Apple’s Mac computers have gone through just two major CPU architecture transitions in their entire history: moving from Motorola 68k to PowerPC processors in 1994, and then switching from PowerPC to Intel x86 chips in 2005. Based on recent reports (and nearly half a decade of rumors), it appears Apple is on the precipice of a third major transition, with strong expectations that Apple will announce the first ARM-based Mac next week at WWDC 2020.
For Apple, the benefits of moving away from Intel chips to custom ARM-based processors similar to the existing A-series chips Apple uses in iPhones and iPad are clear. Not only would Apple gain greater control over the roadmap and features built into the CPUs that power its computers, but Apple would also stand to reap greater profits by cutting yet another third-party supplier out of the equation. On top of that, with Apple’s A-series chips often delivering similar single-core performance and better performance-per-watt numbers compared to Intel x86 chips, an ARM-based MacBook could be just as fast while offering increased battery life.
But for me, the most interesting effect of Apple’s switch from x86 to ARM is how this change could affect the design of future Mac computers. What operating system would an ARM-based Mac run? Will Apple simply update macOS to better support ARM, or is Apple planning another big software transition similar to what we got when the company moved from Mac OS 9 to OSX? In Steve Job’s keynote speech at WWDC 2005, he said all the work that went into OSX set Apple up for the next 20 years. But now in 2020, we’re drawing close to the end of that window. So what comes next?
Furthermore, what potential impacts could the switch to ARM have on product design? So in advance of WWDC 2020 which kicks off next week on Monday, June 22, I wanted to ruminate on some of the possibilities presented by Apple’s big transition.
Over the years, Apple has adapted macOS (and its predecessors) to run on a number of different architectures, and while it won’t be easy, there’s no reason why Apple shouldn’t be able to get macOS running on ARM. And to start, that will probably be the most straightforward solution, putting a new A-series chip in something like the MacBook Air. This would create a super thin and light device capable of handling everyday computing needs, but with better longevity.
The downside to this is that while Apple can tweak or update many of its first-party apps to run on ARM, it could take extra time additional time to adapt more complicated apps like Final Cut Pro, not to mention the countless third-party apps where outside developers might not be so quick to adjust. And while initiatives like Catalyst can help port ARM-based iOS apps over to Macs, that won’t be a magic bullet for everything.
But the bigger problem comes when you consider the Mac Pro desktop, which just got released late last year. Generally, x86 chips with their Complex Instruction Set Architecture (CISC) are better equipped to handle more intensive tasks like video editing or 3D modeling than ARM-based chips with their Reduced Instruction Set Computer architecture. That problem could force Apple to maintain multiple versions of macOS, with one designed to support high-end x86-based systems like the Mac Pro, while also developing an ARM-based version of macOS for less powerful machines.
However, with recent additions to iPadOS like native mouse support and accessories like the Magic Keyboard and its built-in touchpad, iPad Pro has become a totally viable alternative to a typical MacBook for a lot of people. So in the long run, it’s possible that iOS/iPadOS will become the foundation of Mac computers, with macOS being reserved mostly for high-end systems before being phased out completely in four or five years (or more).
Compared to macOS, it feels like Apple spends much more effort refining iOS and iPadOS. With combined sales of the iPhone and iPad responsible for nearly 60% of Apple’s revenue in Q4 2019 versus just 11% for Mac, it just makes more sense to invest resources into what is already Apple’s most popular OS, rather than trying to keep macOS going forever or split its attention between multiple forks. Furthermore, iOS and iPadOS already work well with a number of input methods including mouse and keyboard, touch, and stylus, which isn’t really true of macOS and brings me to my next point.
It’s kind of mind-boggling to think that in 2020, Apple still hasn’t made a single Mac with a full-size touchscreen. Over in Windows land, touchscreen laptops have been widely available for the better part of a decade, and now, some computer makers are working on concept devices that have completely ditched the traditional keyboard and touchpad in exchange for two touchscreens or big flexible displays.
Aside from multiple Apple executives having previously stated that they simply aren’t interested in making a MacBook with a touchscreen, another big reason that makes creating a touchscreen Mac so difficult is the OS itself. macOS’s current UI just isn’t designed for touch. Things like icons and menu settings are too small to easily adjust with fingers, and while Apple could certainly redesign macOS to work better with touch, it would probably be a long and painful process, something Microsoft found out the hard way with Windows 8 and even into Windows 10. Instead, it would might be easier and more successful in the long run to scale up iOS/iPadOS to better accommodate desktop use.
Now at this point, a lot of people could question the need for a touchscreen Mac when the iPad Pro already exists, which is fair. But when you combine a 12.9-inch iPad Pro with a Magic Keyboard, you end up something that weighs three pounds, which is actually heavier and less sturdy than a 2.8-pound MacBook Air. Also, with just a single USB-C port, an iPad Pro isn’t always a great choice as a primary work machine.
So for a lot of Mac users, it would at least be nice to have the option of getting a super-thin laptop with a touchscreen (or hey, maybe even a convertible 2-in-1) instead of being forced to go with a heavier and bulkier iPad Pro.
In the short term, Apple’s big transition to ARM is sure to suffer from some troubles with app compatibility, similar to what the Surface Pro X and other ARM-based Windows devices encounter when you try to run older legacy Windows apps or programs not designed to run natively on ARM.
But if Apple can push through that (which based on previous transitions, seems quite likely), Apple will end up with something it has been working towards for quite some time: a truly unified app store. It wouldn’t matter what device you’re using, and there would be no distinction between the Mac App Store and the iOS Play Store, you could just buy an app once and it would just work on whatever device you’re using.
Recently, Apple has taken a big step towards this with its Universal Purchase system, which allows you to buy an app once but still use it on both Mac and iOS. However, because Universal Purchase is still relatively new and has yet to factor in the need for a new wave of ARM-based Mac apps, Apple hasn’t quite reached its goal of creating one app store to rule them all, at least not yet.
This is one I think every Mac user can get onboard with: cheaper systems. And while it’s not a guarantee, by bringing chip design for all of its devices in-house (not just iPhones and iPads), Apple may be able to cut production costs and lower the price of ARM-based Macs. We’ve sort of seen this happen recently on the iPhone side with the new iPhone SE, which reuses the iPhone 8's body and the A13 chip from iPhone 11 to help bring costs down. And with a full retail price of $400, that means when adjusted for inflation the new iPhone SE is actually the least expensive iPhone ever.
Meanwhile for Macs, a new MacBook rarely costs less than $1,000 even if you got for the most stripped down, barebones model, which ain’t cheap. And with CPUs being one of the most expensive components of a laptop and chips like the 10th-gen Intel core i5 CPU featured in a new MacBook Pro 13 accounting for as much $250 of its price tag, there’s a good chance that by switching to custom ARM processors, Apple could pass along some of those savings to its customers.
But again, I have to caution that it’s still very early and there are a ton of unknown factors at play. Heck, Apple hasn’t even officially announced its first ARM-based Mac yet—though all signs indicate that that will happen early next week. This is simply a piece to muse about all the possibilities, because in this precious time before the big reveal, one of the most entertaining things about tech is just dreaming about what could be.