That does seem like a key difference. I imagine there was a lot of outside pressure on Apple to make that change. With Arm it seems more the other direction.
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.
ARM based MacBook Pro discussion [merged]
- Thread starter Omega Mac
- Start date
-
- Tags
- arm intel macbook pro macos sierra
- Sort by reaction score
The first part is comparing orange juice to apple pie, and the latter is very unlikely to be true.
Meanwhile neither of those points actually solves the multitude of problems mentioned by people in this thread multiple times now.
My point n.9 resolves them all, you don't wanna use ARM? No worries, still can get a 2020 laptop that will last more than 5 years and by then everything will be solved or the paradigm will have already shifted to ARM.
About how unlikely to be true to have an ARM Mac executing x86 apps, we have already WindowsARM doing that, how come Apple wouldn't do it aswell? You guys just want to keep moaning about stuff that hasn't even happened yet, but it's the main trend everytime Apple changes something: cry and complain -> well, maybe it wasn't that bad -> forget about it. See any polemic Apple decision for that, end of the story
Sorry I literally stopped reading when I got to the ridiculous part I quoted.
So let's see what is the 9th commandment.
.... Ok great, let me know how that works out for you. Personally I use my Macs for more than watching YouTube and **** posting on Facebook; I have a business to run, and I can tell you there is a zero percent chance I won't need to run x86 virtual machines in 5 years or even 10. Literally zero.
A lot of people are (I am as well) excited about the Desktop ARM chips, based on how they perform on the mobile side, but there are quite a few things wrong with your post.
Benchmarks for the AX chips are great. but they are benchmarks. We don't know how these chips will perform against Intel/AMD. We'll find out soon enough.
x86 chips are designed to do multiple things well to very well. You can't compare them to NPU's or tensor cores that were designed to do ONE thing. They were specifically tuned for that task. Of course they will do better compared to x86
Catalast ONLY works one way. iOS/iPadOS to MacOS. Not the other way around. Go read Apple's website and you will see it specifically state that.
I know in the IT and software development fields, a lot of people use Macs for bootcamp and VM's. Losing that will only make them buy Windows 10 machines. They don't care about fanboyism. They only want to use the machine that gets their work done.
Yes, in the same way that because the iOS Simulator runs X86 code, you can rip iOS out of it, and run it on a Mac. In other words: no.
That explains all your nonsense and moanings here. A good point to ignore your answers since they don't bring anything constructive to the table.
We don't need to compare the same SoC, RAM, mobo, etc. In the end yes benchmarks are benchmarks, but they are just "perform X intensive task in Y time", so that's why they are used to extrapolate conclusions. End users only care about a machine allowing them you to do exactly the same task than the other machine in the same or less time, that's why I was putting the video edit + rendering (same quality both machines) example. So if the ARM "video editor you wish to use" in the iPad Pro/ARM Mac allows them to work having the same or even better performance, people won't care about which freaking architecture they are using except for the inital lack of software (once that's ironed, they won't care at all). Call it video editor, image editor, web browser, IDE or whatever you like, the point stands still. You don't need to compare apples to apples (architecture + software wise) because for 99% of users that won't matter at all as long as they get good performance doing their task.
Obviously that only works that way since Mac is already x86 and they don't need iPad Pro to be a Mac substitute executing x86 stuff. It's not stated anywhere nor anything prevents them to do the same translation they are doing to execute ARM instructions in an x86 the other way around (from x86 to ARM) once they use ARM in Macs. Apple ain't stupid and will ease the transition from an architecture to another the same way they did more than a decade ago.
Last edited:
Come back to me when you're not comparing a UI porting framework with an inter-arch code translation layer, and we'll talk about 'nonsense'.
You appear to completely misunderstand what Catalyst is. Catalyst has absolutely nothing to do with code execution or architecture translation. Catalyst does not execute ARM instructions on x86 or the reverse.
You're right, Catalyst has nothing to do with the architecture transition. But one has to assume Apple will put a checkbox in Xcode, like they did when they transitioned from PowerPC to Intel.
Or better yet, when the app is downloaded from the App Store, the correct binary is automatically installed for the architecture the user is on. So no "fat binaries."
I'll give you a bit more detail than 'no'. An iPad has very different hardware from a mac. You don't only need the OS to be working, it also needs to support the capabilities of the machine it's running on. I've ported operating systems for a living and the first additional thing to do after the OS code seems to work, is to get a led blinking. When you say 'install macos on iPad', I assume you want more functionality than that. How does the iPad's storage work? Screen? Audio? Touch? Bluetooth? Etc.
That might actually be another reason why Apple started the whole T2 thing. It abstracts certain devices. Just make sure you can work with the T2 and several complex things will work at once.
MS has worked really hard to unify the Office codebase. For over 20 years, they had separate repositories for different platforms though there was some shared logic. In 2018, they announced that they had finally unified these codebases, which has huge benefits. Of course, there is still some platform-specific code in the codebase, mainly for UI (although I read they're moving to React Native nowadays).
This unified codebase already compiles to Windows, macOS, Android and iOS. You won't believe it, but even Office 2019 has two versions for Windows on Intel/AMD, it has a 32-bit and 64-bit version. This way, they can still support really old plugins and extensions that companies use. They compile for so many platforms/architectures that it will be quite easy to release a native ARM build for Windows/macOS. I'm guessing they're working on backward compatibility for those old plugins, running Office natively on ARM and simulating x86 for the plugins.
Last edited:
Thanks for the explanation. I was just wondering because if Apple does switch Mac to ARM then it will use the similar processors.. like A12 and A12X, so technically it should be possible to use iOS and MacOS on iPad and Mac and vice versa.. may be not out of the box, but Apple can support both the Mac and iPad hardware if they choose to. Is it different from Windows running on Intel Mac?
Yes, because Windows is a pretty unique operating system because it has drivers for just about any hardware. So it will run on whatever you throw at it. Linux is comparable, but with any other OS, you'll have trouble getting all your hardware to work. Even if the os itself runs fine. That's why people struggle sometimes with wifi and bluetooth or wake-from-sleep when putting macos on a non-Apple computer. Talking about ARM hardware, the field is so un-mainstream that any OS needs work on drivers before it can use the screen, audio, ... In principle, it's the same as running Windows on a mac, but in practice it's very different because no ARM OS will 'just run'.
Thanks! I thought it's an interesting idea if Apple can release drivers for both the iOS and MacOS on iPad and Mac, and it could be relatively easier than Windows or Linux running on multiple hardware since Apple will control both the OS and hardware. It would be interesting running MacOS on a touch-based iPad.
Last edited:
Yeah, they have control over hardware and software. But you're bumping into the difference in concept. Running iOS on a mac means controlling a touch interface with a mouse. It's what Apple is trying with Catalyst. And vice versa, well, you can attach a mouse to the iPad but when I use sidecar (also macos on an iPad screen), it always seems a shame of the incredible touch screen.
It's a bit like saying: hey, I have this truck engine, let's put it into my sedan and go to the shopping mall with it. Or hey, I have this sedan engine, let's put it into my truck and get 5000kg of sand delivered.
I think people, myself included, mostly bring up Catalyst not to say Catalyst specifically will be what Apple will use to ease the ARM transition (which i don't think is impossible, it's not like they can't add more functionality to it once they actually announce the ARM move or introduce some other similar program). I think it's more so just to point out that based on the history with Rosetta and Catalyst that Apple has the prowess and forethought to do something like that and it's almost certain they will have some kind of similar solution for this ARM transition. You're taking this a bit too literally i think, no one's actually saying right now as is, that Catalyst is currently capable of handling the x86 to ARM move, it's just cited as obvious precedence that they likely have a plan for the move.
The person I was replying to held the mistaken belief that catalyst is a technology that allows ARM binaries to be run on x86 hardware. That was literally in their post that I quoted. They continued with the theory that since catalyst allows ARM binaries to run on x86 then obviously it would be just as easy for catalyst to also allow x86 code to run on ARM systems.
This sort of confusion -- while understandable -- does nothing but muddy the discussion and leads to unjustifiable optimism for what Apple may or may not be able to accomplish technically to ease an architecture transition like we are discussing.
Last edited:
Microsoft already has x86 simulation on ARM.
Some details could be found here: https://channel9.msdn.com/Events/Build/2017/P4171
If Apple has this kind of technology (or similar, better) then transition to ARM in my opinion wont be that bad. We will still be able to run old 'good' x86 applications until developers update them to ARM based instructions (if needed).
I really want to see MacBook Pro in body similar MacBook Air. Could you guys imagine MacBook Air in size of MacBook Pro 15/16? Would be fantastic. I think that ARM processor based machines could do that because these chips are colder (correct my if I'm wrong).
Some details could be found here: https://channel9.msdn.com/Events/Build/2017/P4171
If Apple has this kind of technology (or similar, better) then transition to ARM in my opinion wont be that bad. We will still be able to run old 'good' x86 applications until developers update them to ARM based instructions (if needed).
I really want to see MacBook Pro in body similar MacBook Air. Could you guys imagine MacBook Air in size of MacBook Pro 15/16? Would be fantastic. I think that ARM processor based machines could do that because these chips are colder (correct my if I'm wrong).
I'm going to nitpick a bit here. Linux does have a wider variety of drivers than macOS, but where it lags in driver support is new/bleeding edge hardware. For example, getting Linux to boot on bare metal with my 2019 MBP was a challenge, and some things still don't work (most notably WiFi and Thunderbolt.)
Many times open source developers have to reverse engineer the Windows driver in a VM, re-write a driver from scratch without any sort of vendor documentation. What you usually end up with is very basic functionality. An extreme example of this is the Sound Blaster Z, a sound card released in 2012. It wasn't fully supported in Linux until 2019 thanks to someone using their personal time to reverse engineer the Creative Labs Windows driver and code one for Linux.
On the other hand, old drivers tend to hang around for a long time in Linux. I've got an M-Audio Revolution 7.1 (a PowerPC-era PCI sound card) that is still fully supported in Linux. That is, if you can find a PCI slot to shove it into...
Not read the whole thread but I think Apple should do this as a 2in1 where you can fold it around into a tablet like they do on Windows. In Tablet mode it's actually running iPad OS but with integration back to your Documents. Flip it back to tablet mode and it's running Mac OS. With the ability for developers to bring iOS applications to Mac OS last year along with running on an ARM processor... You'd be able to have the same applications transition between the modes but with different input methods. It feels like it would be really quite easy for Apple to do and kind of a killer type feature. I mean I call it iPad OS and then switching to Mac OS but obviously they'd be combined rather than it running 2 OS's at the same time but the experience would be like that.
Otherwise it's a slim laptop with nice battery life but that's about it. People that buy these cheaper laptops for the most part won't care if it's ARM / Intel or whatever. Well unless Apple can sell it at a really low price but that's un-Apple like normally. However can you imagine the buzz they might generate if they also go for a new 2in1 form factor with iPad and Mac OS in one plus Apple Pencil support, inbuilt 5G, etc? That I think would be huge.
Otherwise it's a slim laptop with nice battery life but that's about it. People that buy these cheaper laptops for the most part won't care if it's ARM / Intel or whatever. Well unless Apple can sell it at a really low price but that's un-Apple like normally. However can you imagine the buzz they might generate if they also go for a new 2in1 form factor with iPad and Mac OS in one plus Apple Pencil support, inbuilt 5G, etc? That I think would be huge.
I guess Apple will use the new iPad with Magic Keyboard as a developer system for the ARM Macs, providing a version of MacOS for developers to load onto the iPAD. I do not think Apple will go for a 2in1 with dual personalities.
I think we have seen the direction Apple is taking in the release of SwiftUI. It looks like Apple want to merge AppKit from the Mac and UIKit from iOS to something common, SwiftUI which can be used to develop the user interface for two different paradigms (mouse vs. touch) with a minimum of differences in the code.
Having created a couple of apps in SwiftUI (one was ported from Windows .NET Forms) it looks like a very interesting approach but still fairly buggy and immature.
So if Apple is serious with starting the transitions to ARM (and merging the SDK) a lot of work remains, it will be very interesting to see WWDC this year, we should get some confirmation about the strategy and timeline.
I think we have seen the direction Apple is taking in the release of SwiftUI. It looks like Apple want to merge AppKit from the Mac and UIKit from iOS to something common, SwiftUI which can be used to develop the user interface for two different paradigms (mouse vs. touch) with a minimum of differences in the code.
Having created a couple of apps in SwiftUI (one was ported from Windows .NET Forms) it looks like a very interesting approach but still fairly buggy and immature.
So if Apple is serious with starting the transitions to ARM (and merging the SDK) a lot of work remains, it will be very interesting to see WWDC this year, we should get some confirmation about the strategy and timeline.
You’ve brought to life what I think the 12” ARM MacBook could be for consumers - Apple would call it the iBook, bringing back that former name.
Owning a 15” 2-in-1 that I have only have unfortunately used in tablet mode only 2-3 times (way too heavy for that), I would love to see something like this from Apple. The 2-in-1 concept was what got me to buy it in the first place. If only it wasn’t 1)buggy in tablet mode, 2) too heavy to use freely as a tablet, and 3) well, it wasn’t an iOS tablet. Great laptop though with all the right specs. For me, I wouldn't need the 2-in-1 larger than 12-13” and of course, slim, light, and cool.
Apple could even offer a regular ARM notebook, then a 2-in-1 option ala Microsoft. iBook and iBook Air. It could be what’s coming for the Air since the market for the Air would be pretty much the same for this device - people who need more than iOS, but don’t needed sustained intensive performance. Interesting to see what Apple has in the works in the general consumer space with ARM.
Last edited: