keyboard, I'm inclined to buy this next one (sad the 14 inch wishful thinking never translated in a single leak). But this is more of a convenience purchase, not a "need" right now; so I could easily wait one more year.

So, some intriguing questions:

  • Will ARM processors run everything? i.e will it be a seamless tansition for us?
  • Would this sole change make you careful about buying the first gen ARM based macs?
  • What other implications would all this mean? (pro/cons)
Hopefully some of you care or simply find entertaining to help/explain all this!

Regards,

ARM (as current version ARMv8) will run anything that a x86 CPU will run.
Microsoft already proofed this by allowing ARM CPU to half emulate x86 binary code (serving x86 binary native arm API libraries and emulate the x86 part).

So there no "x86 only" software.

Now everyone hate emulation so we want native binary on ARM.

The next part is kinda complex as you have to know what a compiler is. Chrome/Safari/Word/Keynote almost every software you are currently using are not build using the language CPU could understand.

Your CPU only understand binary that is 010101s. Those binary contains instruction that CPU could directly run.
Developer usually write "High level programming language" like C/C++/ObjectiveC/Swift/Java/Kotlin.
Those language have to "compile" into binary somehow and this compiler take care of what CPU you are targeting.

You can just ask compiler to compile your code into ARM or any new CPU arch that is supported by the compiler--if your code does not contains any CPU specific code like inline assembly or platform different length variables.

Modern platform usually encourage developer to avoid these as those things caused trouble when upgrading from x86 to x64.
By now if your code runs on x64 then most likely you do not need to modify a single line and it will just works on a ARM64 CPU target.
 
The biggest implication will be the loss of support for legacy software.

Legacy software died with the release of Catalina and its killing off support of 32 bit apps. As such, the number of abandoned apps that are not being actively updated will be minimal. I have no doubt that Apple will release a Rosetta style emulator, too.
 
The concern i have is that they are going to lock it down hard. This 16" Macbook Pro may be my last Mac, particularly if they remove bootcamp functionality.

Agreed. It's one thing to lock down hand-held / portable devices, but if it's going to stay competitive the MacBook will need to maintain broad capabilities, including supporting x86-based software.
 
As others have noted, this will be a staged roll-out, with the most powerful Mac models (the ones being used in corporations) being the last to migrate off of native x86 CPUs. And most of the Macs I have seen in my workplaces (and they are very large workplaces) are used to write software for the cloud, not to run Windows productivity applications (that is done on PCs running Windows natively).


As others have also noted, the most-likely model that will be released first will be in the 12" MacBook Retina form factor. And those used 7W Intel CPUs that were not known for fire-breathing performance. So I think even with emulation, a top-end A-Series CPU could probably hold it's own at a minimum running native x86 apps.
 
  • Like
Reactions: alexandero
Read the OP quote:
"If you are willing to provide a desktop-style cooling solution (i.e. heatsink and fan) there's absolutely no reason ARM can't own x86-64."


Sure, switching to ARM will grow the market, but ARM won't *own* the market like the OP claims. I can't re-call claiming the ARM market would shrink...

For ARM to own the x86-86 market, they'll have to dominate Windows, and that will be a hugely difficult task to do. The chicken and egg situation, like I've referred to.

Windows is also going ARM.
And since Apple just built VM hypervisor for Mac recently it's hard to imagine they will not support running Windows ARM in VM.

Recent Windows Server ARM insider build already shows Hyper-V running on ARM running Windows ARM. (Yes ARM host and ARM client VM)
 
  • Like
Reactions: MacsAre1
Depends on whose SDKs the apps use. On iOS, generally developers have no choice but to use Apple's APIs for everything. On MacOS, developers have been free to use any other compatible API if they wanted, or even bring in additional libraries if needed. It really depends on the app.

I agree that most common apps will be fairly easy to port, especially if they were written for maximum OS X compatibility from the start. But some of those apps written for Linux and X11 that were brought over to OS X later are probably going to be a giant pain to port.

Open source software targeting Linux might already support running on ARM and already runs on ARM with cocoa API and Darwin kernel--iOS.
 
  • Like
Reactions: MacsAre1
Geekbench also inserts a 1 second pause in between each test so thermal limitations of mobile devices don't negatively affect their results.

Very unlikely that Apple will use a mobile device chip in a new ARM Mac. New chip probably be capable of the same thermal dissipation as the current Intel silicon, but typically be run cooler for the same real-world app performance.
 
As an option, sure. As a replacement, it'd put a quick end to Mac use in corporate environments.

and it'd kiss gaming goodbye. Sure you might have iPad games, but the big titles would never port, and losing bootcamp would be the final fork in Mac gaming.
Most games are 32 bit, and as such will already run in Windows ARM (and already don't work on Catalina) today via Microsoft's emulation layer.
 
They will not replace the main CPU of the systems any time soon. Might add another chip to allow it to run certain things (similar to how things get offloaded to the GPU), but an interpretation layer will NOT give the performance everybody is dreaming out.
 
I really hope this doesn’t accelerate the already declining availability of Macintosh ported software. Also hope it doesn’t result in even more lock down and strip down for MacOS.

It will increase the availability of software since any work involved is also useful for ipad and iphone.

And MacOS isn’t locked down (turn off SIP and you can do whatever you want) and it is not “stripped down,” (it can do everything it could do last year and the year before, unless you count removing support for old technologies ”stripping down”)
 
  • Like
Reactions: 09872738
Windows is also going ARM.
And since Apple just built VM hypervisor for Mac recently it's hard to imagine they will not support running Windows ARM in VM.

Recent Windows Server ARM insider build already shows Hyper-V running on ARM running Windows ARM. (Yes ARM host and ARM client VM)
It is, but it’s early days.

Windows has been ported to multiple architectures in the past, with less than successful outcomes.

there’s no guarantees this time round.
 
They will not replace the main CPU of the systems any time soon. Might add another chip to allow it to run certain things (similar to how things get offloaded to the GPU), but an interpretation layer will NOT give the performance everybody is dreaming out.

1) there already is an ARM chip in macs. This is not about that.
2) The increased performance we dream of is for ARM-native, not interpreted, code
 
  • Like
Reactions: 09872738
You got it wrong.
Sun exist because they have SPARC that is compatible to Oracle's database and other middleware back in 90s and early 00s.
x86 was the "new and proprietary" architecture at that point. Almost no one ever though x86 will ever take over the server market back then.

Sun ported Solaris to x86 in 1993. Your timelime is waaay off.
 
  • Like
Reactions: AlumaMac
I'm interested to see how this plays out. Intel's been holding up Mac releases for the last few years and they seem to be falling behind in the CPU wars. An ARM based Mac would be interesting and I can only imagine how much processing power it'd have, if the A-series chips give us anything to go by. I am someone who needs Bootcamp though, so if I were to lose that functionality with an ARM based Mac (which is a real possibility), it'd be hard to move over to macOS exclusively until developers ported their software over.

I'm optimistic though. The switch from PowerPC to Intel took a few years, but look how much good that brought to the Mac. Maybe the same thing will be said about the switch from Intel to ARM, if it truly does happen.
 
Yes! The zillions of macrumors fanboys who screamed for a G5 PowerBook are finally going to get their wish. (except with a RISC ISA and implementation that’s more than 10 times faster, and no water cooling needed!). With probably even more apps available for it.

Or do all of you really want that water cooling?
I totally want that water cooling. 👌😀

BTW, Asus makes a water cooled laptop. The water cooler is a detachable unit that makes the thing ginormous.
 
  • Like
Reactions: firewood
It is, but it’s early days.

Windows has been ported to multiple architectures in the past, with less than successful outcomes.

there’s no guarantees this time round.

Windows NT has been ported to x86 and that's a huge success.
Obviously not guarantee but more competition is better than less right?
 
I wonder if the ARM chip will be a "co processor", capable of running native code for better power consumption?

Getting rid of options to run x86 would be too drastic a change IMHO. (I'd prefer an AMD option)

No way they do this. It would be way too expensive to include both an intel and Ax chip in this hardware.
 
I'm all for this. The only non-Apple apps I use anymore are Photoshop and Lightroom and my understanding is that the iPad versions of these apps are pretty capable, so as long as it will run competent versions of these apps, I'll gladly buy an ARM MacBook.
IPad version of adobe? Really not even close to what a macOS version can do
 
Perhaps it'll just be an iPad Pro with non-detachable keyboard? :p

Actually, I wonder if this wouldn't be a better way to go. If they worked on iPadOS 14 to be more Mac-like in some ways, they could have a product in the $750-$1000 range that would compete with higher-end Chromebooks and mid-range Windows laptops, is designed for (and supports) a touch-screen, blows away the competition in performance and battery life, has cellular support, and have a ton apps that could run on it from day 1.

Why go through a messy Mac transition and fragment that market when you can have a product that plays to Apple's strengths right now?

Also - there are more iOS developers than Mac developers, and apps designed for this mythical product can possibly be ported to the Mac using Project Catalyst.

So, while everyone wants an ARM-based Mac, I want a more Mac-like (and cheaper) iOS laptop.
 
Are there ARM chips that are capable of matching the powerful processors available for the new Mac Pros?
 
Sun ported Solaris to x86 in 1993. Your timelime is waaay off.

And guess when did Sun Fire x86 server becomes popular.
Most bank still purchasing SPARC during early 00s as they want to run software that was built more than 20 yeas ago.

I was an engineer prepare them for bank at that time. Using hot swappable CD-Drive to load OS on them.
 
Last edited:
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.