More integration, more timeline and feature control, reduced costs leading to higher margins...among many reasons
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 Expected to Announce Arm-Based Mac Plans at WWDC, Transition Away From Intel to Begin in 2021
- Thread starter MacRumors
- Start date
- Sort by reaction score
More integration, more timeline and feature control, reduced costs leading to higher margins...among many reasons
Kind of a waste if that's the only purpose for buying an ARM. Plus x86 Macs can already run iOS / iPadOS apps. I assume there has to be a much grander vision for transitioning to ARM
Those of us who enjoy (or need) dual booting for Windows...believe me when I tell you that Apple doesn’t care about that feature and our concerns around it..
one.
single.
bit.
one.
single.
bit.
Yeah but then we get to the same old intel can’t deliver chips on time and in volume problem.
I didn't say that was the only reason.
And Macs cannot already run iOS/iPadOS apps (at least not ones from the app store, unless recompiled, and recompiling doesn't always work - I know, I've been trying to simultaneously support mac with one of my apps and every new release of Xcode breaks things.
Own the whole stack make better products with less reliance on other vendors delivering on time. Make more money.
The most interesting part of that is when he mentions that OS X had been designed to be multi-platform starting in 2000. That must be the case all the way to 2020 because macOS, iOS, iPadOS, etc. all share the same base Unix operating system.
Nothing arbitrary about it.
I have not seen a tablet yet that could fully replace a computer.
We've been hearing this for a decade. Yet ARM's market share in the segments I mentioned continues to be negligible.
Call it what you like. But the huge amount of software that would have to be ported, as well as the lack of industry standard platforms for ARM CPUs are real problems, and the benefits of moving to ARM are questionable.
With the transition of macOS to ARM, I'm curious if Apple rebrands macOS to macOS 11
If Tiger Lake ends up catching up to and/or surpassing Ryzen in any way, as initial benchmarks are suggesting, the AMD crowd on this forum is going to collectively s*** themselves. Which would be fine, as their blather gets old. Not saying I like Intel, they're fat, bloated, arrogant and rely on bully tactics to stay on top, but cracks are starting to show with AMD. Small cracks, but I can already start to see them. Which is a shame if I am right. I hope that I am not.
When you design processors, most of the cost goes into the design. The marginal cost for manufacturing each additional processor is rather small. So when you only sell processors (e.g. Intel), you have to recoup your IP investment in every sale.
But if you're Apple, you could easily, say, stack 10 or 20 processors in a Mac Pro, with only a marginal increase in cost. You can't do that when you have to buy every Intel processor. I wonder if the highest end Macs will be the biggest beneficiaries.
Also, most software is linear, so getting the most advantage out of multiple cores and processors requires software redesign/planning. But when you control the compiler, the language, the OS, and the supporting hardware, there is a lot more potential for optimizing parallelism. This is going to be an interesting experiment.
But if you're Apple, you could easily, say, stack 10 or 20 processors in a Mac Pro, with only a marginal increase in cost. You can't do that when you have to buy every Intel processor. I wonder if the highest end Macs will be the biggest beneficiaries.
Also, most software is linear, so getting the most advantage out of multiple cores and processors requires software redesign/planning. But when you control the compiler, the language, the OS, and the supporting hardware, there is a lot more potential for optimizing parallelism. This is going to be an interesting experiment.
No question, they're righting the ship, though it may still be a little bit before we really see significant gains
They care enough to develop Windows drivers for all their hardware and release them when new Macs come out. That's not a minor undertaking at all.
You’re proving the point.
They would love to drop all of that effort in a hot minute.
When Microsoft released the (newest) version of ARM-Windows, every app in their Windows Store ran automatically, without recompilation, and even without emulation (!). It turns out, the binaries in the Windows Store are not Intel binaries, but a pseudo-binary generated for a virtual processor (this is standard output of the Microsoft tool chain for many years). So running on Intel or ARM didn't really matter -- it is compiled on the end-user's machine before running.
I assume Apple will take a similar approach. Swift was designed to compile to an IL-language. And/or they will use fat binaries.
Whatever approach they take, they will probably force it on developers for the next OS-release in the fall (like they did with 64-bit apps last year). By the time ARM Macs are released, most of the software in the App Store will probably just run, one way or another.
This segment is really small now.
And technically there's no problem for a different CPU arch to shine in this segment anyway. It was like that 20 years ago when PPC/DEC Alpha/Intel was competitors in PC segment.
I don't know what to think. I run Windows 10 via parallels for some win only apps, radio scanner apps, Apple iigs file utility called ciderpress (Ironically not available on macOS) I could get a cheap laptop or even nice/compute stick should arm no longer allow me to run windows programs. It would be annoying to have the extra expense however.
They going through the mire (relatively speaking for Intel) at the moment. Their progress in the last half a decade or so has been nothing special.
I have high confidence Keller will right the Intel ship.
But the Apple Intel alliance will be toast by then.
Azrael.
You can stick with your current Mac. And if Bootcamp still exist ARM Mac can run ARM version of windows that is back compatible to x86 32bit software.
Well I certainly hope, and think, Apple will do a better job with macOS on ARM than the lame attempt by Microsoft - so far - at doing the same with Windows. (Surface Pro X. I bought one back in December and sent it back to MS in January after using it for about a month. Many, many applications are incompatible with ARM. And developers don't seem to be in any rush to fix that.)
As someone who has ported game engines to x64 and ARM devices (iOS, tvOS) I can say this is not usually true.
Very old game engines who pulled tricks and stunts with code that only work in whatever bit-level they're in (i.e., this trick works with 32-bit but you compile for 64-bit and it's broken) have some work ahead of them, sure, but the vast majority of game engine coding for years now has been of the variety where you can just recompile for x64 and you're golden.
The real issue is of support - namely, a lot of old games are way past their support window. They don't get reworked because the developer/publisher is done with that game. It's made its money. It would be nice if they did a recompile for newer systems but then they'd have to support it, etc.
So ironically then your example of FF14 is actually the opposite - they're still actively supporting that game (being an active MMO and all) so it's actually very likely they'll port it to ARM. World of Warcraft was released in the PPC era and was ported to Intel Macs no problem. Expect the same here. Squeenix has like dozens of games on the Switch, also an ARM device, they've got some experience here.
I’m anticipating both more power and better efficiency, along with everything that comes with that.
It really comes down to software. If I can have the same native programs, or in some cases a suitable equivalent, I’m all for it. Plus, porting iOS apps could become a huge upside.
It really comes down to software. If I can have the same native programs, or in some cases a suitable equivalent, I’m all for it. Plus, porting iOS apps could become a huge upside.
iPhone versus Android has shown that most people don’t care about the nitty gritty CPU specs in their devices. Does it work? Does it work well? Ok cool. I’m a pretty big dork and I do not know the speed of my iPhone Xs CPU or how many cores it’s rocking without looking it up. Works pretty great though.