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.
MBP with Apple Silicon....When?
- Thread starter rhyzome
- Start date
- Sort by reaction score
No one knows for sure, obviously. But if the rumor mill is to be trusted, you'll see the 13.3" MacBook Pro making the switch to Apple Silicon and away from Intel sometime between September and Christmas. Apple is probably ready to, from a technology standpoint, make the switch on this machine now. Same with the MacBook Air.
There seems to be no word on the 16". If I had to guess, I'd guess between June of 2021 and January of 2022, with similar predictions for the iMac Pro and/or Mac Pro.
The 13" model ought to not be all that bad. It's the higher-end Macs that I'd be worried about.
If suspect that machine will be awesome for non-x86 code. I would buy one myself but I need to run Parallels (I'm not that scared of buying the first generation as long as the reviews look good). I think the rumors point to the end of this year (that is very fast in my opinion).
they‘ve been working on it for many many years, so end of the year isn’t that fast.
I feel the same way, which puts me in a bind as the current 16" is technically the first revision of that model...
Nope. Windows on ARM will not be licensed to non-OEMs. i.e.: even if it works in Parallels, Microsoft will not let you have it.
And since Parallels is virtualization, and not emulation, you can also kiss regular x86 Windows 10 goodbye.
In essence, the new ARM MacBook will not be able to run Windows at all. At least not at launch.
Well, microsoft won’t let you have it NOW, but we’ll see. It wouldn’t run without modification anyway - the page size is wrong.
Because (unfortunately) I need to run Windows (x86) software. There will be no virtual machines for x86.
Yeah, but even if they let me have Windows on ARM, I don't see why anyone would want it.
It's not the same thing as Windows 10 x86. Very few applications aside from Microsoft's own actually run, if at all. It doesn't even support the same drivers as Windows 10. So in essence, it'll just "look" like Windows 10, but it's almost completely useless for anything else. I should know because I played around with a Surface Pro X and honestly, I'm not impressed at all.
If Microsoft see they could make profit of this because of Apple, pretty sure they will want to open up license keys for Windows ARM for anyone.
You can run x86-32 bits software within Windows ARM. They say support for x86-64 bits is coming soon. So if this is possible, you could, with certainly a performance hit, run x86-64 apps within a virtualized Windows ARM on a ARM Mac through Parallels.
It could be totally useless for a moment, but maybe if we give it a couples of years (years is more accurate time frame when we talk about Microsoft), Windows ARM could be useful.
But yeah, at launch, I don't expect this to work at all..
Considering the chips are an extension of the iPhones A series chips that Apple has been working on for like 12 years, is it really first gen silicon? It's coming in new form factors sure, but this isn't exactly a first gen thing for Apple. I think, like with any transition like this, the main issues will just be waiting for 3rd party developers to port software over. But with Apple already having years of experience making these chips, i don't think there will be problems with them.
And that's why it'll never happen. Boot Camp is already a niche inside of the niche of Apple users. Microsoft will not be able to make a sizable profit off of this venture.
At worst, Apple may be able to bring in as much profit as a small-sized Windows laptop OEM... and that's assuming they'll outfit every single ARM Mac with a version of Windows on ARM.
So honestly, I think you're veering into wishing territories at this point. Realistically, I'd expect Apple and Microsoft to strike a deal on office apps, and Apple will still continue their monopoly on Mac OS. That's the current business model with iPad OS and iOS.
Beside all of that point, why would I try to virtualize Windows on ARM, only so that it'll run another x86 emulation layer on top? That just makes no sense. It makes more sense for Parallels and VMWare to go back and make an efficient x86 emulation layer to run straight Windows 10. And I think that's most likely what will happen in... say 2022. They may be able to introduce it in 2021, but I doubt they'll be able to ship the software that fast.
There will be no x86 virtualization by definition, but there could be emulation. If performance is sufficient, Apple (unlikely) or a third-party could provide an emulation solution that would allow you to run Windows in a window on the Mac Desktop. Unless the Apple Silicon has exceptional performance (general purpose performance, not specialized systems like the Neural Engine and such), the emulation experience will be slower than what we have today on Intel based Macs. If this route is taken, hopefully there is enough performance to allow Windows to run "good enough." There could also be some sort of WINE solution provided, but again since it would be emulation, it would not be as fast as current Macs.
Rich S.
I’m really hoping Apple silicon blows everything else out of the water such that this kind of emulation will work just good enough for me. This will be, honestly, the first time they’ve ever been able to fully jack up the clock speed (and use active cooling even) on their custom chips. I will live with my Intel MacBook Pro until that time.
Most people are not realising that it's not miracle silicon that would make everything instantly better and faster. There will be problems and incompatibility issues during the transition and after.
Nah nah.
From a pure software engineering point of view, having software specifically designed for your silicon is the best you can ever have. You control absolutely everything and this is the only way to really control every single aspect of the product. You control all aspect of the chip and its own development pace. You also control the video rendering, the drivers, the graphics API (metal), so you control the entire graphics stack which is the weakest point of every OS in history.
Yeah, Windows won’t work natively on it. It’s sad, yes. For software engineers, it’s even really bad because it removes the possibility to test on all three platforms on a single computer. At least you will have Linux ARM with Parallels which is fundamental for devs, a lot more than Windows. But for everything else, I think there are only benefits.
The problem won't be with software that is written for the Apple SOC, it will be with apps that are just ported to it. Sometimes it is not as simple as just recompiling an application for a new architecture/OS. There will be bugs that get missed or introduced, performance issues, and some applications that will just get lost or abandoned. Don't get me wrong, I don't mean performance issues because that I think the Apple SOC will be slow, but performance issues related to the fact that you will have to do things differently going from CISC to RISC. If the *best* path isn't chosen, it could run too slow, or even too fast. It will take developers time to learn the idiosyncrasies on how to get the best out of the Apple SOC.
Plus there is the simple fact that if an application developer just tries to take their mobile app to a full-size computer, it may not work as well. Smartphone/Tablet apps tend to be things that have simpler use-cases, and are not as "fully featured" as the bigger computer applications tend to be. Hopefully we don't just get a bunch of iPhone/iPad apps, but I expect that will be the initial case.
Rich S.
Why does it seem that everyone on the planet understands how to use the word "revision", except tech nerds on forums? "Revised" means "look at it again". In writing, no one ever calls a first draft the "first revision!" It's: 1. First Draft, 2. First revision, 3. 2nd revision, etc.
Yeah, I also avoid buying the FIRST VERSION of a new Mac model, and instead wait for the first revision.
Last edited: