Macworld:
With every release of a new iPhone powered by another cutting-edge processor designed by Apple, the rumbling grows. It’s amplified by the perception that the Mac is being delayed and hamstrung by the moves of the Mac’s chip supplier, Intel. It’s the theory that, one of these days, Apple is going to break from Intel and power its Macs with an Apple-designed processor related to the ones in the iPhone and iPad.
And it’s true, the Mac is no stranger to a processor transition. It’s happened three times in the 32-year life of the Mac, so roughly once a decade.
It could definitely happen. I don’t want to say that it won’t, because Apple’s desire to chart its own course and not be beholden to other companies for key parts of its products is well known. Having proven itself a capable chip designer with the A series, Apple could very well dump Intel and strike out on its own.
But I don’t think Apple will.
I would strongly disagree with my colleague, Mr Snell. I think Apple will eventually do this. But not for at least 5 years.
Rumor has it that Microsoft is developing their own Rosetta-like tech for running Intel apps on ARM: http://arstechnica.com/information-technology/2016/11/x86-emulation-rumored-to-be-coming-to-windows-for-arm-in-late-2017/
That would possibly alleviate the boot-camp concern.
I’ve often read stories stating the latest iPhone/iPad is as fast as fairly recent desktops and laptops so if this is true, then it will probably happen sometime.
This directly contradicts the clickbait tidbits article you link to later. Apple is NOT going to do a major architectural shift if they’re marginalizing the Mac platform. Come on.
Given the A series typical speed increases of 1.5-2X for each generation, in 2-3 years it’s going to blow past intel and make it almost a moot point.
The new MacBooks will have an ARM chip in them for the Touch Bar – and from a developer and user point of view it is seamlessly integrated. Also Apple does things in stages – for example, Secure Enclave to Touch ID to Apple Pay.
I fully expect that we will see more ARM chips going into Macs, and year on year they will be taking on more and more of computer’s processing. And one day, maybe there won’t be an x86 chip at all, but rather a group of ARM processors working seamlessly together.
Could this be the reason why the Macs haven’t seem big updates yet? My guess is that at next WWDC Apple will announce the move to ARM which will take them about 12 months for the entire product family. They need to give devs a couple of months to update the apps.
“They need to give devs a couple of months to update the apps.”
I assume this is sarcasm.
That’s what I’ve been thinking (though admittedly, coming from a place of ignorance when it comes to processor architecture). Keep the Intel chips as the primary processor, but start adding ARM chips to take various tasks off of that to speed things up overall. Seems like the best of both worlds, as it allows Apple to boost their computers without an over-reliance on Intel’s product roadmap.
I hadn’t associated the delay in desktops with an architecture change. Does Swift have a significant role in adapting between processors? Will boot camp be a plug in board like the Motorola days, or is running Office or whatever going to be online or iOS? Of course if it can run on iOS, there’s no reason it couldn’t run on A class macs, along with all the other major Windows apps that have iOS ports.
Speaking of desktops, what the hell were they thinking with the Tube? It’s like if the Cube was the only desktop Mac coz it was purty. (Back in the day, or was it the night?)
Or, put differently, if they did it, it would demonstrate rather odd thinking.
Sure. Apple could migrate Macs from Intel to ARM.
But would it be worth it? And not just on the bottom line.
There would be kvetching among the few, the loud, the “pros.”
Developers of major apps (ahem Adobe) would drag their feet.
Fat binaries would take longer to download and install. Etc.
Apple would need to show pros, consumers, and developers that migrating to ARM would provide real benefits. They’d have to pull everyone in with quality and performance, not arbitrarily push everyone over with deprecation.
I don’t think there would be any benefit, other than saving a few bucks on low-end MacBook/iMac CPUs, until the Ax CPUs are vastly faster and cheaper and more power-efficient than the Intel parts.
So yeah. Maybe 5 years.
Maybe Xcode 9 will have a new option: “Fat binary”. For ARM/Intel compatibility.
And Swift 4 could be even more processor-independent than Swift 3.
Doubtful, I know. It all depends on Apple’s long-term Mac strategy. And if/when it will ever make sense from a business standpoint. Lots of factors in that equation.
For some developers, it could be as easy as a recompile. If your app is new, doesn’t do anything low-level, and stays within all of Apple’s guidelines, there’s a good chance that is all it will take.
But for others, it will be man years of effort. It took Adobe and Microsoft a long time to fully port their stuff to Intel.