The rumors are true—Apple is beginning its next major Mac transformation, kicking Intel processors to the curb in favor of its own silicon. At the WWDC virtual conference on June 22, Apple announced the start of its two-year transition to produce all new Macs with its own Apple silicon rather than CPUs provided by Intel and GPUs provided by AMD.
What does that mean for you? Here’s a look at the timeline and technologies Apple announced.
Why Apple is doing this
Apple began its announcement by trumpeting its prowess in making its own silicon, including ten years of iPhone chips (with a total 100-fold increase in CPU power) and specialized versions for the iPad Pro (with a 1,000-fold increase in graphics power). Apple even went smaller, customizing its processors for the unique demands of the Apple Watch.
All told, Apple has shipped over 2 billion of its own processors over the years. With the Mac, the intention is to bring its industry-leading performance per watt to the Mac. The goal is to produce faster, more powerful Macs that use less power.
But it’s not all about performance-per-watt. Apple has built lots of specialized functional units in its processors, from powerful image and video processing engines to the Neural Engine to run machine-learning and AI code. Apple intends for the Mac to make full use of this specialized hardware, developing products, software, and silicon together in harmony as it does with the iPhone, iPad, and Apple Watch.
This will enable Apple to build Macs with capabilities and form factors it simply can’t today.
Apple did not announce its first Mac processors, though. The development kits will use the A12Z processor found in the iPad Pro, but the first consumer products will likely use something else.
Native apps will lead the way
The first Macs to ship with Apple’s own silicon will run the just-announced macOS Big Sur, which we noticed carries the OS version 11.0 rather than 10.16 on Apple’s development system.
When you buy a Mac that has one of Apple’s own processors, you can be sure that it will run native versions of all of Apple’s own apps, specifically optimized for its processors. That includes pro apps like Final Cut Pro and Logic.
But third-party developers will be able to make new apps that are optimized for the architecture as well. “Universal2” apps give you a single binary that can run on both Intel or Apple’s own ARM-based processors. According to Apple, developers can download the new version of Xcode, recompile, and have most apps up and running in a cross-platform manner in just a few days.
Apple specifically called out two of the big developers users are likely most worried about: Microsoft and Adobe. The company demonstrated native versions of Office on Apple silicon, including the same versions of Word, Excel, and Powerpoint that we use on Macs today. Everything was fast, responsive, and fluid. From Adobe, Apple showed Lightroom and Photoshop, each processing rather heavy tasks on very big sets of data. Notably, Premiere was not demonstrated.
Apple appears to be doing a great job of making a case for professional apps on its own silicon. Using a development kit with an A12Z, the company showed editing in Final Cut Pro with three 4K ProRes videos playing at once, and applying effects like titles and lens flare in real time during playback.
And how’s this for native apps? Macs will Apple silicon will be able to run iPad and iPhone apps natively, without a single change from the developer! You’ll just find them right in the App Store, and be able to download them at will. Apple demonstrated this with three apps running at once in separate windows: Monument Valley, Fender Play, and Calm. None of those apps are available for Mac, though they all do have iPad versions.
Rosetta2 for non-native apps
For apps that aren’t re-made for Apple’s silicon, there’s an updated version of the Rosetta software Apple introduced during its PowerPC to Intel transition. Rosetta2 will take apps made for Intel processors and translate them to run on Apple’s own ARM-based processors.
The new Rosetta translates most apps when they are installed, but can also run dynamic real-time translation for things like JavaScript that can’t be translated ahead of time. Apple promises better performance and compatibility than the original Rosetta (which was rather impressive), and even demonstrated two difficult cases: 3D modeling and animation software Maya, and the game Shadow of the Tomb Raider (running at 1080p on an A12Z processor with good performance and detail).
Virtualization will be supported as well; Apple showed a Parallels VM session running Linux to create an Apache web server.
The transition timeline
So when does this all happen? How does it impact you as a user? Apple detailed the timeline, though was vague about many specifics.
Developers will be able to sign up to buy a Developer Transition Kit. It includes a Mac mini enclosure outfitted with an A12Z processor, 16GB of RAM, a 512GB SSD, and all the software necessary to start building Mac apps for Apple’s own processors. Apple will begin shipping those out during the week of WWDC (June 22-26).
Perhaps the biggest surprise was Tim Cook’s promise that consumer will be able to buy the first Mac to use Apple’s own silicon before the end of this year. It will take some time before all Macs include Apple’s own processor—Cook said it is expected to take two years.
Intel isn’t just going to be dropped, however. Cook promised that Apple still has great Intel-based Macs on the way, and that macOS will continue to be built for Intel processors and supported for years to come.
This is similar to Apple’s prior transition from PowerPC to Intel. We should note that Apple completed its product transition ahead of schedule that time, and continued to make Mac OS X with support for PowerPC chips for an addition three years after the transition was complete. Whether the timeline will work out that way again is anyone’s guess.