Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Well, first off we don't know that it will be ARM based. Apple could produce their own x86_64 chip and it wouldn't be a big deal. The real question is how would those chips stack up to intel or AMD. This is still years out, and it is just a rumor. No need to get too crazy just yet.
It cannot be an x86 or x86_64 iSA because Intel owns the Patent for it, and only a few Manufacturers in the world that have license to manufacture it. AMD , VIA and IBM are the only companies i know who have the license. Why would intel hand out more licenses of its ISA?
 
I'd be interested in seeing how they would change the A# chips to go into its Macs? More cores, higher clock speeds, etc.? Right now, the current A10 & A11s seem okay for low level things like iPads and iPhones, but not higher end iMacs, MacBook Pros, etc.
Maybe have an M series chip (M for Mac)?
 
One advantage of a new processor would be that it would force Adobe to rewrite Lightroom Classic and Premier to really optimize them....A lot of the programs would need a lot of work...

The performance gains need to justify the rewrite, as well as the business model to pay for it.

Do keep in mind that in CS4, Adobe delivered a 64-bit version for Windows ... but not for OS X.

There was finger pointing back & forth with Apple, but in essence, it boiled down to two factors: (a) changes in Apple's OS X roadmap, and (b) Adobe wanting the cheapest path, which was to perpetuate old code to avoid the higher cost of a major rewrite.
 
  • Like
Reactions: ssgbryan
Exactly, as I noted a few posts ago. Stick an iPad processor in a laptop where it can have a much bigger heat sink and you’d never have to throttle. Stick it in a desktop where you can have a fan, and you can bump up the frequency quite a bit.

Lots of people on the forums seem to be imagining Ax chips in the MBP.

But I really think that we should be imagining what they’re going to be like in the Mac mini, iMac (not pro) and especially the MacBook.

Mac Mini - massively cheaper entry level Mac, that’s great for education and is super super tiny and runs super cool. No fan. Think Apple TV sized

iMac - even thinner and even cheaper. Runs super cool. No fan. Also widens the gap between the iMac Pro (which stays on Intel for the foreseeable future) and thus Apple get more $$$ from businesses who need the power of the iMac Pro.

MacBook - insane battery life and much cheaper than the current MacBook. Comes with an LTE / 5G option. No fan. Again, widens the price, performance and functionality gap between the MacBook and MBP (which stays on Intel).

And by the way, who do you think will fab all of these chips? I’m betting it will be a company beginning with ‘I’.

(EDIT: I do think that the Pro Macs will have Ax processors eventually. But I suspect that we’ll have a few years of the scenario above first).
 
By the way, I also know a few things about MMUs, cache coherency, and the like, and your point is still not well taken because x86 is as complex or more so in each case. (For work I've done see, for example:

http://ieeexplore.ieee.org/document/641683/
https://www.ecse.rpi.edu/frisc/theses/MaierThesis/

I was also one of the designers of AMD's first 64-bit opteron/athlon 64, UltraSparc V, and Exponential x704, among others.

and for giggles https://www.law360.com/articles/784676/lg-gets-amd-graphics-patent-claims-axed-by-ptab)

Interesting work.
I did x86 cache controller design.
I was at Amdahl in the late 80's doing design in the computer development group.
But I have written papers on cache coherency, audio digital signal processing and unified flow for FPGA to ASIC transition.
My area of specialization was processor architecture, but have spent more time doing custom processors for video encode/decode and compression and not general purpose computing.

But, credentials aside I think you missed my point.

My overall point is that an A11 does not have the equivalent complexity of an Intel desktop processor.
By that I mean the additional peripheral interfaces needed for a balanced system.
Compare an A11 to the newly announced i9.

That's not an A11 with a few tweaks.
Can Apple design a processor to compete? Probably, if they hire enough people.
Does it make sense? Only if the ROI is in the billions of dollars because that is the kind of commitment they are going to need to make. It's not a single processor, they are going to need a family of processors and a processor roadmap.

Right now the logistics and people aren't in line with Apple replacing Intel in 2020.
 
Last edited:
  • Like
Reactions: jgbr and JPack
It cannot be an x86 or x86_64 iSA because Intel owns the Patent for it, and only a few Manufacturers in the world that have license to manufacture it. AMD , VIA and IBM are the only companies i know who have the license. Why would intel hand out more licenses of its ISA?
Apple designs, intel produces. Still possible.
 
This COULD work as long as the new chips implement the Intel instruction instruction set and remain compatible. Just like the way AMD chips work. It compatibility breaks then I'd have to abandon the Apple ecosystem.

What I wonder is if Apple can really do better than Intel. Likely not. So Apple has 2nd or 3rd rate CPU chips in all their products?
 
I can't wait for Apple to ditch x86 and Intel. Hopefully get back to (Open)Power for desktops/workstations and ARM for cheaper devices.
 
And Parallels or Fusion won’t allow you to do so?

If they move to a different instruction set, something not X86 (like ARM or PowerPC) then Parallels and Fusion will stop working. Both of those are VMs not emulators. A VM can run at near native speed an emulator is perhaps 10X slower.

But Apple just might make chips that are like AMD, Intel compatible but lower cost. Cost is Apple's real motivation no matter what they claim
 
Last edited:
I’ll wait to see how it turns out before proclaiming that I’ll leave Apple over this. But it does make me worried that we will go back to the days of having far fewer apps for the Mac. Also with my work computer which is a Mac I do have to use windows on occasion so no longer being able to boot into windows would be a problem there.
 
Lots of people on the forums seem to be imagining Ax chips in the MBP.

But I really think that we should be imagining what they’re going to be like in the Mac mini, iMac (not pro) and especially the MacBook.

The article said replace not augment.
Tim Cook has also said that iOS and MacOS are two different ecosystems and people here are speculating that they will merge. Tim Cook has explicitly stated that isn't the case on multiple occasions.
Also iOS is MacOS. It's a different GUI but it's still BSD.

But back to the point.
With multiple architectures, you are now back to the 68k/PPC changeover with "fat binaries" or emulation of at least one architecture. I see that as less likely, not more.

They also must keep updating the OS for all those orphan machines for at least 5 years after they discontinue making them.

But we are all just speculating.
 
Maybe someone can explain to me the implications. Will we lose the ability to run the apps we use now? For example Microsoft Office? Will they need to be ported? Will Windows Run on this new chip platform? Will we need to depend on third party software support, in other words will we be waiting around while software companies decide whether or not to support Apple's chips and then provide the software?

Bottom line, what will we gain, and what will we lose?
 
When Apple switched to Intel, I started to buy Apple laptops. When they will stop using Intel, I will switch away from Apple. I need to run Linux and Windows on my machine. It will either not work or it will make things more complicated, with little or no benefit for me as a customer.

I've had it with Apple. First they removed 32-bits apps and now this.
 
Last edited:
  • Like
Reactions: ssgbryan
Apple cant get ahead? Hold that thought, now google the Most valuable company on the planet. Wait a moment .... Yup!! That reaction is outstanding.
Well they are definitely good at separating people from their money. No argument from me there.

They invented the modern personal computer then squandered their market position due to bad strategy.

They invented the modern mobile OS then squandered their market position due to bad strategy.

No doubt they have a lot of your money in their bank account.

But they are always relegating themselves to somewhere between second place and also-ran.

Why oh why. Steve's genius came with some emotional issues and everyone understood that.

But Tim could have broken through the Reality Distortion Field and really catapulted them into the mainstream.

Instead he stopped innovating and pinched pennies.

Sigh.
 
  • Like
Reactions: ssgbryan
The article said replace not augment.
Tim Cook has also said that iOS and MacOS are two different ecosystems and people here are speculating that they will merge. Tim Cook has explicitly stated that isn't the case on multiple occasions.
Also iOS is MacOS. It's a different GUI but it's still BSD.

But back to the point.
With multiple architectures, you are now back to the 68k/PPC changeover with "fat binaries" or emulation of at least one architecture. I see that as less likely, not more.

They also must keep updating the OS for all those orphan machines for at least 5 years after they discontinue making them.

But we are all just speculating.
Indeed we are. It’s a report in Bloomberg not an official announcement from Apple, of course. But Bloomberg are often used when Apple want to ‘leak’ something, so I suspect that this is broadly true.

I suspect it’s testing the water and putting the idea out there etc. so that devs get behind the Marzipan initiative fully (which will almost certainly support Ax and x86 at some point soon).

And I do think that Apple fully intend to go with Ax for all of their lineup - just over a few years.

Agree that we won’t get iOS on the desktop (at least I hope not).

This move could leave iOS to focus on touch and AVR interfaces and then leave macOS for traditional GUI computing without either having to ‘squish’ elements from the other (see Windows 10. Ouch).

It will most likely enable them to share technologies even more than before and make both better ie building on things like the Secure Enclave (let’s not mention the touchbar).

For example, I fully expect that we’ll see Pro Macs with Intel CPUs but also a Ax ‘neural engine’ working alongside it powering the same machine learning, AI and Siri etc. that will be on iOS devices too.
 
I can't wait for Apple to ditch x86 and Intel. Hopefully get back to (Open)Power for desktops/workstations and ARM for cheaper devices.

If they do that then everyone who needs even one non-Mac app will be forced to buy two computers. I suspect Apple will only sell apps for consumers to do things like watch youtube. The will have to stick with the X86 instruction set or they break Fusion and Parallels and just as bad they break running Mac OS in a VM
 
Maybe someone can explain to me the implications. Will we lose the ability to run the apps we use now? For example Microsoft Office? Will they need to be ported? Will Windows Run on this new chip platform? Will we need to depend on third party software support, in other words will we be waiting around while software companies decide whether or not to support Apple's chips and then provide the software?

Bottom line, what will we gain, and what will we lose?

1. We might lose the ability is there isn't some emulation built into the OS. When they moved from Motorola 68k to PPC they built an emulation layer into the OS. This worked because the PPC was so much faster than 68k. Emulation probably won't be efficient when trying to emulate the x86. A recompilation layer that translates instructions to native processor instructions and stores the results might work

2. Long term all applications would need to be ported.

3. Windows won't run on this platform unless it's ARM or X86 instructions.

4. For Windows support we have always relied on third parties; VM Ware, Parallels, etc. That was a VM and not VM with emulation.

5. Yes, developers will need to decide to port their applications. See #2.
 
What's stopping Apple from designing their own x86 chips, licenses and patents? Maybe Apple can partner/acquire AMD/VIA leveraging the key licenses/patents that they hold.
 
Maybe someone can explain to me the implications. Will we lose the ability to run the apps we use now? For example Microsoft Office? Will they need to be ported? Will Windows Run on this new chip platform? Will we need to depend on third party software support, in other words will we be waiting around while software companies decide whether or not to support Apple's chips and then provide the software?

Bottom line, what will we gain, and what will we lose?

Yes, you completely lose the ability to run the apps you use now. They would need to be ported. Although porting could possibly be fairly simple in many cases.

Windows will not run on this new platform, at least not the way you want it. It wouldn't give you access to the ocean of Windows apps available today.

And yes, you would depend on third party software support, which would in some cases mean waiting.

Now, some or all of these can be mitigated by Apple. There could be emulation for X86 macOS apps. There is X86 emulation for ARM Windows. Emulation could possibly work reasonably, but right now in Windows it doesn't. Other solutions could be imagined, but are probably not practical.

What you gain is uncertain. Probably not performance, but you might not lose much either. Some think you will gain battery life. Some think the devices may cost less. Both are plausible, but there's really nothing in the article to indicate anything either way.

For someone not using many 3rd party apps, it's quite possible that the transition would be almost unnoticeable.
 
  • Like
Reactions: bopajuice
I'm not talking about Windows RT. I'm talking about Windows 10. With x86 Emulation...

https://www.windowscentral.com/windows-10-arm-not-windows-rt


Yes, there are some limitations (the biggest 2 being no x64 Emulation (yet) and limited Device Driver support (mostly an issue only for legacy peripherals), but most of them won't affect "non-intense" Windows-use:

https://www.theverge.com/2018/2/19/...dows-10-on-arm-apps-games-limitations-support

..and it seems that the product is still somewhat VaporWare; but only "just". Here's someone installing it on a Windows ARM Emulator (Yes, an emulated ARM emulating an x86 CPU!)

https://winaero.com/blog/install-windows-10-arm-qemu/

...and on a Raspberry Pi:

https://www.windowslatest.com/2018/02/09/developer-installs-full-windows-10-arm-raspberry-pi-3/

...so it LOOKS like it might actually be "real" afterall...
Uh huh.

Be careful not to frame this from the perspective of home users as home users are not what matter in computing.

That's what Apple has never understood.

You ran a VM on your weaksauce homemade ARM processor...inside of which you are running a weaksauce experimental variant of Windows...inside of which there is an emulator where you can run x86 applications slowly and stupidly.

Good for you. You are not enterprise IT and you are not a statistically significant user group.

What matters is what will work for a billion users. Microsoft understands that, Apple does not.
 
  • Like
Reactions: ssgbryan
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.