Pixel Envy

Written by Nick Heer.

M Powered

Next Friday, the first Macs powered by processors of Apple’s own design will go on sale. That is a pretty big deal. But what isn’t — at least, not from a user’s perspective — is the transition from one architecture to another. Apple last did this about fifteen years ago when it moved from PowerPC to Intel processors, and it is effectively copying its own playbook from that time. There are even sequels of the same transitional technologies: Rosetta 2 and Universal 2.

It should therefore come as no surprise that the Macs announced today use enclosures that are either identical or nearly so to the Intel model equivalents, and at similar prices. The perception of a smooth transition would be marred if there were a clear line between Intel Macs and Apple Silicon Macs, so it is best to keep everything pretty samey.

One way that this transition differs from the last go-around is the order in which Apple is turning over its lineup. The first Intel models Apple released in 2006 were the iMac and the MacBook Pro — its consumer desktop and its professional notebook. But the first Apple Silicon models are entirely on the lower end of the performance spectrum. The Intel MacBook Air has been entirely replaced by ARM models, as have the two-port 13-inch MacBook Pro models. The ARM Mac Mini is kind of a new line at the lower-end of Intel models. In the case of the 13-inch MacBook Pro, the four-port models updated earlier this year remain available only with Intel processors, as does a higher-end Mac Mini configuration.

As expected, Apple Silicon now has a proper marketing name: all of the Macs announced today run on the M1 system-on-a-chip.1 Aside from one fewer GPU core in the lowest-end MacBook Air configurations, any performance differences between these computers can likely be ascribed to their different thermal envelopes. The MacBook Air doesn’t have a fan, and the MacBook Pro remains pretty thin. Otherwise, the SoCs appear to be pretty much identical; or, if they are different, Apple is not describing them as such: they have the same transistor count, same core counts, and the same RAM options, as memory is now packaged with the SoC instead of being soldered to the board separately. And because so much of this transition is about efficiency, these MacBook models promise remarkable battery life.

Despite the M1 being an apparently entry-level configuration, Apple is promoting big performance gains. Graphics on the MacBook Air, it says, are up to five times faster than the highest-specced Intel model; on the MacBook Pro page, it says that machine learning performance is eleven times faster. Those are big leaps for complex tasks, but we’ve been down this road before. The Intel iMac was said to be two to three times faster than the PowerPC model it replaced, while the first MacBook Pro was apparently four to five times faster. Those tests were conducted using benchmarking tools, while the comparisons this year are being made using real-world tasks. All of this is to say that we can’t know just yet how fast these new Macs are. Even though they are not Apple’s most performative products, could they perhaps out-perform their Intel-based cousins? Or are they modest updates that help guide users and first- and third-party developers onto a new platform?

At this time, only Apple’s engineers know for sure. But I think these products are a great warm-up act for M-branded processors to sweep across the entire line. The M1 appears to be a bigger, badder version of the A14, so it makes sense that it would be placed in these more entry-level products. From a performance perspective, this is only a clue about what Apple’s full capabilities are. Remember, it has committed to switching every Mac to chips of its own design, including its highest-end products. If today’s event felt underwhelming from a “professional” Mac perspective, it will make up for in volume — these are among Apple’s most popular Macs. Today’s announcements are just as much about getting the future of the Mac in front of a huge number of typical users.

It is also only the start of seeing the hardware and software integration that custom-designed SoCs may allow. I was not expecting Face ID today, but I am certain that it is coming in the not-too-distant future. You can imagine for yourself what products might be enabled by own-brand processors focused on, in Craig Federighi’s words, “quiet performance”: a truly miniaturized Mac Mini, perhaps, or a less chunky iMac. A new Cube — why not? But, again, I am not surprised that nothing like that was announced today. This is an internal revolution disguised as a transition, and there is no reason to upset that illusion just yet.

So much of today’s short keynote felt like a love letter to the Mac. I won’t spoil it for you if you haven’t seen it, but there were familiar faces throughout — right to the end. The first half of the past decade, for whatever reason, felt like Apple wasn’t too sure what it wanted to do with the Mac. But recent years have indicated that the company is still strongly behind this unique family of products and truly does care about it just as much as the rest of us, even if it does not always feel like it. This is the most excited I have felt about the Mac in a long time. I am not in the market for a new computer but, when I am, I do not feel worried that I have reason to switch platforms. The Mac still feels like home.

Here’s to — for so many reasons — 2021.

  1. This branding is recycled from the motion coprocessors. Apple still refers to the “M12 coprocessor” on its iPad Pro and iPad Mini tech specs pages. ↩︎