Custom Mac Silicon Frees Apple to Iterate, or Not, as it Pleases

on November 13, 2020

Apple’s event this week brought a few surprises (three new Apple-Silicon based Macs, not just one) as well as some frustrating nonsurprises (no touchscreens, no LTE or 5G, and no new entry-level starting prices for notebooks). Based on Apple’s deliberately vague testing proclamations, the M1 system on a chip (SoC) certainly appears to be a powerful performer that will also offer substantial battery life improvements. We will know more about both in the coming days as reviewers begin the process of benchmarking and real-world testing. What is clear, however, is that the M1 has already caused Apple to radically rethink the role the processor plays in differentiating products in its lineup. Just as important, I believe it will give Apple significantly more freedom to iterate around its Mac form factors, features, and, eventually, prices.

Three New Macs
I won’t go into too much detail about the new Macs, as Carolina covered those details in her excellent day-two column. Like her, I wasn’t surprised that Apple chose to effectively hold the line on its pricing (with the exception of the $100 drop on the Mac Mini) because it needed to establish out of the gate that the M1 isn’t a low-cost alternative to Intel, but a powerful custom-designed replacement that merits like-for-like pricing.

What’s truly remarkable about this product launch is that by using the same chip across a new Mac Mini, MacBook Air, and MacBook Pro, Apple effectively eliminated one of the key ways the PC industry (and Apple itself) has traditionally segmented its products. Processor performance level and branding have always been a primary differentiator in the market. With the M1, Apple says the quiet part out loud by acknowledging that a single chip, placed into three different thermal envelopes, will drive three different performance levels.

That last part is going to fry a lot of people’s noodles, especially those who have traditionally made buying decisions based on the often small but highly marketed speeds and core count of one system’s processor over another. You can see it now as buyers wrestle with a decision between buying a fanless MacBook Air or a MacBook Pro that offers improved performance predicated entirely on the fact that it has a fan that lets the M1 processor run faster, longer than the one in the Air. Ultimately, I expect that the M1 and Apple’s subsequent Mac processors will lead an increasing percentage of their customers to think less about the processor and what its esoteric speeds and feeds mean to them. But this transition will take time, and it will cause some hand-wringing along the way.

One of the early issues with the shift to the M1 is RAM limitations. All the new Macs offer a standard starting RAM allotment of 8GB and a maximum of 16GB. For years it has been notoriously hard—if not impossible—to add aftermarket RAM to a Mac, but with the M1, it is simply not possible because the memory is part of the SOC. This 16GB limit likely isn’t a dealbreaker for most MackBook Air buyers. Still, it has caused a small but vocal minority of Mac diehards to pump the brakes on new MacBook Pro and Mac Mini purchase because they are unwilling to buy a system with less than 32GB of RAM. Here’s the thing: Conventional wisdom (and experience) may dictate that power users need 32GBs, but that may not be the case with the M1’s Unified Memory Architecture. We will have to wait for the benchmarks and real-world testing to know for sure.

Another notable thing about the new M1-based Macs is that they all support just two Thunderbolt/USB 4 ports, whereas some previous versions of both the Mac Mini and MacBook Pro offered up to four. It is unclear if this is an M1 limitation or an Apple design decision. However, this too may be a dealbreaker for some users, who—in the case of the Mac Mini or MacBook Pro—will then need to look back at the legacy Intel-based products still on offer or wait for subsequent product launches.

No Touch, No LTE, and No New Form Factors…Yet
I was not surprised that Apple opted to go with its existing chassis for these product announcements. Particularly in the Tim Cook era, Apple tends to be quite deliberate when it comes to new product designs, so it made sense that the new products look just like the old products. It was also not shocking that Apple did not add a touchscreen to the Mac and did not roll out an LTE or 5G option for the MacBook Air or MacBook Pro. And to many people’s disappointment, the company did not introduce a new lower-priced notebook. However, the fact that Apple did not do any of this week doesn’t mean that it won’t in the future.
In fact, I see that as one of the great benefits of the move to Apple Silicon. While the company decided the shift to the M1 was enough change for 2020, the flexibility inherent in rolling its own silicon—and knowing the ramifications of a future chip in terms of battery life, performance, heat, I/O, and cost—uniquely positions the company to iterate on the Mac in ways it has never done before.

Obviously, there will be new designs, likely in the service of Apple’s obsessive drive to make everything thinner and lighter. With its own silicon on board, Apple will be free to make design changes without waiting on a partner or making concessions for features it deems unnecessary for the Mac. I’m less convinced Apple will add a touchscreen to the Mac, even though many of us have pushed for it for years. However, the support for iOS apps, enabled by the M1, could mean Apple rethinks this position in the future. There is a slightly better chance that Apple eventually rolls out a Mac with cellular connectivity. This would require a fundamental redesign of its notebook chassis, and if it were to happen, it would likely occur using a 5G radio. There is a great deal of interest in connected PCs today due to the massive shift to work from home, but it’s clear Apple won’t be moving quickly to try to catch that wave.

Finally, I expect Apple to eventually roll out more affordably priced Mac notebooks (note I didn’t say low-end). It is instructive to look at how Cook has approached this in his other categories. Traditionally, it was with waterfalled products—last year’s iPhone drops in price, the previous year’s product also decreases in price, and Apple keeps selling them to reach a wider audience. More recently, the company has launched purpose-built products designed to appeal to more value-oriented buyers, such as the second-generation iPhone SE and the Apple Watch SE. I suspect Apple will begin the process here by waterfalling M1-based products into lower price points as it announces new products with next-generation M-Series processors.
By shifting its lineup away from Intel, Apple will no longer have to deal with people always pointing out that it sells products with years-old chips that look dated versus the other PC players. Yes, Dell, HP, Lenovo, and others will always ship products with the latest Intel processor. But, Apple will argue, this two-year-old M Series processor is still competitive because it is custom-designed to run this product.

I’m eager to see the first benchmarks and to test out one of the new Macs myself. If the new M1 performs as well as Apple suggests, then this silicon transition is likely to have a much more significant impact on Apple (and its competitors in the market) than any previous transitions. This has been a resurgent year for the PC category, and things just got a whole lot more interesting.