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.
So M1 Max with 32 Core GPU matches RTX 3080 Mobile.
- Thread starter iBug2
- Start date
- Sort by reaction score
That would also require MS to help. I thought they might be up for that but so far MS’s response has been lackluster for Windows ARM in general - as in no Windows Arm beyond OEM. For Apple CPU and GPU, MS would have to help write the DX drivers and make the necessary kernel changes. Apple can’t just write them. And then there’s the matter of support. You’re talking a massive amount of work. I still think it’s possible but so far MS’s attitude make it increasingly unlikely.
Last edited:
I don’t buy that; DX is well documented and there are at least a couple open source “translation” libraries to learn from. You’re right that it’d be a lot of work and even more support headaches, though. It’d be a tough elevator pitch for sure.
Maybe but I edited the quotes sentence with other issues: namely the changes to Windows kernel to support AS. MS would have to make those. Right now they won’t even sell licenses for standard ARM hardware - which is kind of a necessary first step. It’s why VMWare has said they won’t support Windows even through virtualization.
^ Yeah, these are examples of poorly performing games. We can find examples everywhere, including on "pro Mac" channels.
I'm not sure why Fortnite does not perform better. UE4 does not seem well opitmized.
Alien Isolation is expected to perform poorly since it's an openGL game (why use such old game for a performance test?). Apple's has been neglecting openGL for years, and it uses a translation layer on Apple Silicon. Sleeping Dogs perform even worse, there's obviously a bug somewhere.
I'm not sure why Fortnite does not perform better. UE4 does not seem well opitmized.
Alien Isolation is expected to perform poorly since it's an openGL game (why use such old game for a performance test?). Apple's has been neglecting openGL for years, and it uses a translation layer on Apple Silicon. Sleeping Dogs perform even worse, there's obviously a bug somewhere.
Apple’s SoCs support all of the ARMv8 instructions, so MS doesn’t *have* to do anything. ARM Windows would benefit from Apple’s extensions, but it’s not like it would run worse than on whatever vanilla ARM processor. Apple (or a few enterprising OSS fellas, say the Asahi Linux crew plus some Windows fans) could write drivers and it would probably be good.
VMWare is feeling things out before committing. I don’t think your read is right, though your take on official support makes a lot of sense to me for the time being.
There is no doubt that Apple could implement the drivers necessary to run Windows natively. There is even less doubt that they won’t do that. It’s a lot of work with zero payout for Apple. They don’t even want to support Vulkan in macOS.
Windows would need kernel modifications just to boot on Apple Silicon. A lot of low level stuff is just different: interrupts, between CPU communication… alternative would be Apple writing a low level hypervisor that „pretends“ that this is a regular ARM PC, aka Bootcamp 2.0, but again, a lot of tricky work with zero payout.
Craig Federighi even stated that it would be up to Microsoft to support the M1 hardware but that it was possible because the boot loader is open, just like the Asahi Linux project.
Especially since at the moment MS doesn’t even want to sell Windows on Arm stand-alone licenses so the point of Apple doing all that would be … nothing.
This is all why I stand by the notion that this would have to be a deal between the two companies and while I can imagine reasons for each to do so, at the moment I don’t see it happening. MS attitude is that WoA is OEM-only and Apple’s is that we’ll support boot loading but after that you’re on your own.
Aye but there is a little disagreement over whether he meant through virtualization or bare metal. The wording of the question and answer in that interview is somewhat confused on that point.
Federighi's statement was taken widely out of context. He was referring to virtualization. Apple does open native boot to third-party implementations, but they do not release any documentation which makes it worthless for production use. Asahi Linux can spend time reverse-engineering Apple hardware and patching the kernel because they are passionate idealists. Microsoft will do no such thing.
Virtualization is the simplest and most effective path forward. Frankly, it would help if Apple introduced some "compatibility support" features for Metal that would make it easier to map DX and VK idiosyncrasies to Metal.
Sure I agree with that, but even virtualization is kind of hamstrung by MS insistence that WoA is not for sale. ? Obviously it still works and you can do it, but it makes it feel … not so great.
To be fair taken out of context with reason. If I remember right, the question was pretty geared towards boot camp even if the answer was not so I understand why people think they were talking bare metal support (though as you say that is a little unreasonable to expect of MS without direct support from Apple.)
Well, you know, if that's MS official position, then taking about the possibility of native boot makes even less sense. Yeah, it's a ... suboptimal situation.
That's from Gruber's interview, right? They discuss hypervisors just afterwards where Federighi says out direct that they are not talking about native boot.
Oh absolutely that was part of my point . while I could’ve imagined a scenario where both agreed this was a good idea, neither seem interested. So it’s not going to happen for the foreseeable future.
The one got people excited was the Arstechnica interview who botched it up:
“We are giddy”—interviewing Apple about its Mac silicon revolution
Craig Federighi, Johny Srouji, and Greg Joswiak tell us the Apple Silicon story.
arstechnica.com
And the macrumors copied their version:
Craig Federighi: Native Windows on M1 Macs is 'Really up to Microsoft'
Following the release of the M1 Macs Apple executives have been doing interviews with a range of publications, and today, Ars Technica published...
www.macrumors.com
That’s why people still think he was discussing bare metal. Because it was reported that way in a lot of outfits.
Last edited:
The one got people excited was the Arstechnica interview who botched it up:
“We are giddy”—interviewing Apple about its Mac silicon revolution
Craig Federighi, Johny Srouji, and Greg Joswiak tell us the Apple Silicon story.
And the macrumors copied their version:
Craig Federighi: Native Windows on M1 Macs is 'Really up to Microsoft'
Following the release of the M1 Macs Apple executives have been doing interviews with a range of publications, and today, Ars Technica published...
That’s why people still think he was discussing bare metal. Because it was reported that way in a lot of outfits.
Ah, I see. The article did say "as to native support". Given Apple's stance on the matter and the fact that their hardware is proprietary, and considering what they said in other interviews, I believe that this article misquoted Federighi.
Yup but that’s why I’m pretty sympathetic to people who get that wrong. Heck reading that at first even I thought he meant boot camp 2.0 and DirectX drivers and everything. It took several rereadings as well thinking about it to realize the framing was wrong and that CF was probably talking about virtualization.
Edit: in fact if I remember right it was probably reading posts by you (or someone else on the forum) soon after launch that caused me to go back and reread the article more carefully and realize that he might not have said what I thought.
Last edited:
Since you can't play them unplugged, money wise it's better to buy an Xbox for your a MS kind of guy.
It's much much easier to get a gaming laptop vs. a Xbox Series X. (Sorry, the ubiquitous Series S is really last gen specs masquerading as current gen hardware)
This is a bit misleading.
Yes, Microsoft could support Windows 11 on ARM for the M1 chips. The issue is that would require work on both ends to get it functioning. Of biggest note, the M1 chips do not support UEFI and ACPI, both of which are required for Windows 11.
Is it possible that Microsoft would remove those, but it's also possible that every more game dev studios code directly for Metal and the M1 chips. It remains to be seen what will happen, and I'm not exactly holding my breath. BOTH sides would have to make concessions, and I'm honestly not sure it'll happen. It's much more complicated than Craig would have you believe.
Most of the cost burden would be with Microsoft. In the 15 years of Intel Macs has there been substantial demand for Windows on the Mac?
Intel Macs was a relatively small R&D cost for Microsoft as it uses the same Intel chips as any other PC. This was the same reason why hackintosh became a thing.
With Macs moving nearing its transition to Apple Silicon it is up to MS to make some adjustments to Windows 11 on ARM to make it function on Apple chips.
Last edited: