Why drop this, though? If it works, the main maintenance would be to fix security exploits?
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Apple to Phase Out Rosetta 2 Starting With macOS 28 as Intel Era Ends
- Thread starter MacRumors
- Start date
- Sort by reaction score
You're completely missing the point.
The users of those Apps are the ones that suffer.
This is one of the things I honestly hate about the Apple platform, despite loving many other things about it.
They don't care about keeping older software working.
You can still run Windows applications from the late 90s on Windows 11 and they often work perfectly without modification. Good luck running any Mac software from even half that long ago.
They did it with MacOS 9 to Mac OS X. Then with PowerPC to Intel 32 bit. Then they did it with Intel 32 bit to Intel 64 bit. Now they're doing it with Intel to ARM.
Every decade or so Apple completely breaks older software. It's like a tradition of theirs.
With Rosetta 1 they had licensed code they had to pay for, so they had an excuse. Not so with Rosetta 2. It would be trivial for them to just keep Rosetta 2 going perpetually. But they won't, because they just don't care about backwards compatibility long-term.
They don't care about keeping older software working.
You can still run Windows applications from the late 90s on Windows 11 and they often work perfectly without modification. Good luck running any Mac software from even half that long ago.
They did it with MacOS 9 to Mac OS X. Then with PowerPC to Intel 32 bit. Then they did it with Intel 32 bit to Intel 64 bit. Now they're doing it with Intel to ARM.
Every decade or so Apple completely breaks older software. It's like a tradition of theirs.
With Rosetta 1 they had licensed code they had to pay for, so they had an excuse. Not so with Rosetta 2. It would be trivial for them to just keep Rosetta 2 going perpetually. But they won't, because they just don't care about backwards compatibility long-term.
For the same reason they dropped PowerPC support 5 years after the Intel transition. Apple has never made backward compatibility a priority. If you want to use old hardware and software indefinitely, Microsoft is your vendor.
Do you think a company that spent $50k in 2019 is still using those?
Yes as thats still a DECADE of support and security updates where Windows 11 does not support some 4 year old CPUS and windows 10 security update support ends this year making them only get like 4-5 years of life out of some systems.
What do you mean? People buy stuff for reasons other than those that I personally buy stuff for?
Preposterous
This is surprising considering that they also announced their container runtime which can run x86_64 images. So that's at release already a dead horse. I did hope that Apple keeps Rosetta 2 around precisely because of its usefulness around VMs and containers.
This won’t happen. If the code isn’t compiled for Intel, there are no hacks or end runs.
Would be nice if someone like crossover would create a WINE version from macOS to run older Intel 32 bit at 64 bit applications, right now we could run 32 bit windows games. 
Have to keep a dedicated 32 bit Intel macOS to run all my legacy software.
Have to keep a dedicated 32 bit Intel macOS to run all my legacy software.
Hopefully this pushes HP to update their scanner software. It still requires Rosetta 2.
I wonder if Apple will provide an uninstaller for Rosetta 2 at some point. The only thing that bugs me about it is that the OS offers to install it, but provides no mechanism to remove it once you do.
I wonder if Apple will provide an uninstaller for Rosetta 2 at some point. The only thing that bugs me about it is that the OS offers to install it, but provides no mechanism to remove it once you do.
MacOS X/macOS has never had a native uninstaller unfortunately. It's unfortunate when there have been examples of robust installer systems around for decades (not to mention the Windows example).
More excusable back when most Mac apps were drag and drop out of a .dmg but we seem to be past those days on macOS without a corresponding Apple-standard package manager/uninstaller. Neither for Apple's software nor 3rd-party software using Apple's installation frameworks/formats.
If you have the budget to drop $50,000 per computer, you aren't still using that computer in 2025. It was likely sold off last year at the absolute latest, more likely it was replaced by 2022 and sold to a decomissioner for pennies on the dollar.
Good. Many developers with resources are sitting on the laurels forever.
For instance, Logitech updated their Options+ for Apple Silicon, but not G HUB (used for gaming input devices). Microsoft updated Minecraft's gaming engine, but not the launcher. And so on.
For instance, Logitech updated their Options+ for Apple Silicon, but not G HUB (used for gaming input devices). Microsoft updated Minecraft's gaming engine, but not the launcher. And so on.
You're making an assumption that dropping R2 support = "Devs will update older software"
It's equally likely that software just gets abandoned, which is bad for users (that's us!)
I would tweak that to say there will be no practical solutions starting from Apple's Apple Silicon-only releases. I think it was earlier this year that someone demonstrated Linux running on a 386 emulator running on an 8088. A 32-bit protected mode OS running on an emulated CPU that was neither. Of course completely impractical. It took like 2 days to boot. Whoever did that had to be almost literally insane.
What the community might consider is putting its efforts into things like PureDarwin, GNUstep, Darbat, and/or Darling. I'd be very much into a MacOS-like OS not tied to Apple deprecation schedules that could run commercial macOS applications (supporting 32 and 64-bit Intel binaries and Apple Silicon binaries on either platform via AOT, JIT, etc emulators). Ton of work...
Exactly
I have a 2019 Mac Pro that I use for scientific research and 3D CT and CAD work every day. I’ll update it to Tahoe when it’s out but I’m fine with it being on Tahoe for the foreseeable future. My apps won’t stop working or anything. We have a 5K iMac Pro that still runs most apps and gets security updates even now.
Isn’t it on those developers to update them? Speaking as someone who’s actively developing research apps currently, it doesn’t take that long to update your code, and even if it does, they’re giving you a heads up a whole year in advance.
The only case where it affects users is legacy apps that are no longer being maintained by the developer. But if that’s the case you’d be better off getting an alternative now anyway because one day or the other it’ll inevitably stop working. Eg: when they ended 32 bit support, and for good reason.
I think this is a mistake. Not because we need to run legacy Mac software that hasn't been ported to ARM / Apple Silicon yet, there has been more than enough time for that, but because so much new Intel x86/64 code is being written every day that is not Mac-specific or written by Mac developers yet is used on Macs every day. I maintain codebases that still rely on huge x86 code libraries from third-parties. Some are actively working on ARM versions. Others adamantly believe if it's not broke it doesn't need fixing.
In the PPC days it made sense to kill Rosetta after a few years because the only target devices for PPC code were Macs and there no new PPC Macs. The market and need quickly evaporated. Today, Intel-based computers and the code for them is not going away anytime soon and Rosetta could leverage that indefinitely with little impact to Apple.
In the PPC days it made sense to kill Rosetta after a few years because the only target devices for PPC code were Macs and there no new PPC Macs. The market and need quickly evaporated. Today, Intel-based computers and the code for them is not going away anytime soon and Rosetta could leverage that indefinitely with little impact to Apple.
Last edited:
Imagine spending $50K on a computer to presumably produce paid work and not having an upgrade plan.
That's not how companies work. Anybody buying 10, 20 or 50 high-end Mac Pros is 1) using them to create deliverables for money and 2) absolutely has an upgrade plan set up. Computers in a production environment have a shelf life and any CTO authorizing the purchase of 2.5 mil worth of Macs would definitely be making a business case for buying them, and assuming a certain period of useful life.
And if you're an individual who bought this hypothetical $50K machine, then you had better be using it to create at least that much billable work over 5 years or you're just burning money up for bragging rights or to feed a hobby.
