macOS Tahoe Might Support One Fewer Mac Than Previously Rumored

[sdfox7]

Apple will drop OS support for M1 in 2027. 2020-2027 would be 7 years of support. There is no incentive for Apple to extend the life of those machines because Apple needs people to buy new Macs.

Apple has shareholders to satisfy, so it is a pipe dream to expect M1 to have a longer life just because Apple makes the hardware.

 

[SactoGuy18]

I think MacOS 26 will support any Apple desktop/laptop that runs the M1 or later SoC. Older Intel-based Macs, I have my doubts.

 

[pshufd]

Apple will drop OS support for M1 in 2027. 2020-2027 would be 7 years of support. There is no incentive for Apple to extend the life of those machines because Apple needs people to buy new Macs.

Apple has shareholders to satisfy, so it is a pipe dream to expect M1 to have a longer life just because Apple makes the hardware.

M1 MacBook Pro 14 and 16 came out in late 2021 and the M1 Studio came out in 2022. So you have a case for M1 support through 2029.

 

[ZMacintosh]

they should just drop intel support at this point, continue to release point updates for Sequoia for those if needed.

 

[GioGiusi]

Yep. Which is why I waited for the M1 iMac until I replaced out old 2011 iMac. These computers last forever. Even my 2011, 10 years later, was fine - once I replaced the HDD with a solid state drive. But, the lack of updates for a couple years was the dealbreaker.

My hope was the M1 would give me 6-8 years of OS updates…

You'll probably get more like 15 years, if not more.

 

[seek3r]

Apple will drop OS support for M1 in 2027. 2020-2027 would be 7 years of support. There is no incentive for Apple to extend the life of those machines because Apple needs people to buy new Macs.

Apple has shareholders to satisfy, so it is a pipe dream to expect M1 to have a longer life just because Apple makes the hardware.

I’d bet longer for a few reasons:

* there was a *huge* wave of upgraders in 2020, both because of the ARM machines themselves and because of COV19 and a spike of new computer purchasing to handle WFH.

* Apple is still selling new (not new old stock, they’re supplying new machines) M1 Macbook Airs through Walmart as their distribution channel

Those two above means the install base for M1 is almost certainly far bigger than any given Intel deployment so MTBF will still take a long time, longer than previously or probably subsequently to winnow down userbase of these machines.

* Apple’s income is driven a great deal by services today, if they think M1 users are making them more money staying put and paying for services rather than being forced to choose whether to stay in the Apple ecosystem by needing a new machine (most apropos for those walmart machines) they’ll keep support longer

* the cost of maintaining the codebase and support for M1 machines should be far lower than Intel, both because of whatever Apple’s been having to pay Intel to maintain their licenses for tooling and access and because platform-wise there’s not really a massive difference between M1-3. Yes maintaining testing infrastructure and qa coverage still has a huge cost but given the install base it’s likely justifiable.

I’d bet the M1s will match or beat (I’d bet on beat) the legs of machines like the mid 2012 13” macbook pro which got 9 OS releases over 8 years (plus a couple more years of security patches)

 

[padams35]

Apple will drop OS support for M1 in 2027. 2020-2027 would be 7 years of support. There is no incentive for Apple to extend the life of those machines because Apple needs people to buy new Macs.

Apple has shareholders to satisfy, so it is a pipe dream to expect M1 to have a longer life just because Apple makes the hardware.

Nah, I'd bet M1 will still be supported in 2027 by MacOS28. The M1 iMac skipped past M2 entirely and did not get an upgrade option until the M3 iMac was released late 2023. To not support it in MacOS28 would give less support than the last Intel models are expected to receive. The M1 Wallmart special carries less weight, but stock is still sold new and I don't think Apple will drop support less than two years after that special ends either.

No bet on 2028 / MacOS29 M1 support.

 

[seek3r]

Nah, I'd bet M1 will still be supported in 2027 by MacOS28. The M1 iMac skipped past M2 entirely and did not get an upgrade option until the M3 iMac was released late 2023. To not support it in MacOS28 would give less support than the last Intel models are expected to receive. The M1 Wallmart special carries less weight, but stock is still sold new and I don't think Apple will drop support less than two years after that special ends either.

No bet on 2028 / MacOS29 M1 support.

I legit honestly think we’ll see full M1 support through 2030

 

[Corefile]

Steve would never have allowed the disaster that is Apple Intelligence to occur. That's all on Tim Apple.

 

[seek3r]

Steve would never have allowed the disaster that is Apple Intelligence to occur. That's all on Tim Apple.

I seem to remember more than a few flubs on Job’s watch, including but not limited to:

* The initial 10.5 (Leopard) release, which was an unstable disaster that literally spawned an entire subsequent release that was mostly dedicated to bug swatting: 10.6 (Snow Leopard)

* The iPhone 4 “antennagate” “you’re holding it wrong” issue

* The initial launch of Apple Maps, developed under Jobs, fixed by Cook (which, btw, I think is a close analog to Apple Intelligence, AM was a complete mess that was delayed and then massively fell short of its promises on initial release, it’s fantastic now)

* the ergonomic hell that was the goddamn puck mouse

* iTunes Ping, when Jobs thought it would be a great idea to launch Apple into the social media world using iTunes as the base.

And I’m not even counting things that were just plain product flops (The G4 cube, which Apple has finally managed to create the far superior and thus popular version of in the studio - under cook), the HiFi, etc

Jobs was awesome at Apple, but he was far from perfect. Cook’s doing fine.

 

[Lioness~]

My Macs will be good longer then I'll usually keep them.
My old iPad mini 5 might living on the edge though, even as a timer only.
I'll trade it in soon, so don't lower the high value for it Apple 😉

 

[bzgnyc2]

Yeah, the amount my “playable on mac” section in steam dropped still irritates me

As I said in another comment they just introduced a new M3 variant, so it’s gonna be quite a while on that one, probably 2031 or 2032.

That may be a more likely delineator

I think a few different issues here:
-From a pure I don't want to support the past perspective, we can make a case for all sorts of things
-Just because it would make their lives easier, we all agree commercially and reasonably they can't
-On the other hand, they are dropping things that don't have much commercial justification

All the arguments about the future of M1 or whatever that look at it from a technical perspectie ignore that that doesn't appear to be a driving issue for any of their decisions to drop other models. What's happening to the Intel MacBook Air 2020 will happen to the MacBook Air M1 soon enough.

MacOS as an appropriate alternative to Windows on the desktop (the selling point for all users), and one that is a full unix (the additional selling point for pro users) is the biggest single reason people use Macs, after that is the ecosystem and integrated services. I dont think it would save any money at all, in fact I think it would kill Macs off entirely

I agree -- not arguging for that at that all. That's actually my whole point. Just because something is easier for Apple doesn't make it a good idea. The key is the value -- if retaining support or improving something improves value for the customer then it is good even if it costs Apple money.

Also they have to maintain Darwin and most of the other layers and APIs anyway for iOS etc anyway

Some of the components of Rosetta 2 are integrated into the silicon, support for x86_64 memory ordering and specific x86 flags to use it and other execution optimizations are baked into AS chips. Just opensourcing the software wouldnt be all that useful. That said I doubt Apple will remove any of that any time soon, outside anything else it enables a lot of useful tools in the virtualization/emulation space used by a lot of professionals (me included).

I agree -- I hope they don't kill Rosetta2. I think others' argument was 'kill Intel' -> 'kill Rosetta2" -> 'kill everything I don't use'. All I am really arguing for on this is that if Rosetta2 is too much to maintain they don't kill the IP / block others from offering the capability.

 

[bzgnyc2]

Of course there’s a logical reason. You’ve been sneaking by at under $1000/year for professional use and they’re a business not a charity.

$1000/year is relatively high for Apple. A base Mac Mini M4 will likely come in around $100/year. Given that includes hardware clearly macOS doesn't cost that much to support.

They want you in a new Mac Studio pronto. What’s the confusion?

He already has the iMac Pro, it works, and it works well for his needs. Not everyone looks forward to getting a new machine with a new architecture for which there may not be native versions of some software, finding and buying a new display and calibrating it, etc.

Granted not everything can last forever. Then I will argue -- with some bias I admit -- that I find it slightly less justifiable that Apple is dropping the MacBook Air 2020 the same year -- none of the architectual limitations one can acuse the iMac Pro as having and it is 3 years newer. Only slightly though given the Pro designation and the iMac Pro was leading edge for its time.

 

[sdfox7]

I’d bet longer for a few reasons:

* there was a *huge* wave of upgraders in 2020, both because of the ARM machines themselves and because of COV19 and a spike of new computer purchasing to handle WFH.

* Apple is still selling new (not new old stock, they’re supplying new machines) M1 Macbook Airs through Walmart as their distribution channel

Those two above means the install base for M1 is almost certainly far bigger than any given Intel deployment so MTBF will still take a long time, longer than previously or probably subsequently to winnow down userbase of these machines.

* Apple’s income is driven a great deal by services today, if they think M1 users are making them more money staying put and paying for services rather than being forced to choose whether to stay in the Apple ecosystem by needing a new machine (most apropos for those walmart machines) they’ll keep support longer

* the cost of maintaining the codebase and support for M1 machines should be far lower than Intel, both because of whatever Apple’s been having to pay Intel to maintain their licenses for tooling and access and because platform-wise there’s not really a massive difference between M1-3. Yes maintaining testing infrastructure and qa coverage still has a huge cost but given the install base it’s likely justifiable.

I’d bet the M1s will match or beat (I’d bet on beat) the legs of machines like the mid 2012 13” macbook pro which got 9 OS releases over 8 years (plus a couple more years of security patches)

If you’re right, then imagine how poorly the M1 will run with only 8GB RAM, which is what the bulk of machines have. Plus no AI. IIRC you basically had to special order the 16GB MBA. Apple would have to make the OS lean enough to run well on 8GB RAM, which then discourages hardware upgrades.

The 2012 MacBook Pro MD101 (which I own) was an anomaly. I think it sold and was supported for so long due to to being the last upgradable MBP, along with a SuperDrive. Same with the iPhone 6S/SE being the last phone with the headphone jack, and also supported across 7 iOS versions (9-15).

 

[Blackstick]

I find it slightly less justifiable that Apple is dropping the MacBook Air 2020 the same year -- none of the architectual limitations one can acuse the iMac Pro as having and it is 3 years newer. Only slightly though given the Pro designation and the iMac Pro was leading edge for its time.

Those MacBook Airs were $499 at Walmart toward the end of their cycle. I think it's fair. The iMac Pro is a bigger question but reality is, 2017 was ages ago. My daughter born in June of '17 just turned 8. An 8 year old computer is like a 13 year old dog.

 

[seek3r]

If you’re right, then imagine how poorly the M1 will run with only 8GB RAM, which is what the bulk of machines have. Plus no AI. IIRC you basically had to special order the 16GB MBA. Apple would have to make the OS lean enough to run well on 8GB RAM

Most folks who have 8gb models are doing fine with 8gb models. If all you’re doing is relatively light usage, which is most users, it’s perfectly fine and should be for a few years to come.

Should the base have been 16? Sure. Is 8gb really holding most folks with it back? Not really.

Most people who needed more ram bought more ram

 

[VivienM]

What, bought more than all the market leading PC manufactures?? That's a stretch.

So, I don't know, but my sense is that the 'market leading PC manufacturers' would have been buying a much lower-end mix of chips, at least for laptops. i3s, i5s, the U-series i7s. Not the crazy fancy chips like the i7-48xxHQs and successors. But... maybe I'm underestimating the market for mobile workstations and high-end PC gaming laptops.

 

[VivienM]

Steve would never have allowed the disaster that is Apple Intelligence to occur. That's all on Tim Apple.

Right. He would have understood that artificial idiocy is the netbook and the Flash of the 2020s and would have told all the pundits where to shove their generative AI.

 

[VivienM]

I agree -- I hope they don't kill Rosetta2. I think others' argument was 'kill Intel' -> 'kill Rosetta2" -> 'kill everything I don't use'. All I am really arguing for on this is that if Rosetta2 is too much to maintain they don't kill the IP / block others from offering the capability.

The problem with Rosetta 2 existing is that there are a number of people currently offering Intel-only software for no good reason other than the fact that 'hey, it runs on the current Macs so who cares'.

In fact, I would argue that those are the only people who haven't ported their software to ARM. The Mac loyalists (BBEdit, GraphicConverter, PCalc) did so long ago, the big developers (Adobe, Microsoft, etc) did so a little later but also long ago, the Windows-people-who-don't-follow-Mac-cycles (like VanDyke - the people who make SecureCRT - or Citrix) added ARM support in their next major scheduled release in 2022 or so.

The only people left are the lazy idiots who don't care that their Electron (or non-Electron apps) are running in emulation. The only way to get those people to get on with recompiling their stuff is for Apple to give a big warning when their apps are being opened - "this software will not work on a future release of macOS", like they did with the 32-bit apps. Otherwise their apps will run in Rosetta 2 forever.

Maybe that's the answer - put a big annoying warning sooner rather than later, make it more annoying the next year, and then maybe not actually yank out Rosetta 2 for a while.

 

[VivienM]

Tim Apple might be a safe pair of hands but Apple is being left behind in innovation and future product strategy. The Steve Jobs playbook is now empty which is why Apple's long term future is bleak right now. It's called stagnation and is the killer of complacent companies.

It is? Most of the product line is certainly stronger today than, say, during the 2015-2019 period of peak Iveism.

And Tim Apple has been more willing to compromise, e.g. big phones were clearly a good idea, bringing back ports was clearly a good idea, etc.

 

[VivienM]

The 2012 MacBook Pro MD101 (which I own) was an anomaly. I think it sold and was supported for so long due to to being the last upgradable MBP, along with a SuperDrive. Same with the iPhone 6S/SE being the last phone with the headphone jack, and also supported across 7 iOS versions (9-15).

Most machines from that 2012-2015 period seemed to get a little more OS updates that normal. Same with the retina 2013, 2014, etc.

I suspect a part of it was simply that they didn't have a good technical excuse for dropping them the way they did with, say, the 32-bit EFI systems.

Then by the time you get to the 2016ish machines, you see they are back in a 'want to start dropping things' mood because of the T2, ARM, etc.

 

[escapers]

What about the 2019 MacBook Pro 15.4-inch? It's almost the same as the 16"

 

[VivienM]

What about the 2019 MacBook Pro 15.4-inch? It's almost the same as the 16"

Nope.

I think their logic would probably be - that model was discontinued fall 2019, i.e. 6 years ago, while the 16" continued until the launch of the M1 Pro/Max in fall 2021...

 

[JPack]

What about the 2019 MacBook Pro 15.4-inch? It's almost the same as the 16"

Apple always looks at launch date first, capability second. The 15.4-inch model doesn't pass the first hurdle because it was launched May 2019. The 16-inch model was launched November 2019.

If you work backwards, macOS 26 will be launched September 2025. That puts the 15.4-inch model over 6 years.

Apple always provides only 6-7 years of major OS updates.

 

[pshufd]

The problem with Rosetta 2 existing is that there are a number of people currently offering Intel-only software for no good reason other than the fact that 'hey, it runs on the current Macs so who cares'.

In fact, I would argue that those are the only people who haven't ported their software to ARM. The Mac loyalists (BBEdit, GraphicConverter, PCalc) did so long ago, the big developers (Adobe, Microsoft, etc) did so a little later but also long ago, the Windows-people-who-don't-follow-Mac-cycles (like VanDyke - the people who make SecureCRT - or Citrix) added ARM support in their next major scheduled release in 2022 or so.

The only people left are the lazy idiots who don't care that their Electron (or non-Electron apps) are running in emulation. The only way to get those people to get on with recompiling their stuff is for Apple to give a big warning when their apps are being opened - "this software will not work on a future release of macOS", like they did with the 32-bit apps. Otherwise their apps will run in Rosetta 2 forever.

Maybe that's the answer - put a big annoying warning sooner rather than later, make it more annoying the next year, and then maybe not actually yank out Rosetta 2 for a while.

Another category is large companies that can't because their platform is Windows and they only run on macOS via WINE. Fidelity's Active Trader Pro is an example of this. They have $4.9 Trillion AUM so they are a huge company. If you want to run their program well, then you run on Windows. There's a 20% performance hit on M4 Macs compared to my 2020 i7-10700 Windows system.

Intuit TurboTax is the other reason why I run it on my iMac Pro. I may have to run it on Windows after I lose support on my iMac Pro.