WWDC 2025

[Altaic]

According to ArsTechnica, Rosetta 2 will be phased out after macOS 27:

Apple details the end of Intel Mac support and a phaseout for Rosetta 2

Rosetta app translation features for Intel apps won’t stay around indefinitely.

arstechnica.com

But after that, Rosetta will be pared back and will only be available to a limited subset of apps—specifically, older games that rely on Intel-specific libraries but are no longer being actively maintained by their developers.

That's a bit of a fuzzy definition. I'm wondering what that means for CrossOver. Maybe they have to switch to FEX like Parallels did for the x86 emulation.

That’s a shame, though I could see how Apple wouldn’t want to maintain x86 libs and silicon support. Still, I’m not sure exactly what subset of Rosetta would remain, which is important to understand what the longer term capabilities will actually be.

Original source: https://developer.apple.com/documentation/apple-silicon/about-the-rosetta-translation-environment/

 

[KingOfPain]

That’s a shame, though I could see how Apple wouldn’t want to maintain x86 libs and silicon support. Still, I’m not sure exactly what subset of Rosetta would remain, which is important to understand what the longer term capabilities will actually be.

Original source: https://developer.apple.com/documentation/apple-silicon/about-the-rosetta-translation-environment/

Maybe they just drop the possibility to run x86-64 macOS applications, but leave the binary translator?
Otherwise, the APGPTK would be dead in the water as well.
Or they put the binary translator in a library that has to be bundled with an application that wants to use it?

 

[quarkysg]

Maybe they just drop the possibility to run x86-64 macOS applications, but leave the binary translator?

This would be my guess as well.

 

[throAU]

I swear apple at this point are looking at the windows code base and thinking:

"what can we do that will take microsoft a decade to rewrite the windows UI to copy"

and doing that, whether or not its a good idea or not.

 

[Nycturne]

So I've been playing with SwiftUI in the new Xcode Beta to see how badly iOS 26 breaks some of the workarounds I've had to implement to mimic some things apps do in UIKit.

And boy, seems like they want apps like mine to migrate over to a TabView + Sidebar setup. My implementation of a mini player on iOS conflicts with the new Search Bar positioning on the bottom of the screen, and you can seemingly only attach accessories to the TabView, not a Navigation Split View which is what I use. How fun.

 

[mr_roboto]

That’s a shame, though I could see how Apple wouldn’t want to maintain x86 libs and silicon support. Still, I’m not sure exactly what subset of Rosetta would remain, which is important to understand what the longer term capabilities will actually be.

Original source: https://developer.apple.com/documentation/apple-silicon/about-the-rosetta-translation-environment/

"Rosetta was designed to make the transition to Apple silicon easier, and we plan to make it available for the next two major macOS releases – through macOS 27 – as a general-purpose tool for Intel apps to help developers complete the migration of their apps. Beyond this timeframe, we will keep a subset of Rosetta functionality aimed at supporting older unmaintained gaming titles, that rely on Intel-based frameworks."

So: after macOS 27 we'll still have the emulator, but they will stop shipping fully current x86 versions of most system libraries. Instead, they'll select the most recent SDK version required to support old games, and will indefinitely maintain those libraries as Intel binaries going forward. I expect they'll work out a way to avoid using fat binaries at all; the system will ship as 100% Arm and if it detects you trying to run a legacy Intel game it will download and install the legacy x86 libraries in a separate place, as well as installing the emulator.

Reading between the lines: Apple got the message that many game devs were incredibly pissed about long-tail sales of 32-bit x86 games ending with Catalina. They've belatedly realized they can't push that industry into keeping up with their preferred level of compatibility churn, and that if they're to repair relations, they can't possibly do it again in macOS 28. So now, for the first time in ages, they're carving out an exception to their general policy of "if things are too old they won't run".

 

[leman]

"Rosetta was designed to make the transition to Apple silicon easier, and we plan to make it available for the next two major macOS releases – through macOS 27 – as a general-purpose tool for Intel apps to help developers complete the migration of their apps. Beyond this timeframe, we will keep a subset of Rosetta functionality aimed at supporting older unmaintained gaming titles, that rely on Intel-based frameworks."

So: after macOS 27 we'll still have the emulator, but they will stop shipping fully current x86 versions of most system libraries. Instead, they'll select the most recent SDK version required to support old games, and will indefinitely maintain those libraries as Intel binaries going forward. I expect they'll work out a way to avoid using fat binaries at all; the system will ship as 100% Arm and if it detects you trying to run a legacy Intel game it will download and install the legacy x86 libraries in a separate place, as well as installing the emulator.

Reading between the lines: Apple got the message that many game devs were incredibly pissed about long-tail sales of 32-bit x86 games ending with Catalina. They've belatedly realized they can't push that industry into keeping up with their preferred level of compatibility churn, and that if they're to repair relations, they can't possibly do it again in macOS 28. So now, for the first time in ages, they're carving out an exception to their general policy of "if things are too old they won't run".

What they could d is remove the ability to build x86 binaries on newer toolchains. Then the problem kind of solves itself.

 

[Andropov]

Reading between the lines: Apple got the message that many game devs were incredibly pissed about long-tail sales of 32-bit x86 games ending with Catalina. They've belatedly realized they can't push that industry into keeping up with their preferred level of compatibility churn, and that if they're to repair relations, they can't possibly do it again in macOS 28. So now, for the first time in ages, they're carving out an exception to their general policy of "if things are too old they won't run".

To be honest it always baffles me how long it took for the PC world to switch to 64 bits for good. Some of the 32 bit games that no longer run on macOS were released more than a *decade* after the first 64-bit Mac. Even from big publishers like Activision. This shouldn't have been an issue.

 

[quarkysg]

What they could d is remove the ability to build x86 binaries on newer toolchains. Then the problem kind of solves itself.

Don’t newer toolchains also have to support generating binaries for prior generation OS?

 

[leman]

Don’t newer toolchains also have to support generating binaries for prior generation OS?

That should be easy enough to do, if they want to. Although, as we know Apple they are likely to just drop support, to the ire of open source community

 

[KingOfPain]

Some of the 32 bit games that no longer run on macOS were released more than a *decade* after the first 64-bit Mac. Even from big publishers like Activision. This shouldn't have been an issue.

I guess the game companies wanted to support users of 32-bit systems as well. They could have built a fat binary with x86 and x86-64 code, but then they would have had to test both as well.
Sometimes I really wish that Apple had switched to Intel maybe half a year later, then there would have been no 32-bit Intel Macs at all.
But Apple was desperate to get new laptops out, since G4 was outdated and G5 simply unusable for laptops (and IBM didn't care, having gotten contracts with Microsoft and Sony for the Xbox 360 and the PS3, respectively).

 

[Andropov]

I guess the game companies wanted to support users of 32-bit systems as well. They could have built a fat binary with x86 and x86-64 code, but then they would have had to test both as well.

Oh 32-bit support was a requirement in those years for sure. But supporting both 32 and 64 bits from the ground up shouldn't have been difficult (migrating a 32 bit game to 64 bits is a different thing, though).

I guess the game companies wanted to support users of 32-bit systems as well. They could have built a fat binary with x86 and x86-64 code, but then they would have had to test both as well.
Sometimes I really wish that Apple had switched to Intel maybe half a year later, then there would have been no 32-bit Intel Macs at all.
But Apple was desperate to get new laptops out, since G4 was outdated and G5 simply unusable for laptops (and IBM didn't care, having gotten contracts with Microsoft and Sony for the Xbox 360 and the PS3, respectively).

Yeah, they went backwards (and then forwards again) during those 6 months, from 64 bit PowerPC (the G5) -> 32 bit Core Duo -> 64 bit Core 2 Duo.

Still, this was a 20 years ago, it's bizarre to me that ramifications of that transition are hunting us to this day.

 

[dada_dave]

Reading between the lines: Apple got the message that many game devs were incredibly pissed about long-tail sales of 32-bit x86 games ending with Catalina. They've belatedly realized they can't push that industry into keeping up with their preferred level of compatibility churn, and that if they're to repair relations, they can't possibly do it again in macOS 28. So now, for the first time in ages, they're carving out an exception to their general policy of "if things are too old they won't run".

In addition to that, Rosetta 2 is kind of important to GPTK and Apple’s gaming push so that level of functionality will still be needed going forwards.

 

[tomO2013]

I gotta say, highlight of WWDC for me was iPad OS. Big YESS all around…
I love the improvements to files app, I love the new window manager improvements and I appreciate the ability to have background tasks.
This is bringing MacOS and iPadOS that much closer and I think finally justifies an m4 iPad Pro for what those capabilities and workflows will enable.

 

[Jimmyjames]

I gotta say, highlight of WWDC for me was iPad OS. Big YESS all around…
I love the improvements to files app, I love the new window manager improvements and I appreciate the ability to have background tasks.
This is bringing MacOS and iPadOS that much closer and I think finally justifies an m4 iPad Pro for what those capabilities and workflows will enable.

I think it allows Apple to meaningfully serve the lower end market. People who want a basic computer, without having to compromise the Mac.

 

[Cmaier]

I gotta say, highlight of WWDC for me was iPad OS. Big YESS all around…
I love the improvements to files app, I love the new window manager improvements and I appreciate the ability to have background tasks.
This is bringing MacOS and iPadOS that much closer and I think finally justifies an m4 iPad Pro for what those capabilities and workflows will enable.

I played with iPadOS 26 last night for a bit. Performance aside, the window management stuff is already pretty great. I haven’t quite figured out all the gestures (or at least i don’t have muscle memory for them), but it’s so much better than their last several attempts at it. I hope next year they go the rest of the way, and treat other non-windowed apps in Expose as if they are desktops that you can drag stuff to, etc. (Right now I couldn’t. Maybe I was supposed to be able to. I dunno. Expose was a little confusing to me, but maybe I wasn’t doing it right or maybe it’s not fully ready yet).

 

[swk9]

I'm curious if we've discovered any hint in the new OS releases about the foldable hybrid that's reportedly under development. Given the high price I'd imagine it's more likely to be a Mac instead of an iPad but not sure how Apple will address the hybrid part. Is it just going to be a touch screen Mac that can also run iPad apps?

 

[jbailey]

I played with iPadOS 26 last night for a bit. Performance aside, the window management stuff is already pretty great. I haven’t quite figured out all the gestures (or at least i don’t have muscle memory for them), but it’s so much better than their last several attempts at it. I hope next year they go the rest of the way, and treat other non-windowed apps in Expose as if they are desktops that you can drag stuff to, etc. (Right now I couldn’t. Maybe I was supposed to be able to. I dunno. Expose was a little confusing to me, but maybe I wasn’t doing it right or maybe it’s not fully ready yet).

The windowing gestures are mostly broken in this release according to the release notes.

 

[diamond.g]

How is Apple handling the molasses feeling input lag when recommending at least 30 fps for MetalFX Frame Interpolation? FSR/DLSS FG both recommend frame rates over 60, and even then have antilag/reflex technologies to help things along.

 

[Yoused]

This is bringing MacOS and iPadOS that much closer and I think finally justifies an m4 iPad Pro for what those capabilities and workflows will enable.

A maxed out Macbook Air (32GB/2TB, upper-bin SoC) is $700 (39%) cheaper than a maxed-out iPadPro (16GB/2TB, not nanotextured). The iPadPro has a slightly larger (in pixels: 7.5%) display, which is certainly not enough to justify the difference in price. Of course, the touch screen, along with 5G, is definitely a touch above the MBA, though, to get the most out of it, you would want a Pencil 2.

For the difference in price, I would like to be able to do more with the iPadPro than it is presently capable of. The MacBook does allow me to side-load things like Inkscape and Gimp, which are pretty decent tools for basic use, and the Apple productivity apps are definitely better than the iPadOS versions. If the Pro offered me a little more, I would consider it a practical purchase. But I still cannot build practical programs on it.