As ever, the moment that I point out that a traditional procedure is either unavailable or outmoded on Apple silicon Macs, I’m besieged by cries of “but that’s how I’ve always done it”. So did I for many years, but that doesn’t mean that we shouldn’t change our practices when our Macs change. So how are Apple silicon Macs so different that we should change how we use them?

Reliability

The most likely serious problem we’ve all planned for in the past is hard disk failure. While the great majority ran reliably for at least three and often five years, there have always been rogues that ground to a halt earlier, sometimes audibly. While SSDs aren’t perfect, failure rates appear far lower, and comparable with other solid-state components, as we should expect. It might be comforting to think your emergency external disk could get your Mac out of trouble, but the chances of using it instead of its internal storage are fast approaching zero.

Thankfully, hardware disk failure was unusual, but damage or corruption to a hard disk’s contents, particularly to macOS, were more common. Routinely running a disk checking utility to detect and repair file system errors, and procedures such as ‘clean’ installs and Combo updates, were well-established defences. Switching to APFS has certainly helped, but best of all, the Signed System Volume introduced in Big Sur has made those procedures redundant. If your Mac has a T2 or Apple silicon chip, the integrity of its system is checked every time it starts up (Apple is more vague about how well this is supported by Intel Macs without T2 chips, though). That’s not just its file system metadata, as checked by Disk Utility, but all the data in every single file on the System volume, a feature not previously available in macOS.

Recovery mode is relatively recent, having been introduced in Mac OS X Lion 12 years ago, and has been completely redesigned for Apple silicon. For some essential tools like Startup Security Utility and engaging Safe mode, there is no alternative, and in the absence of any third-party disk repair utility, it provides as complete a repair facility as you’ll get.

There are other factors involved too. Apart from hard disk failures, the most common cause of hardware problems in the Macs I’ve owned and been responsible for have been graphics cards, usually one of the hottest components. Heat is bad news inside computers, and components that heat up and cool down repeatedly during normal use may be more likely to fail as a result. At one time this was often blamed on the change to lead-free solder; more recently in PCs it appears attributable to sheer temperature, that can melt other components. Not only do Apple silicon Macs incorporate their GPUs in the main chip, but they run far cooler than our older Macs did.

In both hardware and software an Apple silicon Mac is thus designed and built for reliability. We need to reappraise the measures we take to ensure that reliability, in the same way that we know not to try defragmenting SSDs.

SSD cost and performance

It’s often claimed that Apple solders in SSDs in order to profit from the mark-up it applies to their cost. There are other and more compelling reasons for this design choice, which I believe has been driven not by marketing but by engineering.

SSDs used for internal storage in Apple silicon Macs are different from those we might order from Amazon. For a start, they’re blisteringly fast, with transfer rates as high as 7 GB/s, more than double the maximum of just over 3 GB/s achievable over Thunderbolt 3 or 4, and that’s with full encryption enabled. This is possible because the SSD’s controller is built into the chip, and incorporates in-line encryption.

There are two Apple silicon models whose SSDs aren’t soldered in: Mac Studio and Mac Pro. Replacing the internal SSD in either of those requires the whole Mac to be restored in DFU mode, a process that needs another Mac, and isn’t just a matter of swapping the SSD. That’s required because the internal SSD stores the great majority of the pre-boot ‘firmware’ used by Apple silicon Macs, to support their Secure Boot.

Contingency

For many of us, our Macs aren’t optional accessories, they’re how we make our living, run our business, or keep our essential records and correspondence. No matter how reliable their design and engineering, we have to consider the possibility that they might fail. Planning for that contingency is much more than installing macOS and a few useful utilities on an external disk, which is in any case more likely to fail than any major component in our Mac.

Many of us are equally or more dependent on our cars, yet we don’t feel the need to keep a spare engine on hand, or a second vehicle simply to cover for those circumstances when the first car has a problem. If you rely on your car to get you to work, then you’ll surely have a fallback plan, and in the coming winter that has to take into account severe weather like snow.

Just as we have to plan backing up our data, so we have to rehearse in our mind what we’d do in the event of sudden failure of a Mac. For some who work to tight deadlines (like me), that means always having a secondary system in addition to the Mac we use for ‘production’. In other cases, it might mean arranging hire of a replacement, and most businesses can insure against those costs. These are all part of planning for business continuity in the face of whatever fortune throws at us. Simply repeating the success formulas of the past isn’t necessarily going to address the problems of the future.