There’s increasing chatter about a new A-chip MacBook to cover the low end and better compete with Chromebooks. There’s also more and more indication the Mac Pro is EOL with the Ultra version of the Studio essentially able to take its place in all but a very small subset of fringe cases that Apple likely believes it does not need to cater to. And just recently, there’s been a leak of Mac hardware expected to be released in the not too distant future that is hinting at finally a larger iMac equipped with a Max-level M chip.
So if we assume that those rumors are not entirely wrong, I’m curious what holes people believe Apple is still leaving in its lineup. Apart from some large tower (that I’d be willing to believe Apple just doesn’t need to offer), is there any type of “regular computer” that Mac does not cater to?
And conversely, are there areas where there is just too much Mac going on? Is there a Mac that could be axed to refocus efforts on perhaps one of the so far neglected spots in the “regular computer” space? Eg. is a non-Pro M4 14" MBP really an important SKU to offer when there is already a 13" and a 15" M4 MBA, not to mention the only marginally more expensive 14" MBP equipped with an actual M4 Pro?
Obviously, Apple will do whatever they do and that’s fine. I’m more interested in hearing folks’ takes on what paths they see lying ahead and how they value these different options. Disclaimer: thread likely to contain ample speculation so if that isn’t your thing –> cmd-w is your friend.
Apple I think sells lots of the low-end MacBook Pro to businesses that want the better display and the cooling that comes with it, so I think that Mac will remain in the lineup for a long time to come.
I still think that there is a world where Apple will have both a processor that is more powerful than the Ultra for a Mac Pro, as well as perhaps GPU and neural engine only co-processors with a super-bus that will allow them to connect to the M-series chip at high speed, as there will always be be customers that want workstation-level capabilities in a Mac that just won’t be able to be handled by the Studio. So I think there will be an upgrade to the Mac Pro someday as well and it will remain in the lineup.
Apple’s Mac business seems to be doing well the last five years since the M1 was introduced, and I’m not sure that Apple is really missing much of anything in the lineup. Maybe a larger screen iMac, but I’m guessing that Apple isn’t selling many iMacs these days and instead wants to steer people to discrete Mac mini or Mac Studio with external displays.
I also won’t be surprised if Apple continues to improve iPadOS and more low-end Mac users will simply buy iPad instead.
In 2025, I think a modified matrix would make sense, e.g. something catering to Consumers, Prosumers, and Pros for laptops. Something similar would work for desktops, though I’m not exactly sure how I’d describe the desktop range.
I think there is a place for both iMacs and non-All-in-One Macs across the full range of desktop users. It’s actually not so different from what Apple actually offers, but I think Apple has lost the simplicity of how Jobs spoke about the range of available models.
Only the thing that Apple has chosen to not develop since the end of the old PowerMac series. An affodable desktop system with internal expansion capability.
With today’s architecture, this might be something like a Mac Studio with one or two PCIe slots and one or two M.2 slots. So not all expansion has to be via Thunderbolt ports.
But it’s never going to happen.
Regarding the Mac Pro, Apple blew that market by making it little more than a Mac Studio in a large case. It’s (more or less) what I was describing, but with too much expansion space and far too expensive.
The thing that made the Intel Mac Pro popular (in its day) was the ability to install massive amounts of RAM (up to 1.5 TB) and storage (four 3.5" drive bays and two 5.25" bays). This was great for those needing a high-powered workstation or server. But you can’t get anything like that with Apple Silicon for any price.
Regarding redundancy, I think that since the invention of Apple Silicon, there really isn’t much that distinguishes the MacBook Pro and MacBook Air series. There are only minor differences, and only minor variations in price. The big difference only appears if you max-out the configuration, with the Pro allowing higher-end configurations.
As for what I would really like them to bring back, I would really like them to bring back the 11" MacBook Air. I love my 2011 model, and when circumstances eventually force me to replace it, I’m not going to be able to find anything that small. It’s not a desktop replacement, and never will be, but it’s wonderful for basic computing when I’m away from my desk, where the desktop system lives.
Here’s the issue I see with expansion cards on Apple Silicon Macs.
Who has drivers written for expansion cards for macOS ARM?
It’s a bit of a chicken or egg situation as I see it. If a card needs driver support (e.g. PCI GPU cards and/or audio interfacrs), without it the card won’t work. And the driver support won’t come until a card can be used or is commercially viable for the effort required to make the driver Apple Silicon native.
Apple has also complicated things by locking down the kernel (deprecating kexts). You get a choice of reduced kernel security and driver signing, or a secure system that may not run the cards. And then there’s that detail of actually spending the time to write the driver.
Unless Apple comes up with a solution to this conundrum, I don’t think it’s likely that they’ll build hardware for the market segment formerly targeted by the expandable Mac Pro. And it shouldn’t be “if we build it they will come” – because with Apple, that hasn’t been the case for a long long time if ever.
This was my first thought also. It greatly simplified a very confused lineup, and would have eased design and manufacturing issues.
My primary machine, for as long as I can remember, has been a laptop. There’s been no thought over this period of buying a true desktop machine, and the vast majority of my friends and family are the same. I’m sure there are people for whom a Pro Tower machine might be handy, but given Apple’s glacial development of such a machine it tends to suggest it’s not a high turnover product.
If they produced a small, cheap, laptop I would buy one for when I, or my family members, travel. I’ve always felt a little uncomfortable dragging my main machine all over the world for fear of loss or critical damage. I’d also use it when I’m in the camper trailer travelling locally. I could use a MB Air, but a cheaper alternative would be nice.
I think the current Macs represent exceptional performance with reasonable value, but I also think they’re drifting into a confusing, overlapping lineup reminiscent of the old Performa/clone era. I don’t see that as a good strategy.
Since the Darwin kernel is open source, and ARM64 compilers are readily available (including the one in Xcode), this shouldn’t be any more difficult than developing a driver for macOS on Intel.
This is the big deal. Apple has created a lot of system services so you don’t often need a full-blown kernel driver, but if your device does require one, you may need to make a hard decision.
The solution, which I think Apple has implemented for some classes of devices, is microkernel services. You provide generic device drivers for low level buses like PCIe, along with APIs so board-specific device drivers don’t need full kernel access, but can run in a sandbox-like environment that restricts it to its own resources.
I think the biggest impediment to such a design is a business decision (convincing Apple that they should do the development work), not a technical limitation.
Good post, and I agree with what you’re saying. The example that comes to mind is the work that they’ve done under the hood that has allowed ExFAT and FAT file systems to run in user space as opposed to kernel modules. I think these are the same underpinnings that MacFUSE has available now as an option.
And you’re right on - there are business decisions that hold this back, not technical issues.
With Apple Silicon, the maximum they appear to offer right now for RAM is 512 GB. That seems useful , but it strikes me that they’ve decided to dispense with addressing the needs of monster users like CERN with over a petabyte a day of data. I imagine it’s a small population of users at that scale even worldwide.
The lineup seems pretty simple to me at the moment, echoing Jobs’ matrix:
Consumer: MB Air (laptop), Mini, iMac (desktops)
Pro: MBP (laptop), Studio, Mac Pro (desktops)
There are configuration choices to make for all the models, but surely we’re not arguing for a single setup for each in the interest of simplicity? Everybody gets 16 GB / 1 TB in black?
I don’t see a substantial gap (eg, where there are massive amounts of sales being missed) in there.
I was mulling a post along those lines in response to the excellent comments by @Shamino and @Technogeezer, but I felt it would take a full article to do justice to the issues involved, albeit for a diminishing customer base. I launched a lengthy diatribe about the Mac Pro since at least the Trash Can model, but I abandoned it. Too much bile for the holiday season!
I think the bottom line is that if you have such specialized needs or high-end requirements that you can’t configure a current Mac to accommodate them in 2025, macOS really isn’t the right platform for you. That is only going to become more true as time marches on. For those demanding use cases, it’s Linux, one of the dwindling commercial big iron platforms, or even Windows.
The days of a diverse and thriving macOS expansion card market and of competitive Mac-based compute clusters at the very high end are at least 10-15 years behind us.
, but it strikes me that they’ve decided to dispense with addressing the needs of monster users like CERN with over a petabyte a day of data. I imagine it’s a small population of users at that scale even worldwide.