Updated July 27. Article originally posted July 25.
The Mac’s move to ARM is an exciting one for Apple, but it is one with some major problems along the way. One of the biggest problems is the existing user base of Intel machines. Apple needs to retain support for the Intel platform. And that’s going to lead to an awkward few years with some substantial issues around every corner on the path to ARM.
July 27 update: The move to ARM offers Apple many advantages, but one of the key areas it believes it can leverage is the performance of ARM over Intel. The latest benchmarking figures bear that out. Chance Miller reports:
“New benchmarks have leaked today that show the Developer Transition Kit running Geekbench 5 Pro natively on the Mac mini…
“The results show a single-core score of 1098 and a multi-core score of 4555. This compares to the non-native of 800 on the single-core test and 2600 on multi-core. For comparison, the entry-level $999 2020 MacBook Air achieves a Geekbench score of 1005 on single-core and 2000 on multi-core.”
This is an improvement on the initial Geekbench results, where the Mac mini based Developer Transition Kit was using a virtual machine to run the benchmarking app. These numbers, from a native app, show a significant step up from Intel. Many factors will impact the performance of a MacBook on ARM, but the headline numbers are looking good for Apple as it moves away from Intel.
While the Intel machines are not going away overnight, Apple has quite clearly stated that its future is ARM based. The engineering efforts will be focused on the new architecture, the designers will be working with the new freedom offered by the increased chip efficiency, and marketing is going to be all about the new frontier. Apple’s primary focus will not be on Intel. Maintenance? Yes. Pushing the envelope? Unlikely.
Developers will also be looking towards the new platform. Although they will be just as conscious of their existing customer base and the need to support them, as Apple extends the MacOS platform and presumably brings in features from iOS and iPadOS the hooks may be restricted to the ARM platform. In other words the older apps can be maintained, but the full capabilities will only be unlocked on ARM hardware.
Finally there’s the loss of value on the Intel machines. Apple hardware has traditionally held its value for a much longer period than the equivalent PC hardware. Selling an older Mac to afford a new Mac is a time honoured tradition in the community.
It’s likely that as the years progress, the going rate for Intel hardware is going to drop far faster than in previous years, subtly pushing up the price of a move to an ARM-based machine (something Apple has already started to do).
One big mitigating factor here is Apple’s past performance and long-term support of older devices.
MacOS 11 is compatible not just on the current Mac machines, but can be installed and run on Mac machines going back to early 2013 in the case of that year’s MacBook Air. Apple is not going to suddenly give up on a seven year support window and the existing user base.
What may happen is that the 2021 iteration of MacOS may bring the drawbridge up a little quicker, perhaps only stretching back to 2015 rather than 2014. Any withdrawal, in my opinion, is going to be on a curve rather than a cliff-face.
Apple has managed this transition before, as it switched from the PowerPC platform over to Intel.
Announced in 2005 to start in June 2006, it was completed by late 2007 (a two year project, hmm that sounds familiar…). Although the full Mac range had switched over, PowerPC machines continued to be supported until MacOS 10.6 Snow Leopard.
That was a four year window of official support. Apple will have had a huge staff turnover since then, the corporate memory and knowledge remains.
The transition period is fraught with danger and there are massive problems along the way. Apple has worked hard to minimze the risk, but it is there. No matter how smooth the process, users are going to find it incredibly difficult to find the right balance.
On one side is the ‘full speed ahead, get on the ARM train as soon as possible’ crowd. On the other side is, I suspect, a much larger crowd that will be remaining with Intel for some years to come. They could be companies with a fixed IT budget and a fleet of Mac machines; business users with a specific workflow that requires third-party apps that may or may not be ported to ARM in the future; and individuals will have invested heavily in their personal Mac hardware.
Apple’s biggest problem may not be the technicalities of the new architecture, but keeping everyone in the Mac community satisfied with their Apple product as the bumpy transition goes on in the background.