Apple Expected to Announce Arm-Based Mac Plans at WWDC, Transition Away From Intel to Begin in 2021

[Detektiv-Pinky]

<snip>

I totally concede we need to take that with a large pinch of salt seeing as it is just a Geekbench score, but there are real world videos on Youtube showing how fast the iPad can handle things like exporting video compared to traditional laptops.

I looked into this recently and found that the 'improvement' in speed mainly results from the fact that iMovie on the iDevices is able to cut videos without having to re-render the whole video. Whereas iMovie on macOS insists on rendering the whole video, even if I only trim it a little. This alone results in a huge difference in export times, but has absolutely nothing to do with CPU performance.

 

[Azrael9]

This news really cuts short the lifespan of the new Mac Pro. Nobody will want those in a few years.

It's 'on notice' to become 'legacy' kit.

Still...it's 'Pro' kit....for 'Pro' users...they can still use it for their 'Pro' work.

It's not like the Mac Pro wasn't dated when it came out. It doesn't use the latest Nvidia kit, it doesn't have PCIE4 (and even AMD have PCIE4 motherboards....) And you can get 64 core AMDs...but just not on Mac Pro.

It's not very price accessible either. So not 'that many' will buy it. Even less 'pro' customers to upset than before those that cleared out during the 6 year trash can debacle... So I doubt many will miss it when the rump that was left after that are replaced with Mac ARM Pro... Which I'm sure will be a more than adequate performer. Beating the intestines out of Intel's single core performance won't be that much of a stretch for Apple. Intel have been a sitting duck for the last 5 years or so and they are still...struggling to get out the mire they got themselves in.

The G5 quads are a footnote in history. The 'over engineered' Intel Mac flag ship will suffer the same fate, cpu wise, at least.

Azrael.

 

[thunng8]

The export is done by hardware and not by the CPU itself. Fixed-function hardware vs. general-purpose CPU. Therefore it is not a good comparision.
Fixed-function hardware is very fast but also inflexible. The encoding quality is usually worse than very good software encoders and improvements are not possible due to lack of updates.

Here you go: https://www.laptopmag.com/au/reviews/laptops/new-ipad-pro-2018-129-inch

Adobe Lightroom being much faster on the iPad Pro than the MacBook Pro and dell XPS 13 and surface pro. I personally tested the iPad Pro 2018 against a similar power usage MacBook 12” 2017 i5 model (actually not really similar notebookcheck measured the power usage and iPad Pro still use a lot less than the maccok) and it wasn’t even close. The iPad Pro ran circles around it on comparable software.

When Apple announces the high end A14 based Macs it will be a bloodbath. You’ll see many charts with 2x performance figures while using less power.

 

[Azrael9]

I deeply hope that this transition will not happen. When I switched away from

Apple to Intel...

Azrael.
[automerge]1591793879[/automerge]

Here you go: https://www.laptopmag.com/au/reviews/laptops/new-ipad-pro-2018-129-inch

Adobe Lightroom being much faster on the iPad Pro than the MacBook Pro and dell XPS 13 and surface pro. I personally tested the iPad Pro 2018 against a similar power usage MacBook 12” 2017 i5 model (actually not really similar notebookcheck measured the power usage and iPad Pro still use a lot less than the maccok) and it wasn’t even close. The iPad Pro ran circles around it on comparable software.

When Apple announces the high end A14 based Macs it will be a bloodbath. You’ll see many charts with 2x performance figures while using less power.

Sound post.

The iPad's performance is jaw dropping. No fan. And some of those benchmarks are outrageous. :O

My favourite bit highlighted in bold. 'Bloodbath.' I think Intel's single core is vulnerable. ...and I wouldn't be surprised with 2x metrics. I think that's what Apple's probably aiming for. Any emulation needed? No problem.

They want this transition. But they'll want to leave their Intel partnership from a position of strength. I think they'll want to bury Intel as they leave by the front door.

That's going to hurt in the morning, san.

Azrael.

 

[Detektiv-Pinky]

Not sure this is good news, they are already a monopoly of sorts and developing your own chips will only lead to higher prices and less choices because of no real competition.

I share your concerns, it will be a sad day for me, when Apple goes down this road.

There was a time about 10 years ago, when it looked as if Apple was taking over Universities with the Mac. Nowadays I see fewer and fewer Macs in the engineering departments, because the current annual OS updates make it difficult to stay on top with the required software tools (Matlab, Cisco Packettracer, Eagle, ...).
I guess, that when Apple goes fully proprietary, a lot of the current software developers will simply stop making the move. Even fewer people will use the platform for productive work and the Mac as I know it simply dies.

I came to the Mac as a beautiful, low-maintenance Unix-Box with a likable UI and well thought out procedures (tight PDF integration, compiler, system-wide spell checking, nice graphics toolkit,...).
It has been a good time...

 

[panjandrum]

I've already been screwed by them dropping 32-bit support on Mac. So many old games that are just unplayable now. And this change will get the rest.

Yep, and just for fun, go to a new Windows machine with a 64-Bit OS and try and run your old 32-Bit games. Guess what happens? They usually work! Sure, sometimes they need some tweaking, but they usually work. Windows has far, far, far, far, far (you get the idea) greater legacy support than the Mac. It's all so sad because despite much of the griping many of us do here, there are still some great advantages to the Mac, but Apple just keeps throwing up wall after wall after wall in the way. Apple is their own worst enemy, over and over again (that whole butterfly keyboard disaster is possibly the single best example of this). Apple has demonstrated, several times, that they are perfectly capable of supporting legacy software in ways Microsoft has never really had to do (when moving once to entirely and entirely different OS and once to an entire different CPU), and yet they can't be bothered to do even basic legacy support long-term, with all the $$$ they now have. Lame and shame.

 

[rp2011]

To do nothing and just stay with intel would be the easiest safest thing for Apple could do CPU wise. Just do nothing and wait for Intel. People at home running mathematica, calculating weather patterns or modeling the brain on a Macbook Air are fuming at the possibility of the switch. But i highly doubt Apple would risk so much switching to something else unless it would offer better results.
As a iPad pro user, I already know that there is no way this device could even exist if it had to run on an Intel chip. That the software is optimized for device is the icing on the cake.

 

[Homy]

Dieter Bohn writes: "Windows offers a roadmap of where things could go awry for the Mac. Windows on ARM still has unacceptable compromises for most users when it comes to software compatibility and expectations. I say this as a person who walked into those compromises eyes wide open, buying a Surface Pro X. I essentially use it as a glorified Chromebook and it’s very good at being that thing, but there’s no way Apple would want that for its Mac users.

Speaking of things Apple wouldn’t want: ARM-based Windows computers are slower. Unless you’re able to stay within those Chromebook-esque constraints, things get real chuggy real fast. We’ve all been assuming that Apple’s much-vaunted prowess at making fast ARM chips for iPads will translate well to Macs, but there’s no guarantee that’s true until we get to test them ourselves.

Another thing I’ve learned is that using a Windows computer with an ARM processor actually requires a higher level of technical expertise, because you need to know what won’t work and why going in. Basically, 32-bit Windows apps can be emulated in ARM, but more modern 64-bit apps cannot. And short of Googling (or, uh, Binging) around for a decent chunk of time, it’s difficult to know if an app you need will work.

Yet another thing we can take away from Windows is the idea that ARM and Intel versions can co-exist. It’s within the realm of possibility that Apple intends to support both x86 and ARM based Mac for the foreseeable future instead of just managing a transition. Windows is sticking to a plan to support both x86 and ARM. When ARM-based laptops and tablets started getting released, the message was “Here’s a cool new thing you can get if you want, but the reliable old thing isn’t going anywhere.” That’s the Windows way. If Apple were to take that tack, it would mean a sigh of relief for everybody who needs to buy a Mac for the next year or three. But it would also mean another potential pitfall. Windows on ARM simply isn’t getting the developer attention and support that standard Windows gets.

If Apple goes the Windows route and declares that it has no plans to sunset x86 support, then it needs to ensure that both ARM and x86 Macs feel equally supported. If it goes the classic route and declares that the future of the Mac is on ARM, then it needs to assuage concerns that every Mac in existence right now will become obsolete before its time. Neither path is easy."

So let's stop speculating and patiently wait to see what Apple has to say on WWDC.

 

[Azrael9]

Yep, and just for fun, go to a new Windows machine with a 64-Bit OS and try and run your old 32-Bit games. Guess what happens? They usually work! Sure, sometimes they need some tweaking, but they usually work. Windows has far, far, far, far, far (you get the idea) greater legacy support than the Mac. It's all so sad because despite much of the griping many of us do here, there are still some great advantages to the Mac, but Apple just keeps throwing up wall after wall after wall in the way. Apple is their own worst enemy, over and over again (that whole butterfly keyboard disaster is possibly the single best example of this). Apple has demonstrated, several times, that they are perfectly capable of supporting legacy software in ways Microsoft has never really had to do (when moving once to entirely and entirely different OS and once to an entire different CPU), and yet they can't be bothered to do even basic legacy support long-term, with all the $$$ they now have. Lame and shame.

Well, fair point on Windows. Can't argue with that.

Apple moves on and chainsaws the past. It's not like Mac users don't know what they're signed up for.

Apple do high prices and legacy chopping. It goes with the territory.

Yes. Apple are their own worst enemy at times. eg. butterfly keyboard, a £6k tower that replaces a £2.5k one... I'm sure there's a list somewhere... And as soon as they could? They stripped out the emulation layer for PPC. Deprecated Open Gl. Hosed the legacy ports on Macbook.

We'll see if it leads to a brighter future. But there can be pain along the way.

There are still great reasons to own a Mac. Many of them.

Azrael.
[automerge]1591795095[/automerge]

To do nothing and just stay with intel would be the easiest safest thing for Apple could do CPU wise. Just do nothing and wait for Intel. People at home running mathematica, calculating weather patterns or modeling the brain on a Macbook Air are fuming at the possibility of the switch. But i highly doubt Apple would risk so much switching to something else unless it would offer better results.
As a iPad pro user, I already know that there is no way this device could even exist if it had to run on an Intel chip. That the software is optimized for device is the icing on the cake.

Spot on.

Apple must have high confidence that they can deliver with this move.

Apple want to control the whole thing. They're a 1 trillion company now. They're no longer at Intel or M$ or Adobe's behest.

Azrael.

 

[Neirin]

If ARM MacBooks can't run windows I'll have to buy an older intel MacBook pro

 

[Azrael9]

I share your concerns, it will be a sad day for me, when Apple goes down this road.

There was a time about 10 years ago, when it looked as if Apple was taking over Universities with the Mac. Nowadays I see fewer and fewer Macs in the engineering departments, because the current annual OS updates make it difficult to stay on top with the required software tools (Matlab, Cisco Packettracer, Eagle, ...).
I guess, that when Apple goes fully proprietary, a lot of the current software developers will simply stop making the move. Even fewer people will use the platform for productive work and the Mac as I know it simply dies.

I came to the Mac as a beautiful, low-maintenance Unix-Box with a likable UI and well thought out procedures (tight PDF integration, compiler, system-wide spell checking, nice graphics toolkit,...).
It has been a good time...

I hear what you're saying.

Yet strangely, Apple are selling 20 million Macs per year now. And installed base of 100 million. More than when they had universities under their thrall. And if they sell 20 million Mac ARMs in their Apple Stores (they will...) they'll have that user based equalled in at least 5 years. And be part of a much bigger iOS universe which may have dramatic effects.

Apple are in the popular mindshare now. Apple stores, which I never thought I'd see in my lifetime.

'Macs' are in 1 billion iOS devices. Under the hood? The same thing.

And next year we'll have 'Macs' that run on native Apple cpu.

I never thought that would happen either..

Azrael.
[automerge]1591795579[/automerge]
I don't understand the 'crocodile tears' for Intel. There was a lot of gnashing and wailing over PPC. But it had come to the end of it's thermal rainbow.

Intel started ok in that context. But they've been a profligate advocate of 'Mac' cpu performance over the last 5 years or so. It sat on quad cores for years. Any advancements on more cores come from AMD pushing Intel's complacency and they're on 16 cores to intel's 10 core (which appears to be Intel's current consumer limit...with 'great' thermals...on that...)

Evolutionary. Hardly dramatic beyond the initial performance per watt advtantage it had over the PPC going into laptops.

Intel will be like PPC. In a couple of years we won't be talking about it. 'Yeah,man, no way you could have fit that volcano hot Intel chip into a iMac Air 'z' thin edition...and those Intel chips...man...twice as slow...'

And to get an Intel cpu. Quite easy and affordable to buy a 'Tower' for £1000 that will run rings around an iMac.

Running a modestly priced PC tower and an affordable 23 inch iMac may be a common sense thing to do. At least in my case.

Azrael.
[automerge]1591795736[/automerge]

If ARM MacBooks can't run windows I'll have to buy an older intel MacBook pro

If it lasts 5-7 years...it should still work ok doing what you want it to do. And by then, nobody will be talking about Intel on Mac (or Windows if M$ keeps going the way they are...)

Azrael.

 

[jlc1978]

Whilst I 100% look forward to this and ANY improvements that can be made to help computers run faster.
I still wish the people who say how fast ARM runs on a Phone/Tablet could be made to understand, both those devices are running a VERY light OS and apps designed specifically for them.

To me that is the key - the chip and OS are designed specifically for each other.

If Apple put an ARM chip in a, dare I say it "Proper Computer" running a full heavyweight OS like MacOS or Windows 10, then load up full heavyweight programs to run on it. Photoshop, 3D Cad etc etc.

And the ARM chips does all this and blasts past Intel and AMD I will be shocked/stunned and very happy indeed.

I suspect however this is not possible (yet) And Apple will come up with some cheat?
Like a special OS and/or special Apps to make them able to run on some cut-down laptop.

Apple would have the same advantage in a PC as a tablet or phone - designing the OS specifically for the chip and vice versa. They do not have work with generic processor. In addition, they could off load some tasks to other Arm processors in the system designed for specific tasks. In the end, the tight integration between the hardware and software give Apple the best shot at pulling it off.

Please tell me I'm wrong if you think I am
I would love to be proven wrong.

I have no idea if you are right or wrong but I doubt Apple would be seriously looking at an ARM processor in Macs if they didn't think the could make a better machine with it than with "Intel Inside." Only time will tell.

 

[Detektiv-Pinky]

<snip>
The iPad's performance is jaw dropping. No fan. And some of those benchmarks are outrageous. :O

Azrael.

These benchmarks mainly show dedicated graphics functions that are well suited for offloading to silicon. And sure, integrating those functions into a single chip yields some measurable benefits.

However, in computing there is always this tradeoff between implementing functions in hardware or software. Hardware is faster, software is more flexible. On a media-consumption device I may favour battery life and ease of operation. On a general purpose computer I rather like the ability to install and develop software and functions without vendor lock-in.

 

[jlc1978]

Not to mention Macs won't be compatible with industry standard PCs anymore; say goodbye to virtualization and Boot Camp.

I am not sure if it means the end of Windows on a Mac. MS already has an ARM based Windows, even if it is not fully completely compatible with all Windows 10 programs. Much will depend on what future MS sees for the ARM, if Apple produces a capable machine with a much longer battery life laptop manufacturers may reconsider using the ARM in a machine; and work with MS to develop a real Windows.

Even so, that does not mean ARM Mac couldn't run a VM and Win10, just we'd just be back to the Performa/VirtualPC world. While that may not satisfy gamers, users who just run business programs probably wouldn't see much of a performance hit and still be able to use a Mac as needed. Anecdotally, I am one of those users. I run a VM just to be able to develop in PowerBI and occasionally troubleshot a formatting issue when a Mac Office document is opened on a Winboxen. As long as the Mac could do this at the speed of an average mid priced PC laptop I would be able to do what I need to do. YMMV

 

[Azrael9]

To me that is the key - the chip and OS are designed specifically for each other.

Apple would have the same advantage in a PC as a tablet or phone - designing the OS specifically for the chip and vice versa. They do not have work with generic processor. In addition, they could off load some tasks to other Arm processors in the system designed for specific tasks. In the end, the tight integration between the hardware and software give Apple the best shot at pulling it off.

I have no idea if you are right or wrong but I doubt Apple would be seriously looking at an ARM processor in Macs if they didn't think the could make a better machine with it than with "Intel Inside." Only time will tell.

Apple have moved away from middleware software with X-Code and Swift. And now, with the hardware? The same. The A chip.

Part of the Mac's problem was middleware. It didn't sell enough volume to dictate specific market standards to it's own advantage. It only adopted Open GL to try to get some gpu parity and access to market after Quickdraw 3D.

But it was always a 2nd class citizen in its implementation, updates and what developers brought to the table. 2nd hand performance and ports.

Sounds like the Mac ARM chips? The one problem they won't have...is gpu or graphics. Nor single core performance. For everything else? Many cores. MANY cores. As many as it takes to bury the competition.

With Metal. That ends. With Mac ARM. That ends. It's going to take a few more years to turn the wheel. Of this synergy between software and hardware hegemony aligning for 'Mac ARM' (being the last piece....) But when it does...

Azrael.

 

[jlc1978]

These benchmarks mainly show dedicated graphics functions that are well suited for offloading to silicon. And sure, integrating those functions into a single chip yields some measurable benefits.

Which is what I'd expect Apple to do to drive performance, whether on one or several chips.

However, in computing there is always this tradeoff between implementing functions in hardware or software. Hardware is faster, software is more flexible. On a media-consumption device I may favour battery life and ease of operation.

My guess Apple is eyeing battery life as one reason to consider going to ARM, and has never been especially concerned about making it easy to do things they think they can do better in HW or SW for that matter.

On a general purpose computer I rather like the ability to install and develop software and functions without vendor lock-in.

But isn't that what programming for an Apple device does?
[automerge]1591797155[/automerge]

We're going to have the same thing happen if they transition to ARM. I'm pissed because this is unnecessary; the x86 architecture is fully capable. At least with PowerPC -> Intel they had a good reason for the switch.

We can always reformat the Mac to just run Windows...

 

[Detektiv-Pinky]

Which is what I'd expect Apple to do to drive performance, whether on one or several chips.

My guess Apple is eyeing battery life as one reason to consider going to ARM, and has never been especially concerned about making it easy to do things they think they can do better in HW or SW for that matter.


But isn't that what programming for an Apple device does?

Well, for me a computer is a tool and not an object of religious devotion.

Whenever I can, I write my texts in LaTeX, my presentations in Markdown, my programming in Python and I like to take my data with me from one cloud-provider to the next.

I find it convenient that I have software tools on my Mac that 'transition' easily between different operating systems and I daily exchange files and data with very many people without regard of their specific computational platform. I couldn't care less what type of CPU runs in any of these machines, as long as it is an open platform that facilitates these types of workflow.

I have an 2011 iMac and a 2014 Macbook on my desk. Performance wise these machines run just fine, battery life is still good on the Macbook. What difference does an ARM processor make here? Other than a faster obsolescence of current hardware and potentially even higher walls around the OS in the future...

 

[cmaier]

The way I understand it is that a 128 bit processor can manipulate 128 bit pointers in a single cycle. Using bank switching or, worse, multiple 64 bit registers would slow things down.

wikipedia notes

Notably amd64 has modes where integer operations are 32 bits even though addressing is 64 bits. My point is you can do the same thing to get to 128 bits. No need to define 128-bit integer operations.

 

[whiteboytrash]

I share your concerns, it will be a sad day for me, when Apple goes down this road.

There was a time about 10 years ago, when it looked as if Apple was taking over Universities with the Mac. Nowadays I see fewer and fewer Macs in the engineering departments, because the current annual OS updates make it difficult to stay on top with the required software tools (Matlab, Cisco Packettracer, Eagle, ...).
I guess, that when Apple goes fully proprietary, a lot of the current software developers will simply stop making the move. Even fewer people will use the platform for productive work and the Mac as I know it simply dies.

I came to the Mac as a beautiful, low-maintenance Unix-Box with a likable UI and well thought out procedures (tight PDF integration, compiler, system-wide spell checking, nice graphics toolkit,...).
It has been a good time...

Then you’ll also remember when software companies including Apple would charge for an OS release. First it dropped to $99 then free. I also don’t think you can get any more ”low maintenance“ than iPad OS hence why it’s likely Apple wants to move away from Intel to reduce the maintenance footprint it has to maintain on behalf of intel.

Lastly software developers will go where there is critical mass and a distribution system that providers them revenue. Apple has a market share on ARM based devices to keep developers from going hungry.

 

[firewood]

For myself, I use Mathematica extensively in my workflow. As it stands now, I would not be able to use Mathematica on an ARM platform.

False. Mathematica not only runs only run on an ARM platform, but you can get Mathematica for free on Raspberry Pi's (a popular educational ARM platform) running Raspbian. See: https://www.wolfram.com/raspberry-pi/

 

[Lounge vibes 05]

I hope Apple realizes, Intel Macs released as early as this year can’t get just 2 more measly macOS updates and that’s it. During the Intel transition Leopard became the last PowerPC release; even though there turned out to be Snow Leopard PowerPC beta. I think they were still capable of at least 3 more universal releases considering how much G3 and G4 Macs still work to this day. They are almost appliance like.


Considering they have been supporting Macs as old as 2012 with macOS updates, a decent 5 years of support is the most respectable thing to do.

For those who splurged on a MacPro, they should get at least 10 more macOS updates, cause that’s not a computer you throw out after 5 years or even 10 years.

Anyway, I better start collecting them old Intel Macs. I certainly will be watching this closely to see if it’s viable upgrade or switch back to a Windows machine in the interim.

[automerge]1591802098[/automerge]
I don’t think updates will be too big of an issue. Apple went on a computer killing spree between 2009 and 2012. Mountain lion, which came out in July 2012, cut off support for some computers at weren’t even four years old. But ever since 2012, they have been very cautious about cutting off computers. Same with iOS. The iPhone 3G only got about two years of support. Meanwhile, the iPhone 5s got almost 7. So I think they are going to at least support these computers for the next 6 to 8 years

 

[pankajdoharey]

I hope they only do this for low end macs. I use too many virtual machines for this ever to work for me professionally.

Well if they are planning a 12 core processor it could easily work for you, what are the problems you can imagine?

 

[k27]

Well if they are planning a 12 core processor it could easily work for you, what are the problems you can imagine?

This does not help him when he needs to virtualize x86 software.

 

[ruka.snow]

Well if they are planning a 12 core processor it could easily work for you, what are the problems you can imagine?

Even a 80 core ARM processor isn't going to run Docker images and full on VM's if they are x86. Though the comment does bring back 'fond' memories of Microsoft Virtual PC for Mac.

 

[thequicksilver]

I don't think any of us can claim to be surprised that this is coming.

I was around at the time of the Intel announcement. It was genuinely shocking when CNet broke the news on the Friday before WWDC that time. But the smoke signals that *something* had to change had been around for a while. Missing the 3GHz G5, no PowerBook G5, and G5 desktops running hot and expensive.

In the end, the existing PPC Macs basically had an Intel chip dropped in the existing chassis to get a major performance and efficiency boost.

This time around, I think it's been more subtle. We know Intel have lost their way, but I can't help but feel that Apple have wanted it to happen in some ways. This way they will control almost the entire product, software and hardware, kill the Hackintosh scene, and make their profit margins even healthier.

I briefly had a new i5 MacBook Air last month, but returned it when I saw that performance of 60fps video playback was full of frame drops, and paled to the performance of my base spec 2018 iPad. The poor cooling in the MBA did not help either. I feel confident they could drop a decent ARM processor into that MBA chassis, where the cooling solution is suddenly adequate for ARM processor and the Intel machine is left looking like a relic. Don't forget that the first MBP was announced as being up to *4-5x faster* than the outgoing Powerbook G4.

The question is how much of the range they can move to ARM and get a performance gain. Air and mini - almost certainly. MBP? Possibly the 'lower end' 13", 15" unsure. iMacs sure, but less confident on the iMac Pro or Mac Pro.

Personally I'm hoping it is not a full ARM transition, and that Intel will remain an option at the 'Pro' notebook and desktop lines. I like having a Mac I can run Windows on where needed, and my new MBP 13 is a lovely machine. But we approach another fork in Mac history.

Intel have brought us Mac users a lot. I hope whatever comes next is as good. But given recent steps to 'lock' users in (application signing, T2 chip, soldered components etc), I do have some trepidation.