Intel Expecting Apple to Transition to Custom ARM-Based Chips Starting in 2020

[MacsRuleOthersDrool]

My only concern with this is will this change still allow us to run Boot Camp and Parallels Desktop on ARM based Macs?

There will be a "Boot Camp for ARM" that will work in conjunction with MS' Window 10 64-bit for ARM, which has been out for about a YEAR now.

As far as Parallels and VMWare goes, that's a different story. If Apple develops/licenses a JIT Compiler-based ARM -> x86 "Rosetta", then those products might remain viable. BIG "if"!
[doublepost=1550874681][/doublepost]

Ah, sweet déjà vu… Remember the days of waiting for the magical G5… and then Intel was going to solve that dilemma forever and ever and ever?

Yeah, and then Intel decided to start slow-walking their CPU roadmaps...

Apple could for sure have predicted that back in 2005, right?
[doublepost=1550874982][/doublepost]

I think many people are missing something here:

What effect will this have on running Windows natively on the Mac (i.e. through Boot Camp)? I know not everyone does this, but many people do and I don't think this is going to be possible on an ARM based system is it?

I know there will be emulators that will pop up, there always are, but they consume overhead.

It's called "Windows 10 for ARM 64-bit", and has been a full-fledged, downloadable MS PRODUCT for about a year now. It doesn't EMULATE x86 code, it Cross-Compiles it "Just In Time" into native ARM on the first running, and then after that, simply launches the ARM-NATIVE version.

this is exactly how Apple handled the 68k -> PPC transition, and it worked SO well that Apple didn't even rewrite largish portions of MacOS until System 9.0.

 

[firewood]

The reason many developers have chosen a mac is simply because it is the one device that can be used to target all viable platforms to develop for:

Web development
Mac development
iOS development
Android development
Windows development
Linux and BSD development

You can already do Linux, BSD, and Android development on ARM based products (Raspberry Pi, Chromebook, etc.) A Raspberry Pi can already run ARM Docker images. MS is working on a Visual Studio that runs on Windows 10 on ARM. If Apple does ship an ARM macOS product, I assume there will be an Xcode for Mac and iOS for it in the App store.

So a hypothetical ARM MacBook will still be a fully multi-platform capable developer platform.

Note that Xcode already includes all the tools needed to create fat binaries which include both arm64 and x86-64 slices. What's missing is something like the frameworks for Marzipan.

 

[MacsRuleOthersDrool]

So this is how Mac will die.
To all saying cool and great arm gpus are fast and maybe even better that iris pro etc. well blame apple and their greed for not putting a dedicated gpu on the Macs.
As for cpu power don't make me laugh arm going against i7,...... not even against an AMD.

Would you like some ketchup with those words when you have to eat them?
[doublepost=1550875672][/doublepost]

Yes I do.

And that was APPLE'S fault, HOW, exactly?

 

[firewood]

This sounds like an unmitigated disaster. There is a ton of software that will never be ported to ARM.

Yup. Nobody buys MacBooks because there was a ton of Apple II software that was never ported to Mac, a ton of Mac software that was never ported to 32-bit clean System 7, a ton of 68k code that was never ported to PPC, a ton of PPC code that was never ported to Intel, and a ton of x86-32 code that won't be ported to x86-64. What did I leave out?

 

[MacsRuleOthersDrool]

hahah those who think ARM CPU is anything like an Intel or AMD chip. Apples and Oranges dependent on tasks. Its amazing when you can code and run specific things only on an ARM aka iPad Pro performance. Yet you have an intel and AMD CPU's which can do so much more and you'd see that in plenty of benchmarks with real world tasks and proper computing.

And it's definitely different when you can architect a CPU from the ground-up, like Apple can with ARM.

Oh, and BTW, NO ONE currently in business has more ARM experience than Apple. Period.

 

[PKoz]

I did not say the iPhone did not bring more people. I said Bootcamp brought many people

I said thin and light, and iOS developers brought them. You said, oh you must be new. Kinda sounds like you did. Now that the perspective is made clear you seem to see otherwise.

 

[MacsRuleOthersDrool]

Cooling... in a Mac? Never going to happen. They love thermal throttling too much.

What a ridiculous statement.

 

[Lankyman]

Thank God for the plucky Brits and ARM.

We're much happier dealing with Hauwei, letting them build our mobile infrastructure and maybe the odd back door too.

Oh BTW ARM was sold to Japan's 'SoftBank' which also has links to Hauwei.

 

[MacsRuleOthersDrool]

How about rendering, video editing, etc?

That's MUCH more SOFTWARE dependent than HARDWARE.
[doublepost=1550876252][/doublepost]

I'd wager that there won't be a full transition at first. Perhaps, migrating the non-Pro models to these Apple based chips will be the plan. Pro models will probably still utilize Intel chips.

I've only been saying this for over a YEAR now. Welcome to the party!

 

[cube]

I said thin and light, and iOS developers brought them. You said, oh you must be new. Kinda sounds like you did. Now that the perspective is made clear you seem to see otherwise.

I said you must be new because you implied Bootcamp did not bring many people.

 

[MacsRuleOthersDrool]

plenty more examples where an intel or AMD CPU will wipe the floor with an ARM based chip!

For now...

And are you comparing against APPLE ARMs. Not all ARMs are alike.
[doublepost=1550876497][/doublepost]

In synthetic benchmarks but not in real world performance.

Prove it.
[doublepost=1550876712][/doublepost]

As someone who is mainly doing music production and mixing on my iMac, I fear the new ARM chip future. If this happens this soon, I hope there is a good(!) working compatibility layer and the devs will port the DAWs and plugins rather fast with Apple providing some good migration paths and APIs.

Music production has always been an Apple niche. I can't imagine Apple not thinking that through.
[doublepost=1550876819][/doublepost]

Absolutely.
I wish I could run GeekBench for a living.

There is more to performance than raw "speed".
Ergonomics, screen size, colour accuracy — external storage… hell, there is quite a list where a desktop or laptop still beats an iPad. Hands down. Any race. Any time.

He was commenting on the speed of the ARM SoC in his iPad; not iPad vs. Mac.

Sheesh!

 

[Spock]

just in case you don'tknow... there was a period in the 90's, where Apple's isolation with no compatibility to windows, (amongst some other problems like extremely high prices that didn't match their value offerings, proprietary ports not easily adopted by the mainstream,) nearly saw apple bankrupt. Including a measly 1% total computer share.

what you're endorsing and claiming can't hurt Apple is a near repeat of exactly the behaviour that nearly doomed them in the past.

Apple at least has other industries such as phones now to fall back on should the Mac lineup tank like it did last time. But to claim that there's no risk here is foolish

at some point they have to care when people leave. it's not going to be one day they wake up and 100% of their customers left overnight cause of one move.

it's little moves, lose a few hundred here. lose a few thousand there. these death by a thousand cuts eventually erodes brand trust. this is something Apple needs to be careful off, that they don't alienate and drive away too many customers.

In the 90’s Apple was in trouble because of high prices, model confusion, the Mac clones, poor running software, poor management, tons of projects like the Newton, QuickTake, printers and even laptop batteries that were catching fire. And Mac OS was starting to show its age. Mac OS had support to open documents from PC formatted disks, Photoshop, Office etc. were all available for the Mac I could go on but compatibility with Windows was not the reason Apple was hurting in the 1990’s.Here’s 1995 product lines

• Performa 580CD
• Performa 588CD
• Macintosh LC 580
• Performa 5200CD
• Performa 5210CD
• Performa 5215CD
• Performa 5220CD
• Power Macintosh 5200 LC
• Performa 6200
• Performa 6205CD
• Performa 6210CD
• Performa 6214CD
• Performa 6216CD
• Performa 6218CD
• Performa 6220CD
• Performa 6230CD
• PowerBook 550c
• Power Macintosh 6200
• Power Macintosh 7200
• Power Macintosh 7500
• Power Macintosh 8500
• Power Macintosh 9500
• Performa 5300CD
• Performa 5320CD
• PowerBook 190
• PowerBook 190cs
• PowerBook 5300
• PowerBook 5300c
• PowerBook 5300ce
• PowerBook 5300cs
• PowerBook Duo 2300c
• Power Macintosh 5300 LC
• Performa 6260CD
• Performa 6290CD
• Performa 6300CD
• Performa 6310CD

 

[MacsRuleOthersDrool]

lol and see serious windows only apps LOVE THAT....not!

How would they even "know"?

The Windows x86 Application (not "App") is Just-In-Time Compiled on the "First Run" into ARM-NATIVE Code, THEN the ARM-NATIVE Code is what actually RUNS.

NOT "Emulation". "Cross-Compiling".

Same way Apple transitioned from 68k to PPC. And that worked so well that Apple was able to avoid re-writing chunks of MacOS basically forever.

And that was with WAAAAAY slower CPUs!

 

[0837990]

You can already do Linux, BSD, and Android development on ARM based products (Raspberry Pi, Chromebook, etc.) A Raspberry Pi can already run ARM Docker images. MS is working on a Visual Studio that runs on Windows 10 on ARM. If Apple does ship an ARM macOS product, I assume there will be an Xcode for Mac and iOS for it in the App store.

So a hypothetical ARM MacBook will still be a fully multi-platform capable developer platform.

Note that Xcode already includes all the tools needed to create fat binaries which include both arm64 and x86-64 slices. What's missing is something like the frameworks for Marzipan.

I have a Raspberry Pi 3+ B that runs Linux. I am quite familiar with Linux on ARM. It's not really there yet, though.
And Windows 10 on ARM might be there, but are the apps? No... I really don't need the OS as much as what I run on the OS.

Personally, I don't want to transition from x86 to ARM either. I do large scale multicloud environments in AWS, Azure and GCP. None of which are ready for ARM.

I expect at least 5-10 years for a production ready ARM solution to be ready for enterprises and serious mission critical solutions. When ALL of that is in place, then I might consider an ARM based computer. But that computer needs to give me low-level access just as x86 does today.

 

[MacsRuleOthersDrool]

It will become a mobile OS if an ARM is introduced.

Based on WHAT, exactly? The fact that you can't imagine anything else?

I GUARANTEE you that there are iPad Pros in Apple Labs running macOS Sierra AT THIS VERY MINUTE, with a Click <-> Touch layer added.
[doublepost=1550877392][/doublepost]

Maybe the new modular Mac Pro will be ARM based.

No.

That's too big of a leap right now.

 

[Tooldog]

Apple Inc. is a business. They exist to make money. That is their only purpose. They do not care how frustrating or expensive things are for us. If you don't like it stop giving them your money.

 

[MacsRuleOthersDrool]

I am very curious to know how this affects the upcoming Mac Pro, and if this is the cause for the delay.

Don't think so. Not directly, anyway.
[doublepost=1550877560][/doublepost]

This sentence should read: “This transition will greatly increase the number of Mac apps available, and it will greatly cut down on the overall quality of Mac apps.”

Why would it have to?

Because YOU think that ARM == iOS?

 

[Bin Cook]

We're much happier dealing with Hauwei, letting them build our mobile infrastructure and maybe the odd back door too.

Oh BTW ARM was sold to Japan's 'SoftBank' which also has links to Hauwei.

The brains are still in Cambridge, that'll do for me.

 

[MacsRuleOthersDrool]

if they want one app that works on all devices they better get some touchscreens on the mac asap

Funny: Because from what I have heard/seen, outside of Surface commercials on TV, VERY few people actually USE their Touchscreens for anything more than the occasional "tap" (and that, mostly because the trackpads on ALL Windows laptops SUUUUCK!).

 

[Think creative]

Sweet, so by 2022 it will all be worked out. I'd still love to have a mac pro be able to be a great windows gaming machine as well. I'd drop $5000 to have a desktop machine that could do both that and run my publishing business on. I guess I can keep dreaming.

 

[theluggage]

The question is, does it make business sense?
I personally don't think so; but I don't run Apple.

The big advantage is that Apple could specify their own CPUs with precisely the processor speed, number of cores, cache, GPUs etc. that they need for their computer designs. One factor in some of the current discontent with the Mac range & speed of updates is that they're totally at the mercy of what CPU models (particularly the different CPU + iGPU permutations) that Intel deigns to release.

just in case you don'tknow... there was a period in the 90's, where Apple's isolation with no compatibility to windows, (amongst some other problems like extremely high prices that didn't match their value offerings, proprietary ports not easily adopted by the mainstream,) nearly saw apple bankrupt. Including a measly 1% total computer share.

You may think Windows has the upper hand today, but its nothing like the 1990s. The 1990s was a DOS/Windows-only desert when "Wintel" came within a hair's-breadth of achieving a total monopoly. Pretty much every other competing personal computing platform got ground into dust. Frankly, Apple did well to cling on to a 1% share - the only way they could have done better would probably have been to drop the Mac and start making PC clones.

The internet, the rise of mobile computing, the success of Linux as a server platform, more powerful computers and operating systems that reduce the need for hardware-dependent code in applications have opened things up significantly.

For starters, MacOS is now Unix and has access to a ton of software written for Linux/BSD. More and more applications - particularly odd little bespoke things that would have had no chance of being ported to Mac - are web-based, and inter compatibility between browsers has improved beyond recognition. More and more developers are accustomed to supporting multiple platforms (because of iOS and Android).

Apple have pulled off architecture switches in the past in far more challenging conditions.

As I've said - the loss of x86/amd64 virtualization and dual-boot is going to be the biggest loss - but unless Apple are stupid and try to switch overnight, that might not be such a big deal by the time 2022 swings around.

Especially if the alternative might be Apple releasing Xcode for Windows/Linux and dumping the Mac.

 

[Reindeer_Games]

Based on WHAT, exactly? The fact that you can't imagine anything else?

I GUARANTEE you that there are iPad Pros in Apple Labs running macOS Sierra AT THIS VERY MINUTE, with a Click <-> Touch layer added.

In a lab is not proven tech yet I'm afraid. You are trolling yourself right now-quit, we squashed this already.

If macOS and Pro lines remain independent then I'm personally good and my concerns are handled.

History tends to repeat itself-I'd rather not. Thats all.

 

[MacsRuleOthersDrool]

ok--fine. But here is my concern:

what about boot camp? I this going to screw with my windows 10 when I decide to buy a new Mac. Second, what about the programs that came to Mac once apple transitioned from G-processors? Are they all going away then?

Windows 10 for ARM 64 bit has been a PRODUCT from Microsoft for over a YEAR now (and for 32 bit ARM back to 2016!). It handles x86 by automatically and silently Cross-Compiling the code into ARM-Native Code. Then it runs THAT. No "Emulation" Required!!! Also, An ARM-Based Mac would have an ARM Boot Camp. Boot Camp is pretty much just a set of Hardware Drivers.

See:

https://docs.microsoft.com/en-us/windows/arm/
[doublepost=1550879203][/doublepost]

In a lab is not proven tech yet I'm afraid. You are trolling yourself right now-quit, we squashed this already.

If macOS and Pro lines remain independent then I'm personally good and my concerns are handled.

History tends to repeat itself-I'd rather not. Thats all.

I'm fairly certain that they won't do something as foolhardy as immediately commit the entire Mac line to an "off the cliff" transition like that. I give that at least another 5-10 years, and maybe even never.
[doublepost=1550879321][/doublepost]

will they dumb down macOS ?

Why would they have to?
[doublepost=1550879684][/doublepost]

No, of course not. Someone might come out with an emulator, like the old Virtual PC, but it won’t run Intel code natively.

But it CAN do what Apple did DECADES-ago (and with GREAT success!) for their 68k -> PPC transition, and what MS is doing RIGHT NOW with its Windows 10 on ARM 64-bit: Just In Time Cross-Compiling. You cross-compile the x86 Code to NATIVE ARM Code either on Install or "First Run", then RUN THE NATIVE ARM CODE...

NO EMULATION NEEDED. When it Runs, it IS an ARM-NATIVE Application.

 

[JayMBP]

This would be the end of Mac as we know it... or even Apple.

The transition from PPC to X86 was difficult, but it also made the Mac compatible with what the rest of world is doing (as in Intel/Windows monopoly). Because they made Mac easier to work with and realize the potential of using a Mac without having to take many compromises. Apple had the full support of software devs.

Google, with much less burden on its shoulder, can't really seem to quite crack the windows dominance. I can't think how Apple is able to pull it off.

And quite honest though, I have the latest iPad Pro... and it's so far away from replacing my Mac that I don't know where to begin to talk about it.

I'd just put it this way... some things are meant to stay the way they are. The way I feel is that Cook & Co are betting that by merging both OSes, they can get devs for develop for one thing and make it work on the other relatively easily.

Nevertheless, they have to check first where the market is at. Unless Microsoft is also calling it quit on X86 architecture, going without the full support of devs will easily create a scenario that I can already envision:
The mobile version (UI-wise) will be too underpowered, while the desktop version will be too dumbed down due to the common denominator effect.

Devs either create a core that's targeting the desktop users (plenty of power) with the mobile version severely slowed down or a core targeting mobile devices with a desktop version that's too overly simplified.

 

[MacsRuleOthersDrool]

Apple gives zero F's about that - sorry man
I feel ya', but honestly they don't care and I don't really blame them.

Boot Camp is simply a set of Drivers to bridge Apple Hardware and Windows' I/O APIs.

In fact, MS could actually build-in "Boot Camp" into Windows; but for fairly obvious reasons, does not.

Long story short: It will be nothing for Apple to rewrite Boot Camp for ARM.