First Arm-Based Macs to Be 13-Inch MacBook Pro and Redesigned iMac, Launches Coming in Late 2020 or Early 2021

[Bujo Botaro]

The transition between power pc to Intel CANNOT be comp

the PPC-to-Intel transition happened inside of 18 months. I don't expect any different this time. If Apple is transitioning to ARM it's because they believe they can go all the way. Otherwise no point doing it.
[automerge]1592759016[/automerge]

Don't be surprised if Bootcamp is dead.

The transition from PPC to Intel was fast only because there was a very efficient translation layer created between the two architectures. However, this is not the case for Intel -> ARM. This is evident from all the ARM Windows laptops and tablets we've seen lately. Really powerful chips, but the real-world performance in x86 emulation is lackluster at best. I honestly think that Apple won't put the ARM chips in their pro-line-up anytime soon, maybe they'll have two versions or something.

 

[smulji]

Apple being a subscription and services company

They are not a subscription and services company. Three-quarters of their revenue still comes from selling hardware.
[automerge]1592773760[/automerge]

There haven't been any leaks around chips needed for the desktop Macs

There weren't any leaks either during the PPC to Intel transition. When SJ introduced it he basically that macOS was leading double life for five years inside of Apple's labs running on Intel chips. No one knew a thing.
[automerge]1592773898[/automerge]

However, this is not the case for Intel

That's an assumption. You can't use what MS is doing with Windows on ARM and assume that Apple is going to use the same playbook. For one, MS is not using an in-house custom designed ARM processor. It's predominantly designed by Qualcomm with some help by MS but it isn't anywhere near the same thing that Apple is doing. Qualcomm is essentially the Intel of the mobile device world.

 

[rwxx]

They are not a subscription and services company. Three-quarters of their revenue still comes from selling hardware.

I'd argue that almost all of their revenue still comes from selling hardware. The bulk of their "services" revenue is Apple Care and App Store commissions. No hardware sale, no "services" revenue.

 

[cmaier]

I'm not convinced of that. Obviously I have no data to back this up, but my feeling is that the Mac user base isn't really growing. I think Apple's Mac sales are mostly to existing Mac users. Mac unit sales have been flat for years. A switch to ARM isn't going to lure new Mac users...unless Apple surprises us all with a cheap entry-level Chromebook-type product (doubtful). Will some of the folks who came to the Mac after the Intel transition now leave the platform? I think there's a very good chance that will happen and I don't see new customers lining up to take their place.

The new functionality and interoperability with iOS, even if it lures a tiny percentage of iOS users, would multiply mac sales by many times.

 

[jimbobb24]

Guess I have to buy the last intel iMac I can this year. Wasn’t planning to upgrade but I want to stay with Intel as long as possible.

 

[rwxx]

There weren't any leaks either during the PPC to Intel transition. When SJ introduced it he basically that macOS was leading double life for five years inside of Apple's labs running on Intel chips. No one knew a thing.

"Knew"? Maybe not. But suspected? Most definitely. The PPC architecture was lagging and anyone who knew anything about NeXT knew that NEXTSTEP ran on a variety of hardware platforms. The switch to Intel wasn't a surprise.

 

[petvas]

Apple will not abandon all professionals out there, especially after the last three years and the investments they made to the Mac platform. Going only ARM without a good (and long) transition plan, would be suicide. I don't believe this report at all. I am confident though that Apple will introduce an ARM entry level Mac at some point and that it will transition completely to ARM, but first the hardware (ARM CPUs) and software (macOS & 3rd party apps), must evolve accordingly. This will not happen in 12 months. I believe that it will take 3-5 years for this transition, especially when talking about Pro software.

 

[Zen_Arcade]

Kuo says performance will be 50% to 100% higher than the Intel equivalent.

Would be nice if I finally no longer had to explain to everyone on here over and over again that Arm (and RISC) have inherent advantages - it takes up half my day

And we love you for it!

👍

 

[Dozer_Zaibatsu]

Literally just bought the New 2020 Macbook Pro..........
mistake?

Keep it for three years, during which it will be singing in its prime. Just in time to upgrade to whatever third-gen of ARM-based laptops, which will inevitably be touted as improving the unforeseen adoption problems with the first gens.

 

[rwxx]

The new functionality and interoperability with iOS, even if it lures a tiny percentage of iOS users, would multiply mac sales by many times.

What new functionality and interoperability? Are you suggesting that the next version of macOS will offer ARM-only features? That would be a huge surprise. I suspect the next version of macOS for ARM and macOS for Intel will be identical in features, meaning there won't be any difference in functionality and interoperability. How will an ARM Mac lure in an iOS user who still hasn't made the switch? What is going to finally push them over the edge and convince them to buy a Mac if they haven't done so already? Your argument sounds like wishful thinking to me. Apple has been unable to grow the Mac user base for years. I don't see how a switch to ARM makes the platform more compelling, but it's pretty evident how the switch makes it less compelling for certain users.

 

[cmaier]

Apple will not abandon all professionals out there, especially after the last three years and the investments they made to the Mac platform. Going only ARM without a good (and long) transition plan, would be suicide. I don't believe this report at all. I am confident though that Apple will introduce an ARM entry level Mac at some point and that it will transition completely to ARM, but first the hardware (ARM CPUs) and software (macOS & 3rd party apps), must evolve accordingly. This will not happen in 12 months. I believe that it will take 3-5 years for this transition, especially when talking about Pro software.

The millions of macs in coffeeshops and planes and campuses running office applications (which are the vast majority of them) will be replaced by 2x the speed ARM macs and the users won’t care. The rest of us will just have to live it. It will be a fast and complete transition, and in the end it will be fine.

 

[Bujo Botaro]

That's an assumption. You can't use what MS is doing with Windows on ARM and assume that Apple is going to use the same playbook. For one, MS is not using an in-house custom designed ARM processor. It's predominantly designed by Qualcomm with some help by MS but it isn't anywhere near the same thing that Apple is doing. Qualcomm is essentially the Intel of the mobile device world.

Yes, it's an assumption, but it is also an educated guess backed by recent developments in technology. I highly doubt Apple has a hidden trick up their sleeve for a large enough performance boost to cover for translation inefficiencies. Yes, Apple has an in-house custom-designed ARM processor. Yes, in terms of performance it rivals Intel's laptop counterparts. What then? Unless they come up with the largest breakthrough in the software of the decade, they still cannot overcome the translation inefficiencies. I would say the switch is premature at the moment, and it would be better to wait until software companies start to write for ARM. Meanwhile, perhaps a switch to AMD might be better for the consumers.

 

[cmaier]

What new functionality and interoperability? Are you suggesting that the next version of macOS will offer ARM-only features?

Not necessarily the next, but maybe. There are already T1/T2-specific OS features. Touchbar-mac specific features. All sorts of features that only work on certain macs. And they can build AI stuff on their chips that Intel doesn’t have (including both generating ML models), etc. The whole point of this is they can control their future and they can differentiate. They do that with iPhones - build things into the silicon to give them an advantage over everyone else who just uses Qualcomm - so we know that’s how they think.

 

[ahmed.hassa]

Is it likely then that Bootcamp will be discontinued?

 

[Victor Mortimer]

I really want a redesigned high-end iMac, but it looks like I should wait a year or so.

No, you should get it now, otherwise you're risking being stuck with ARM garbage.

 

[cmaier]

Yes, it's an assumption, but it is also an educated guess backed by recent developments in technology. I highly doubt Apple has a hidden trick up their sleeve for a large enough performance boost to cover for translation inefficiencies. Yes, Apple has an in-house custom-designed ARM processor. Yes, in terms of performance it rivals Intel's laptop counterparts. What then? Unless they come up with the largest breakthrough in the software of the decade, they still cannot overcome the translation inefficiencies. I would say the switch is premature at the moment, and it would be better to wait until software companies start to write for ARM. Meanwhile, perhaps a switch to AMD might be better for the consumers.

Kuo says the CPU is up to double the speed of Intel. Throw in some hooks for x86 emulation (in another thread I posted what I, as a CPU designer, might do), and you’re looking at running x86 in emulation at around ¾ speed. Faster if you are running MacOS x86 applications (because nothing in the SDKs needs to be emulated - it can just shim to the native versions).

Remains to be seen, of course, but it doesn’t require technical feats of impossibility if the speed is really what Kuo says.
[automerge]1592775123[/automerge]

Is it likely then that Bootcamp will be discontinued?

It is certain.

 

[MikeZTM]

Is it likely then that Bootcamp will be discontinued?

That is not related to ARM transition.
Apple could discontinue the Intel Bootcamp at anytime. And Apple can provide ARM drivers to enable ARM Bootcamp if they want.

It's a pure business decision rather than a technical one. Though I bet they will not support Bootcamp during first year.

 

[Dozer_Zaibatsu]

I'm actually surprised how many people on here are cautiously optimistic considering the history on MacRumors of the reactions to the initial iPod and the announcement of the Intel transition.

From a technical standpoint, I am genuinely interested in what this can be. We're talking about RISC chipsets with multiple little cores. We've been seeing zig-zag in processor increments, and this is going long on the "zag." It is yet to be proven what can really be done with recent advancements in ARM unleashed. Apple IS, if anything else, a company that makes consumer devices. They're not out to play games with their main customer base. (Though sometimes it may feel that way.) I am assuming that are confident that they can give the users a better Mac experience, and I am intrigued.

From an IT guy perspective, I am a little bit grim that Apple is intent on keep their machines quite locked down. You can't replace parts on them as it is, (not really), and they want to keep your data in the cloud and will lock it out of access to you on your own machine. It feels like Apple is going with the model of leasing their customers devices rather than actually selling them. The user experience may still improve, but it is a wound in the side of open source.

Your move, PC makers.

 

[rp2011]

I don't see the relevance of "running the entire ecosystem natively". It's not like there are iOS apps running somewhere on the iCloud server farms. Most of the code of iCloud, Siri, etc. probably isn't even Swift. And even if it were, its CPU arch wouldn't matter much.

If you own the source code, by and large, arch changes are just a recompile away these days.

Well, as Linus Torvalds said, there is a great benefit in deploying your software in the environment that it was created in, which is why he thinks X86 took over the server space in the first place.
With more and more devices running on ARM, this may change.

“If you develop on x86, then you're going to want to deploy on x86, because you'll be able to run what you test "at home" (and by "at home" I don't mean literally in your home, but in your work environment).

Which means that you'll happily pay a bit more for x86 cloud hosting, simply because it matches what you can test on your own local setup, and the errors you get will translate better.
Actual hardware for developers is hugely important”

 

[dlewis23]

I'd say most devs don't require windows... I think most require a local stack that's closely resembles what they run in the cloud. To me, that means Docker/Kubernetes and any container that runs in that ecosystem... and VIM

That really depends on what your developing. Since Windows is the most popular OS you're going to actually want to test what ever you're building there at some point on a actual Windows install.

Pretty sure Microsoft has been compiling Windows 10 for ARM for awhile now.. Both Apple and Windows have an interest in ditching x86.. it's just a matter of time. So VMs running Windows 10 for ARM will be a thing soon after Apples ARM transition. I wouldn't be surprised if Apple personally asked/worked with Microsoft to help with ARM research and Windows 10 ARM.

And Windows 10 on ARM is crap, there is little to no use of it and little to no apps that actual work on it. x86 emulation is garbage and show. As it is right now it's useless.

 

[petvas]

The millions of macs in coffeeshops and planes and campuses running office applications (which are the vast majority of them) will be replaced by 2x the speed ARM macs and the users won’t care. The rest of us will just have to live it. It will be a fast and complete transition, and in the end it will be fine.

Microsoft would have to bring Office with native support for ARM processors, otherwise I (and many others) would not get an ARM based Mac. I am sure that they will do it, but it will take them some time..
Running software in emulation is not nice (but better than nothing). I think that tomorrow we will get the first glimpse of the future. Of course at the end everything should be fine, but this will not happen as quickly as you might think. For consumers yes, but for pros definitely not.

 

[3characterslong]

Yeah, the screen size thing is fishy to me too. Why would they keep the dated bezel design?
[automerge]1592757622[/automerge]


First gen iPhone was pretty great. Airpods too. And iPad.

Revisionist history, the first iPhone, while revolutionary, was NOT a great product and neither was the iPad. The first MacBook Air was a dog, the first (and only) redesigned Mac Pro 2013 was a completely waste of components...... I could go on, if you'd like.......

 

[Jyby]

Science Rules!

 

[cmaier]

Microsoft would have to bring Office with native support for ARM processors, otherwise I (and many others) would not get an ARM based Mac. I am sure that they will do it, but it will take them some time..
Running software in emulation is not nice (but better than nothing). I think that tomorrow we will get the first glimpse of the future. Of course at the end everything should be fine, but this will not happen as quickly as you might think. For consumers yes, but for pros definitely not.

When I first got my intel Mac i had to run office in emulation (or boot over to windows). It was ok.

My actual fear is that apple doesn’t even plan to provide a rosetta-like emulation of x86 apps. That would be painful.
[automerge]1592775724[/automerge]

Revisionist history, the first iPhone, while revolutionary, was NOT a great product and neither was the iPad. The first MacBook Air was a dog, the first (and only) redesigned Mac Pro 2013 was a completely waste of components...... I could go on, if you'd like.......

I had the first iphone and the first ipad. The first iPhone, in particular, was amazing. Coming from a treo it was revolutionary. I have no idea what you are even talking about.

 

[DeepIn2U]

The 13 inch Macbook Pro is very much a consumer laptop £1300.

As is the 24 inch (muted) iMac. £1250.

That covers £££ of the Mac sales.

So A14x goes into them 1st.

Yes. The intel cpu did debut for devs on the Mac Tower as a 'beta' piece of kit to get porting.

I wonder if they'll use the Mac Pro this time....with Mac ARM chips in them. (I doubt it. It will likely be another machine...perhaps a preview iMac 24 incher 'Beta' to get 'portin' their apps on.)

Azrael.

I beg to differ the price of a machine/computer does not equate to whether it’s a consumer or professional machine.

the 13MBP not the 2020 base model, is a higher performing machine than the MBA. I specifically mentioned the Core i7 which is not 1300 Euros but nice try.