Apple Plans to Ditch Intel and Use Custom Mac Chips Starting in 2020

Just because the technology can do it, doesn't mean Apple will..

I reckon this change in 2020 if is done, will probably be the same as what happened when we moved from PowerPC (Rosetta) to Intel in 2006.. (dynamic binary translator)

Apple introduced backward compatibility as in "Rosetta' for older apps to function.. But over time they abandoned Rosetta and now its all Intel..

I predict the same would be true also here with Intel to ARM in Macs, (or whatever the new chips will be) If Apple went through with it.
 
Most big problems that aren’t tied to x86-64 are phishing. And hardware can’t stop that.

While x86-64 does have significant issues, a move to ARM doesn't solve all those issues either. Apple's ARM chips are/were all vulnerable to Spectre, so security is still an issue.

IIRC, Apple claims to have issued patches, but has anyone been able to verify independently that the patches fix the vulnerabilities? Also, does anyone know if ARM CPUs are also vulnerable to the newer SGX SDK exploit or is it an Intel only exploit?

I've highlighted your assumptions.

Sure, currently supported software will likely offer a new version that will be ARM-compatible. But there will be lots that either won't get the update (not supported) or will require buying a new version (or like you pointed out with Adobe, subscribing to new software).

I heard the same thing about USB-C, how everything would be USB-C not that the MB/MBP only have that port. Yet even years later, almost everything still comes in USB-A.



Sure, people don't generally care about the inner workings. But they'll absolutely care when things don't work.

Not being able to run x86/x64 Windows, for instance, is a pretty big deal. Every single person I know with a Mac in engineering runs Windows via bootcamp.

Another significant issue facing any transition to ARM is going to be the need to rewrite device drivers. I can see many companies refusing to write new drivers and using the transition to drive sales of new hardware.
 
Another significant issue facing any transition to ARM is going to be the need to rewrite device drivers. I can see many companies refusing to write new drivers and using the transition to drive sales of new hardware.
Apple seems to write their own drivers now anyways. For instance, my Canon printer driver is from apple and not Canon. Similarly Epson directs me to use Apple's Image Capture for my scanner.

While I agree with you, that a platform change will drive away developers in various forms. I wonder how printers/scanners (and other peripherals) will be impacted.
 
You can't be serious with this statement.

The reason Mac took off was because of their switch to Intel processors. Developers had less of a battle to port their apps. Lazy? You clearly have never ported anything beyond a simple "Hello" app.

Some of you have chosen to forget the PPC days and how few apps there were. :D

This is different if they basically convert to iOS. Most of the apps already exist.
 
You make some very good points, although I have never seen people on a car forum bragging about how much profit a company makes. :D

True! But maybe it's because no car companies have found a way to make a profit :p

(I say that jokingly. I have no idea if any are profitable)
 
As a sort of Apple fan, I hope this idea doesn't work. Try it out for a few years on the MacBook, let Intel catch up in that space, then return to full Intel and let that be that.

I don't think this is the end of the hackintosh either, there are a heap of intel machine's running MacOS today, and they'll be supported for a long time with upgrades, MacOS works on that architecture.

Anyway, either it doesn't work, or if it does its just for their small laptops (maybe a new Mac mini) and the real powerful stuff stays on Intel.
 
So many people are stressing out because they assume that leaving Intel means leaving x86. That's just not the case.
 
Agreed. CPUs are a commodity item and not a differentiator. Apple should focus on the end user experience and leave the CPU design to those who specialize in it.

Apple specializes in cpu designs, and their cpu designs differentiate iPhones and iPads from competing products.
 
Apple specializes in cpu designs, and their cpu designs differentiate iPhones and iPads from competing products.

In what ways? And in what way can Apple make a difference by spending resources to create their own line of CPUs? They lose the ability to buy at low cost from a large producer of them. I don’t see how Apple could beat their price structure, but I could be wrong there. What special aspects of the Mac OS need or would benefit from anything Apple would do with a CPU that isn’t already being done by Intel? It just seems expensive and pointless to focus so much on an area that makes so little difference.
 
In what ways? And in what way can Apple make a difference by spending resources to create their own line of CPUs? They lose the ability to buy at low cost from a large producer of them. I don’t see how Apple could beat their price structure, but I could be wrong there. What special aspects of the Mac OS need or would benefit from anything Apple would do with a CPU that isn’t already being done by Intel? It just seems expensive and pointless to focus so much on an area that makes so little difference.

Apple chips are less costly than intel’s because risc uses less die area for the same power/performance, and costs are a function of the die area and number of mask layers.

By making their own cpus Apple can include things like the secure element, Face ID logic, Apple Pencil support, High refresh rate circuits, etc all on the cpu, which drastically reduces system cost. A lot of this would benefit Macs too.

Right now Apple is two generations ahead of their mobile cpu competitors.
 
Wouldn’t take them two years to switch to AMD.

My guess is that they are developing a semicustom x86 / ARM solution (MCM?) with AMD. Their relationship on the graphics front seems good and AMD has both deep ARM and x86 experience so it would make sense. That's purely a guess.

My original point though is simply that leaving Intel does not mean abandoning x86.
 
Last edited:
Just because the technology can do it, doesn't mean Apple will..

I reckon this change in 2020 if is done, will probably be the same as what happened when we moved from PowerPC (Rosetta) to Intel in 2006.. (dynamic binary translator)

Apple introduced backward compatibility as in "Rosetta' for older apps to function.. But over time they abandoned Rosetta and now its all Intel..

I predict the same would be true also here with Intel to ARM in Macs, (or whatever the new chips will be) If Apple went through with it.

But you have to remember switching to Intel opened up a whole new world to include Microsoft.

Switching to an all Apple platform limits the experience unless all the other major players come on board.
 
From apple's perspective it makes total sense.

- one architecture to support
- their processors are getting faster leaps and bounds more quickly than intel (the A10X is a beast at the power level it runs in)
- no one can build a mac clone with pc hardware
- they can integrate their own features at their own pace
- they can build their own processors a lot cheaper than they'll pay intel or AMD

And most importantly - their product line is not at the mercy of another vendor's technological process or lack thereof.

Given iOS is bigger than Windows these days, and most software is either cross-platform or being written for mobile, i don't think losing x86 compatibility will hurt.

Those who use macs to run windows are niche users. The previous reason to do so was for a proper version of MS office, or gaming, but now even MS are pushing people to 365, which runs in a browser, that's not such a big deal. Apple doesn't care about running Windows games on a mac, and mac hardware is generally not gaming hardware anyway.
 
After coming back to Apple 15 years ago, the shift to ARM processors would be a point of moving back to Microsoft.

Back in 2003, when I moved back to Apple, Windows was a complete mess, and Apple "just worked". I need Intel compatibility to run PC software in VMWare, and also run Linux in VMWare.

If Apple moves to ARM, I'd have few reasons to stick with Apple. Besides, Windows is far more stable and nice-looking now.
 
In what ways? And in what way can Apple make a difference by spending resources to create their own line of CPUs? They lose the ability to buy at low cost from a large producer of them. I don’t see how Apple could beat their price structure, but I could be wrong there. What special aspects of the Mac OS need or would benefit from anything Apple would do with a CPU that isn’t already being done by Intel? It just seems expensive and pointless to focus so much on an area that makes so little difference.

When Apple buys chips from a Intel, Intel has to make a profit. That means everyone is paying for the R&D. But Apple is still going to use Intel to fab their chips regardless. And if they’re designing them, they can even play fabs off of each other. “That’s fine, we can just go to Samsung”
 
My guess is that they are developing a semicustom x86 / ARM solution (MCM?) with AMD. Their relationship on the graphics front seems good and AMD has both deep ARM and x86 experience so it would make sense. That's purely a guess.

My original point though is simply that leaving Intel does not mean abandoning x86.

They won’t do that, because then they would be reliant on AMD for a wide range of chips in the future, each of which would require that sort of inefficient collaboration. It’s not like ps4 or xbox where they need one new chip every 7 years.

I guarantee you it’s ARM.
 
Wouldn’t take them two years to switch to AMD.
They won’t do that, because then they would be reliant on AMD for a wide range of chips in the future, each of which would require that sort of inefficient collaboration. It’s not like ps4 or xbox where they need one new chip every 7 years.

I guarantee you it’s ARM.

I guess it all boils down to how valuable you think the x86 ecosystem will be to them in 2 years. 100% ARM makes sense only if you think they don't care enough about it or you think that there will be some kind of emulation solution that performs well enough (which would still need to be licensed from AMD or Intel). I am skeptical about them throwing away x86 that quickly and also that emulation will give them the x86 performance they would want to offer.

Tim Cook could probably search the office couch and come up with enough change to purchase AMD so if things get far enough along that is always an option for him to take it all in-house.

All conjecture of course.
 
I guess it all boils down to how valuable you think the x86 ecosystem will be to them in 2 years. 100% ARM makes sense only if you think they don't care enough about it or you think that there will be some kind of emulation solution that performs well enough (which would still need to be licensed from AMD or Intel). I am skeptical about them throwing away x86 that quickly and also that emulation will give them the x86 performance they would want to offer.

Tim Cook could probably search the office couch and come up with enough change to purchase AMD so if things get far enough along that is always an option for him to take it all in-house.

All conjecture of course.

If they buy AMD the cross license with intel is voided, so they wouldn’t be allowed to make x86 chips.
 
in 2 years intel will be 7500 single core and over 100k multi.

Screenshot_20180411-112006_Samsung Internet.jpg
 
If they buy AMD the cross license with intel is voided, so they wouldn’t be allowed to make x86 chips.

If the agreement isn't renewed, Intel won't be allowed to make x86-64 chips. That would be a bit of an issue for them.
 
Last edited:
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.
Back
Top