The official statement from Apple (emailed to developers 10 days ago) is that macOS 27 is the “final release to support Rosetta”, so the title is a bit off.
They also say:
> Please note that Rosetta functionality for older, unmaintained gaming titles that rely on Intel-based frameworks will continue to be supported.
I interpret that to mean just enough of Rosetta and Intel frameworks will continue to be around, at least for macOS 28. Not specified which ones, or whether it stays any longer than that.
I’m pretty curious of what that will look like exactly, because there’s a fair amount of system frameworks/libraries needed to get to a bare minimum “hello world” AppKit app. Add on top any number of other frameworks that might be used by “older, unmaintained” games that Apple sees fit to keep supporting. Does this ensure OpenGL is kept on life support? Will they consider Wine important enough to support, perhaps even after they drop native Intel games?
Apple seems to slightly care about supporting Codeweavers/CrossOver from things I've seen, which indirectly makes Wine, Rosetta 2, and GPTK "important enough to support" since they're important features
Wait, so.. how are we supposed to test Intel builds of our macOS apps from now on?
I get it that macOS has to evolve, but that doesn't mean all apps have to drop Intel support at the same time.
On hardware-level apps like my Lunar app I have plenty #if arch(arm64) because some features like reading the brightness nits or reading ambient light is different or completely missing based on the architecture. I need to test the UI differences based on what features are available.
I don't see it viable to stay on macOS 26 for this, especially if we're going to see breaking changes again with the display and window server subsystem like we did with Tahoe. M5 support for Gamma table changes is still broken after so many months [0]
> Wait, so.. how are we supposed to test Intel builds of our macOS apps from now on?
You don't. You could stay on an old MacOS. Apple would prefer that you tell your customers to stop being poor and buy a new computer. They will make your situation increasingly unbearable until you do.
The overwhelming majority of people haven't needed a new computer since 2016. The current economic situation makes a new computer a worse value proposition than it's been in 35 years. Vendors are responding to this situation by manufacturing obsolescence. Microsoft pulled the same stunt with Windows 11's TPM 2.0 requirement.
I think it's a stretch to call Apple's ARM transition "planned obsolescence". The M-series chips are very clear improvements on what came before and there is a clear rationale for that transition.
We're talking here about an OS that hasn't even come out yet, that will get years of security support, for computers that Apple hasn't been selling for several years now. Seems pretty reasonable.
I said "manufactured," not "planned." I don't think Apple intended to do this at the outset. Tim Cook wasn't leaned back in an office chair, twirling a moustache saying "yes, let's make every mac made before 2019 SUCK!"
If it was planned, Rosetta 2 would have never existed in the first place. It would have been a qemu fork haphazardly crammed into Xcode.
There was no "planning" here. Here's how I imagine it went: a developer whined about tech debt, management seized an opportunity to generate revenue, neither party considered, yknow, humans, and now we're here.
Eerily similar story here. My wife was using her 2017 MBP (the one they got sued over) and she adored it until Tahoe suddenly caused Chrome to run like hot garbage. I bought her an open-box M3 Air. She likes the color. It doesn't provide any more value to her life than her 2017 MBP did, and yet we're out $1000 because Apple said so.
That's overly dramatic. I don't think a new Macbook Air today is a worse value proposition than some Mac from 35 years ago. I just checked Apple prices from 1991:
- Mac Classic II, the slowest of the bunch, $1.900, or about $4.661 today
- Quadra 900, the fastest model in 1991, was $7.200 ($17.663 today)
- PowerBook 170 was $4600 ($11.285)
"Value" and "price" aren't the same thing. A new computer in 1991 cost more, but it also covered a vastly increased set of use cases versus a machine from 5 years prior (assuming the hypothetical 1991 computer buyer had even owned a computer before). Today, you can buy a used MBP with an M1 and it will do everything a new MBP can do, and the differences compared to a new machine will be imperceptible to most users.
Plenty of people would even be perfectly happy on an x86 Mac, too. Sure, there would be a perceptible difference compared to a new machine, but not enough to justify the price. That's what obsoleting Rosetta is about, it's about artifically making x86 Macs so unbearable that would-be happy users have no choice but to buy something else.
IIRC Apple supported 10.5 extra long because of it being the last PowerPC MacOS. I wouldn't be surprised if they do something similar here. Keep an intel mac around, and you should be fine
Arguably if you're shipping new fat binary code today, you should already have an Intel Mac around to test, because there might be subtle differences between Intel-on-Rosetta2 and Intel-on-Intel.
I hope they keep around the underpinnings for Rosetta 2 (without the macOS parts) just to keep supporting Intel virtualization for things like Docker. Heck then anyone who really needs to run some old Intel app can run a virtualized older version of macOS.
But I wonder if they're eager to drop support for the Intel TSO memory model from their CPUs.
Oh yeah, I had forgotten about the weird "games" exception. At least that means they'll keep parts of Rosetta 2 around in the code, but they could also end up doing some weird whitelisting for the specific games they want to support and not let anyone else keep using it.
I read somewhere that the part that allows a virtual machine to use Rosetta inside the VM is sticking around.
MacOS on ARM can't directly virtualize an Intel OS using Rosetta today using the native virtualization framework, you need something like qemu for that. But you can use an ARM linux VM with the Rosetta framework installed internally to run x86 containers, which is I think how docker desktop and similar alternatives are handling it.
Wine can run on aarch64 with FEX reasonably well already, no special instructions or hardware acceleration required. There's a bit of extra overhead, but that shouldn't be a problem for old games on modern hardware, they should run about as well.
Interesting, do you know if performance is roughly comparable with Rosetta 2, i.e., are we talking about a few percent of overhead or something more dramatic? (For CPU-bound code; I understand that the overall effect will be smaller due to GPU code probably being executed host-side in native code etc.)
Because they've invested billions into making their own chips? And those chips have pretty damn good performance/power draw/cooling etc. I think the answer is pretty straightforward.
This seems less about why it won't be supporting Intel and more about why Rosetta 2 will be going away, which seems mostly related to cleaning up code that is no longer necessary once Intel is not supported.
I'm curious what options that leaves for docker. I assume the pattern of building/running linux/amd64 containers on MacOS is pretty widespread.
Edit: "Apple says that it will continue to support older, unmaintained gaming titles with Rosetta along with software running Intel binaries in Linux VMs beyond macOS 27 . There could also be future security fixes."
- https://www.macrumors.com/2026/02/16/macos-tahoe-26-4-rosett...
No Apple citation shown for that, though seems plausible.
As a consumer, I’d like to see the end of “universal” builds for various apps. It made sense for a while but downloading and installing ~60% larger bins just doesn’t make sense 6 years later.
If it's anything like the last time they dropped intel, I wouldn't be surprised if they make sure darwin still builds on x86. Going all arm all the time is good business, but it isn't like apple to not have some kind of fallback
Whatever. We have public utility OS, all the hardware vendors should be forced to provide open-source working drivers after they stop supporting their hardware.
If they are afraid of IP leak, well, they can continue support.
My desktop I built in 2012 is still working running ubuntu, even after Intel & MS decided that it is EOL with the release of windows 11.
Wow and darn I guess last support update to fully depreciate intel MacBooks. Used prices already are cratered.
They are great heavily supported Linux machines though. They work out of the box gorgeously with numerous distros and being usbc is nice. For $100-200 for a mint condition model, it isn’t so bad.
>They are great heavily supported Linux machines though.
Since the release of Touch Bar based Macs (which contain apple silicon) this has not been the case. The Macs that are well supported by linux and work very well were abandoned long time ago.
Bad headline. This tweet attempts to explain why Rosetta 2 will no longer work. Which is because the OS no longer supports the Intel platform. That does not explain why the OS does not support the Intel platform.
Because it costs them money to maintain it, and they'll make more money when people upgrade to M series?
In all seriousness, it's a little lame. Consider that the Intel Mac Pro (2019 model) was still selling in 2023! That's not that long ago, and those were their highest end machines in terms of memory capacity. The "new" Mac Pro has since been discontinued...
Though buying a Mac Pro in 2023 would be a bit weird. The writing was on the wall. By that point the M1 was almost out for three years and even the Mac Studio with M1 Ultra had been out for a year.
(IMO it stopped making sense buying an Intel Mac after the M1 Air or if your want to be generous the M1 Pro/Max-based MacBook Pros.)
Maybe Microsoft will finally update the Minecraft launcher to support Apple Silicon. Last I looked they tried to close the bug report, someone reopened it, then there was a system migration and I lost track of it.
It’s almost like they did the work to get the actual game running on Apple Silicon, but installed Rosetta in the process, then just forgot about the launcher.
I always refused to install Rosetta on my Mac, so I could get a big warning if I was about to install something that wouldn’t work in the not too distant future.
I am missing something ? If I read the link in xcancel.com correctly, it says what I would look at as "intel emulation" will be removed in the next release.
So, it looks to me application vendors who depends upon this emulation was given proper notice of this removal. So I think you should complain to the vendors instead of Apple.
Most times I tend to criticize Apple, but this time seems Apple just moving on to avoid "bloat" and "cruft" from being carried forward in future releases.
OpenBSD does things like this all the time and they get praised for it, which I agree with. Apple did the same with this and some people are upset :)
The official statement from Apple (emailed to developers 10 days ago) is that macOS 27 is the “final release to support Rosetta”, so the title is a bit off.
They also say:
> Please note that Rosetta functionality for older, unmaintained gaming titles that rely on Intel-based frameworks will continue to be supported.
I interpret that to mean just enough of Rosetta and Intel frameworks will continue to be around, at least for macOS 28. Not specified which ones, or whether it stays any longer than that.
I’m pretty curious of what that will look like exactly, because there’s a fair amount of system frameworks/libraries needed to get to a bare minimum “hello world” AppKit app. Add on top any number of other frameworks that might be used by “older, unmaintained” games that Apple sees fit to keep supporting. Does this ensure OpenGL is kept on life support? Will they consider Wine important enough to support, perhaps even after they drop native Intel games?
Apple seems to slightly care about supporting Codeweavers/CrossOver from things I've seen, which indirectly makes Wine, Rosetta 2, and GPTK "important enough to support" since they're important features
I read that as "Rosetta2 for 32 bit" will still be around, somehow.
I wonder if Apple cares about docker/podman which uses rosetta on amd64 images
[dead]
Wait, so.. how are we supposed to test Intel builds of our macOS apps from now on?
I get it that macOS has to evolve, but that doesn't mean all apps have to drop Intel support at the same time.
On hardware-level apps like my Lunar app I have plenty #if arch(arm64) because some features like reading the brightness nits or reading ambient light is different or completely missing based on the architecture. I need to test the UI differences based on what features are available.
I don't see it viable to stay on macOS 26 for this, especially if we're going to see breaking changes again with the display and window server subsystem like we did with Tahoe. M5 support for Gamma table changes is still broken after so many months [0]
[0] https://developer.apple.com/forums/thread/819331#819331021
> Wait, so.. how are we supposed to test Intel builds of our macOS apps from now on?
You don't. You could stay on an old MacOS. Apple would prefer that you tell your customers to stop being poor and buy a new computer. They will make your situation increasingly unbearable until you do.
The overwhelming majority of people haven't needed a new computer since 2016. The current economic situation makes a new computer a worse value proposition than it's been in 35 years. Vendors are responding to this situation by manufacturing obsolescence. Microsoft pulled the same stunt with Windows 11's TPM 2.0 requirement.
I think it's a stretch to call Apple's ARM transition "planned obsolescence". The M-series chips are very clear improvements on what came before and there is a clear rationale for that transition.
We're talking here about an OS that hasn't even come out yet, that will get years of security support, for computers that Apple hasn't been selling for several years now. Seems pretty reasonable.
I said "manufactured," not "planned." I don't think Apple intended to do this at the outset. Tim Cook wasn't leaned back in an office chair, twirling a moustache saying "yes, let's make every mac made before 2019 SUCK!"
If it was planned, Rosetta 2 would have never existed in the first place. It would have been a qemu fork haphazardly crammed into Xcode.
There was no "planning" here. Here's how I imagine it went: a developer whined about tech debt, management seized an opportunity to generate revenue, neither party considered, yknow, humans, and now we're here.
I have a MacBook from 2017 and and m3 air today.
For day to day tasks there is no difference.
I have a MacBook Pro from 2016 and an M4 Pro from last year. There is a night and day difference.
I think "M series chips are no better than ten year old Intel chips" is a take that would be somewhat difficult to sustain, given the data.
Eerily similar story here. My wife was using her 2017 MBP (the one they got sued over) and she adored it until Tahoe suddenly caused Chrome to run like hot garbage. I bought her an open-box M3 Air. She likes the color. It doesn't provide any more value to her life than her 2017 MBP did, and yet we're out $1000 because Apple said so.
That's overly dramatic. I don't think a new Macbook Air today is a worse value proposition than some Mac from 35 years ago. I just checked Apple prices from 1991:
"Value" and "price" aren't the same thing. A new computer in 1991 cost more, but it also covered a vastly increased set of use cases versus a machine from 5 years prior (assuming the hypothetical 1991 computer buyer had even owned a computer before). Today, you can buy a used MBP with an M1 and it will do everything a new MBP can do, and the differences compared to a new machine will be imperceptible to most users.
Plenty of people would even be perfectly happy on an x86 Mac, too. Sure, there would be a perceptible difference compared to a new machine, but not enough to justify the price. That's what obsoleting Rosetta is about, it's about artifically making x86 Macs so unbearable that would-be happy users have no choice but to buy something else.
I still prefer my pre-2016 Intel Mac since I can do more things that I want to do on it than my newer M4.
Keep a macOS 26 machine around for testing. All Intel Macs will be stuck on 26 as well, so testing under 26 is probably best anyway.
Virtualize macOS 26 for testing purposes: https://eclecticlight.co/2025/01/21/what-can-you-do-with-vir...
> Wait, so.. how are we supposed to test Intel builds of our macOS apps from now on?
Isn't this a general form of 'how do we deal with the transition from a to b?'
If your client's can get intel Mac's, then you should be able to get one. If they can't, why do you need to keep supporting intel Mac's?
IIRC Apple supported 10.5 extra long because of it being the last PowerPC MacOS. I wouldn't be surprised if they do something similar here. Keep an intel mac around, and you should be fine
> Wait, so.. how are we supposed to test Intel builds of our macOS apps from now on?
In a older version of the OS running in a virtual machine?
Keep an Intel Mac around or drop support.
They followed the same path when moving from PPC to Intel.
And 32-bit to 64-bit.
Keep an Intel Mac around?
Arguably if you're shipping new fat binary code today, you should already have an Intel Mac around to test, because there might be subtle differences between Intel-on-Rosetta2 and Intel-on-Intel.
It works until that machine dies and you need to scramble for a solution (again).
Same way you test them now?
I hope they keep around the underpinnings for Rosetta 2 (without the macOS parts) just to keep supporting Intel virtualization for things like Docker. Heck then anyone who really needs to run some old Intel app can run a virtualized older version of macOS.
But I wonder if they're eager to drop support for the Intel TSO memory model from their CPUs.
Apple will keep Rosetta 2 support for Intel virtualization. See https://developer.apple.com/documentation/apple-silicon/abou...
Oh yeah, I had forgotten about the weird "games" exception. At least that means they'll keep parts of Rosetta 2 around in the code, but they could also end up doing some weird whitelisting for the specific games they want to support and not let anyone else keep using it.
The page doesn’t actually say that explicitly
I read somewhere that the part that allows a virtual machine to use Rosetta inside the VM is sticking around.
MacOS on ARM can't directly virtualize an Intel OS using Rosetta today using the native virtualization framework, you need something like qemu for that. But you can use an ARM linux VM with the Rosetta framework installed internally to run x86 containers, which is I think how docker desktop and similar alternatives are handling it.
Same here. Would be very sad to lose Wine capabilities as well, and presumably these have minimal macOS dependencies.
Wine can run on aarch64 with FEX reasonably well already, no special instructions or hardware acceleration required. There's a bit of extra overhead, but that shouldn't be a problem for old games on modern hardware, they should run about as well.
Interesting, do you know if performance is roughly comparable with Rosetta 2, i.e., are we talking about a few percent of overhead or something more dramatic? (For CPU-bound code; I understand that the overall effect will be smaller due to GPU code probably being executed host-side in native code etc.)
Because they've invested billions into making their own chips? And those chips have pretty damn good performance/power draw/cooling etc. I think the answer is pretty straightforward.
Related? "Apple will phase out Rosetta 2 in macOS 28" https://news.ycombinator.com/item?id=45692116 24-oct-2025 335 comments
This seems less about why it won't be supporting Intel and more about why Rosetta 2 will be going away, which seems mostly related to cleaning up code that is no longer necessary once Intel is not supported.
I'm curious what options that leaves for docker. I assume the pattern of building/running linux/amd64 containers on MacOS is pretty widespread.
Edit: "Apple says that it will continue to support older, unmaintained gaming titles with Rosetta along with software running Intel binaries in Linux VMs beyond macOS 27 . There could also be future security fixes." - https://www.macrumors.com/2026/02/16/macos-tahoe-26-4-rosett...
No Apple citation shown for that, though seems plausible.
AFAIK all ways of running docker on MacOS rely on a amd64 virtual machine with Linux kernel in it.
docker --platform linux/arm64 depends on amd64 ?
I bet it must feel good for the macOS engineers to remove the intel support. Probably much easier to do development for the OS as well
As a consumer, I’d like to see the end of “universal” builds for various apps. It made sense for a while but downloading and installing ~60% larger bins just doesn’t make sense 6 years later.
If it's anything like the last time they dropped intel, I wouldn't be surprised if they make sure darwin still builds on x86. Going all arm all the time is good business, but it isn't like apple to not have some kind of fallback
Whatever. We have public utility OS, all the hardware vendors should be forced to provide open-source working drivers after they stop supporting their hardware.
If they are afraid of IP leak, well, they can continue support.
My desktop I built in 2012 is still working running ubuntu, even after Intel & MS decided that it is EOL with the release of windows 11.
Wow and darn I guess last support update to fully depreciate intel MacBooks. Used prices already are cratered.
They are great heavily supported Linux machines though. They work out of the box gorgeously with numerous distros and being usbc is nice. For $100-200 for a mint condition model, it isn’t so bad.
>They are great heavily supported Linux machines though.
Since the release of Touch Bar based Macs (which contain apple silicon) this has not been the case. The Macs that are well supported by linux and work very well were abandoned long time ago.
Anonymous mirror: https://xcancel.com/Lina_Hoshino/status/2046112493320458649
Focusing is about saying no.
— Steve Jobs
https://youtu.be/H8eP99neOVs (WWDC '97)
This is something Microsoft will never learn, it's not in their DNA.
Bad headline. This tweet attempts to explain why Rosetta 2 will no longer work. Which is because the OS no longer supports the Intel platform. That does not explain why the OS does not support the Intel platform.
Because it costs them money to maintain it, and they'll make more money when people upgrade to M series?
In all seriousness, it's a little lame. Consider that the Intel Mac Pro (2019 model) was still selling in 2023! That's not that long ago, and those were their highest end machines in terms of memory capacity. The "new" Mac Pro has since been discontinued...
Though buying a Mac Pro in 2023 would be a bit weird. The writing was on the wall. By that point the M1 was almost out for three years and even the Mac Studio with M1 Ultra had been out for a year.
(IMO it stopped making sense buying an Intel Mac after the M1 Air or if your want to be generous the M1 Pro/Max-based MacBook Pros.)
But it does?
> Rosetta 2 requires almost the entire OS to have Intel support.
The implication here being that (almost) the entire OS having Intel support is not trivial.
Because Apple is the King of Deprecations. And they get away with it.
> Because Apple is the King of Deprecations.
Google might wear that particular crown: https://killedbygoogle.com
Apple is the King of Hardware Deprecations. Google is the King of Software Deprecations. You're both right.
They are the Kings of Apple Ecosystem Deprecations - not just hardware. I'm comparing them to the x86 and the Windows ecosystem.
Google is the God-King of Killing software.
Maybe Microsoft will finally update the Minecraft launcher to support Apple Silicon. Last I looked they tried to close the bug report, someone reopened it, then there was a system migration and I lost track of it.
It’s almost like they did the work to get the actual game running on Apple Silicon, but installed Rosetta in the process, then just forgot about the launcher.
I always refused to install Rosetta on my Mac, so I could get a big warning if I was about to install something that wouldn’t work in the not too distant future.
Hopefully Sonos will finally get around to it as well.
executive decision
I am missing something ? If I read the link in xcancel.com correctly, it says what I would look at as "intel emulation" will be removed in the next release.
So, it looks to me application vendors who depends upon this emulation was given proper notice of this removal. So I think you should complain to the vendors instead of Apple.
Most times I tend to criticize Apple, but this time seems Apple just moving on to avoid "bloat" and "cruft" from being carried forward in future releases.
OpenBSD does things like this all the time and they get praised for it, which I agree with. Apple did the same with this and some people are upset :)