The biggest issue with a Mac with ARM CPU is MacOS developers would need to maintain two separate binaries. One for Intel and the other for ARM.
That's exactly what they did during the PPC to INTEL transition and that went pretty smoothly. I think the same was true for the transition from 32-bit to 64-bit hardware.
ARM is great for power saving, and in some tasks they exceed the slower Intel CPUs. But there are other tasks they're much slower at, and generally even the best ARM CPU is miles off the mid range Intel ones. I don't really see the point in switching architecture again, people aren't complaining about the battery life on MacBooks, which is really the only advantage x86 in a laptop has.
Maybe you missed it but the latest Apple SoC beat out Intel chips in speed tests. iPads are out-performing some MacBooks now.
There are two things that make ARM CPUs natural move for Apple:
- more power for less energy - decreased dependency on Intel
...the first is becoming fact. The second is pretty obviously another pain point, as Intel has blown deadlines and been unable to provide updated chips for higher-end Macs. The writing is on the wall.
If you actually read my post you'd find I said "in some tasks they exceed the slower Intel CPUs. But there are other tasks they're much slower at, and generally even the best ARM CPU is miles off the mid range Intel ones." It does not beat the midrange Intel CPUs in most tests, it beats the low range one in some tests.
Yes it's good that Apple will reduce its dependency on Intel, but at the same time everyone is getting close to the physical limits of CPU speeds, it's now down to the efficiency of the architecture. x86 is not efficient whereas ARM is much more efficient, but eventually Apple will hit the same walls that Intel has - albeit with less architectural overhead. Another issue with non-Intel is that all the software written to take advanced of instruction set extensions like SSE will have to use generic instructions on Apple's ARM CPU, making those programs much much slower. Currently those advantages can be ported from other platforms easily, as it's the same CPU. But companies aren't going to spend thousands of programmer hours (or hundreds of thousands of $) to rewrite those routines to take advantage of whatever task-specific acceleration instructions Apple adds.
There's a huge number of things to consider when switching away from Intel, a lot more than a emotionally charged "yeah lets boot Intel coz I've historically associated them with Microsoft and they've hit the laws of physics which won't apply to Apple"
Intel hasn’t hit the laws of physics. They’ve hit the laws of what they can deliver reliably.
I did read your post. Like I said, Apple ARM CPUs are outperforming Macbooks, it’s already happened. Way back with the iPhone 6s over the MacBook:
“This year, the A13 has essentially matched best that AMD and Intel have to offer.”
...mobile device form factor aside, it is clear that Apple ARM CPUs offer more powerful for less energy. Obviously one day they will hit a wall — that’s a universal truth about any computing bus so it’s pointless to say. The point is today, and in the relevant future, ARM is a better option for Apple’s needs.
... but ARM's disadvantages way outweigh the advantages imo.
IMO anyone who owns a MacBook Air is the market for the first generation ARM Mac - they will appreciate the performance and battery life advantage, and there might even be a price advantage. Hopefully Intel will get out of their stagnation and make it impossible for Apple to convert totally to ARM. Either way, there will be some exciting times ahead.
As smooth as the last transition was it took a long time, almost 8 years. I'm in the market for a new laptop now but I expect to get 7 years out of it and I expect full support from Apple and application developers during that time. Sitting here with a 2013 15" MBP, I could stay the course for maybe another 2-3 years if I was not limited to 16GB of RAM which is far short of the 32GB I need.
If I were a Mac Pro user I would expect far more than 7-years, at least 10-15 years but just for security updates and application compatibility. At least there is an option to flip the INTEL models over to Windows once Mac OS and application support drops off.
Are you saying you’d expect 10-15 of OS and application support for a Mac Pro purchased today? That’s a tall order. I work in enterprise with Windows desktops and servers and I don’t know of any of our machines having that kind of support.
No that's not what I meant. Maybe I wasn't clear. I expect at least 7-years of full support and 10-15 years of OS support for security updates and application compatibility (Apple applications that come with MacOS).
My 2008 15" MBP was supported by new OS releases for about 8-years (2008-2015/16 or 8-years) with it's last OS security update in 2018 (10 years total). My 2013 15" MBP was supported by new OS releases for at least 7-years (2013-Current 7-years) and will be supported in the next OS release and possibly many more after that (8+ years) if the security updates go on for another 2-3 years after that (10-15 years total is realistic).
Based on that I see 7-years with full support and 10-15 for OS security updates and application compatibility as realistic, since that's pretty close to what they have done already. I wouldn't expect them to support hardware repairs for that long but from their "Vintage and obsolete products" page it looks like 6-7 years is the current average. Looks like my Early 2013 MBP just fell out of service recently but late 2013 MBP is still serviceable.
So what I said is pretty much what they are doing now.
10-15 years of security and compatibility support? With macOS? Well, good luck.
ARM chips are much more power-efficient than Intel x86 chips, and generally offer better performance-per-watt. That's due to a variety of reasons, including a simpler instruction set, the use of fewer transistors and overall slower clock speeds. ...
x86 processors execute x86 instructions by translating to an internal RISC format (or micro operations) like earlier processors did. ARM processors have no such translations because the instructions are already RISC. Internal hardware translation is of course very inefficient (I used to choose assembly instructions that executed in 1 cycle when programming in assembly for such processors) and apparently done by Intel to be able to change processor backend hardware at will (designers flexibility at the cost of processor complexity and performance). Compilers that target x86 assembly have a problem choosing specific instructions because Intels hardware backend has a fuzzy execution of specific instruction combinations and cannot be predicted performance wise (logically it should pan out, but I'm not 100% sure about that). Intels x86's try to do way too much predicting code and scheduling instruction parts, confusing compilers which reduces the efficiency even more. ARM doesn't need al this. As a consequence Intels x86 processors burn power while ARM sips it.
Another important performance difference has to do with chip fab production. Intel has its own fabs (which cost billions of dollars) and have long be the feature size champion (size in nm of transistors on chip). They thus where able to produce reasonable fast complex processors because others produced chips with a larger feature size (and chip speed, density and watts all go up, up and down quadratically inversely proportionally to the feature size). A few years ago TSMC became feature size king, because ASML sells them the best UV waver stepper fabs, thus ending Intels reign.
I think Apple’s roadmap with ARM is not the MacBook, but the iPad. It seems obvious they are slowly converting it to a full fledged computer, running on ARM.
Why the iPad and not the MacBook? Another simple answer. The instruction set of the iPad and iPhone has always been ARM. The instruction set for their computers has always been X86 (post PPC).
As the iPad grows in power and utility, developers will continue to expand their SW offerings and Apple can gauge if ARM can overtake or match X86 in consumer SW. When they are happy the transition is sustaianble, they can start moving the ARM chips into thier computers.
Whatcha bet?
The software would have to follow —which is why that transition probably won't be day and date with the lower-end models.
I wonder what fraction of Mac Pro users are using the Mac Pro primarily to run Apple software, especially FCPX and LPX. My hunch is that it's a big fraction -- maybe 80% or more. If so, then Apple can make sure that those apps are ready to go on an ARM-based Mac Pro.
The reason I suspect that FCPX and LPX are the main uses for the Mac Pro is that it seems custom built specifically for that software. The hardware that makes the Mac Pro special isn't the Xeon -- instead it's Afterburner and the GPU modules. In a lot of ways, the current Mac Pro has more in common with those old highly specialized SGI workstations than with other previous Macs. If you just want a lot of CPU power to run non-Apple software, you're probably not going to buy a Mac Pro --- there are FAR more economical (and powerful!) options available (for example, https://system76.com/desktops/thelio-major-r2/configure ).
The biggest issue with a Mac with ARM CPU is MacOS developers would need to maintain two separate binaries. One for Intel and the other for ARM.
That's exactly what they did during the PPC to INTEL transition and that went pretty smoothly. I think the same was true for the transition from 32-bit to 64-bit hardware.
The vast majority of developers with apps on the App Store will not need "fat" binaries. They'll upload their apps to the App Store compiled to BitCode and Apple will download a compiled x86 version to existing Macs and an optimized ARM-based version to the new Macs.
Just as iOS app developers upload bitcode versions and Apple delivers a version of the app optimized for the processor and device on which it's being downloaded.
Catalyst isn't in any meaningful way support for any "ARM architecture". It's support for iPadOS and for multitouch as a UI. For devs making a Catalyst app that runs on iPad and then macOS, that's really the point: iPadOS and macOS. The macOS side of it is an underpinning of "iOSonMac" -- well, now "iPadOSonMac" as an operating layer/framework.
Any differences will end up being most certainly bugs attributed to Apple's technology and come down to devs awaiting fixes and providing workarounds to their customers until Apple supplies bug fixes.
... but ARM's disadvantages way outweigh the advantages imo.
IMO anyone who owns a MacBook Air is the market for the first generation ARM Mac - they will appreciate the performance and battery life advantage, and there might even be a price advantage. Hopefully Intel will get out of their stagnation and make it impossible for Apple to convert totally to ARM. Either way, there will be some exciting times ahead.
If they own a MacBook Air a clamshell iPad would be a better solution! The next gen of iPadOS will have all of the metaphors of a Mac laid out in iPadOS. There is no reason to cannibalize the Mac (at least yet) The better direction is to swing to AMD Ryzen and Threadripper for now, or work with AMD to create a hybrid A-Zen chip!
I see Zen based CPU's as the transition and then the hybrid. If makes no sense to go ARM in the Pro market systems.
Apple is Pushing out affordable expandable macs that can last 5+ years. The Paltry mac mini/imac are lackluster and limited for user who want more out of a mac without paying $5k for it. With the possibility of new ARM chip for desktops this will limit the x86 ability of *cough* users to have a device that's flexible. Apple used to be for Everyone and open now it's tailored for the wealthy and ignores the users who made it and supported it in the past.
Apple is Pushing out affordable expandable macs that can last 5+ years. The Paltry mac mini/imac are lackluster and limited for user who want more out of a mac without paying $5k for it. With the possibility of new ARM chip for desktops this will limit the x86 ability of *cough* users to have a device that's flexible. Apple used to be for Everyone and open now it's tailored for the wealthy and ignores the users who made it and supported it in the past.
Apple is Pushing out affordable expandable macs that can last 5+ years. The Paltry mac mini/imac are lackluster and limited for user who want more out of a mac without paying $5k for it. With the possibility of new ARM chip for desktops this will limit the x86 ability of *cough* users to have a device that's flexible. Apple used to be for Everyone and open now it's tailored for the wealthy and ignores the users who made it and supported it in the past.
I've been hearing this bolded part (emphasis mine) for 30 years.
The biggest issue with a Mac with ARM CPU is MacOS developers would need to maintain two separate binaries. One for Intel and the other for ARM.
That's exactly what they did during the PPC to INTEL transition and that went pretty smoothly. I think the same was true for the transition from 32-bit to 64-bit hardware.
The vast majority of developers with apps on the App Store will not need "fat" binaries. They'll upload their apps to the App Store compiled to BitCode and Apple will download a compiled x86 version to existing Macs and an optimized ARM-based version to the new Macs.
Just as iOS app developers upload bitcode versions and Apple delivers a version of the app optimized for the processor and device on which it's being downloaded.
That’s not at all the issue for developers: the tool chain has always built a far binary and as far as I can recall these never been any issue developers have had with that.
What devs HAVE had to worry about is that they are in fact two separate build products. Meaning two test targets. Two debug targets, etc. For simpler, higher level stuff that sticks with top level toolbox stuff that’s approaching a zero issue, but for projects that require getting deeper into coding optimizations, or those involving 3rd-party projects that have lower-level code and optimizations, that’s where you most likely end up with compatibility issues.
Apple has worked to mitigate those and has learned a lot from the 68K -> PPC and PPC -> x86 transitions, but that can’t ever guarantee that issues won’t crop up.
The more I think about it, the more it seems that Apple isn’t going to release ARM replacements for their lower-end Macs first. I think they’re gonna want to make a big performance statement right at the off.
If I were a Mac Pro user I would expect far more than 7-years, at least 10-15 years but just for security updates and application compatibility. At least there is an option to flip the INTEL models over to Windows once Mac OS and application support drops off.
If you were a pro user you would be getting a new computer in less than 7 years
... but ARM's disadvantages way outweigh the advantages imo.
IMO anyone who owns a MacBook Air is the market for the first generation ARM Mac - they will appreciate the performance and battery life advantage, and there might even be a price advantage. Hopefully Intel will get out of their stagnation and make it impossible for Apple to convert totally to ARM. Either way, there will be some exciting times ahead.
As smooth as the last transition was it took a long time, almost 8 years. I'm in the market for a new laptop now but I expect to get 7 years out of it and I expect full support from Apple and application developers during that time. Sitting here with a 2013 15" MBP, I could stay the course for maybe another 2-3 years if I was not limited to 16GB of RAM which is far short of the 32GB I need.
If I were a Mac Pro user I would expect far more than 7-years, at least 10-15 years but just for security updates and application compatibility. At least there is an option to flip the INTEL models over to Windows once Mac OS and application support drops off.
Are you saying you’d expect 10-15 of OS and application support for a Mac Pro purchased today? That’s a tall order. I work in enterprise with Windows desktops and servers and I don’t know of any of our machines having that kind of support.
No that's not what I meant. Maybe I wasn't clear. I expect at least 7-years of full support and 10-15 years of OS support for security updates and application compatibility (Apple applications that come with MacOS).
My 2008 15" MBP was supported by new OS releases for about 8-years (2008-2015/16 or 8-years) with it's last OS security update in 2018 (10 years total). My 2013 15" MBP was supported by new OS releases for at least 7-years (2013-Current 7-years) and will be supported in the next OS release and possibly many more after that (8+ years) if the security updates go on for another 2-3 years after that (10-15 years total is realistic).
Based on that I see 7-years with full support and 10-15 for OS security updates and application compatibility as realistic, since that's pretty close to what they have done already. I wouldn't expect them to support hardware repairs for that long but from their "Vintage and obsolete products" page it looks like 6-7 years is the current average. Looks like my Early 2013 MBP just fell out of service recently but late 2013 MBP is still serviceable.
So what I said is pretty much what they are doing now.
10-15 years of security and compatibility support? With macOS? Well, good luck.
...macOS releases were every 2 years from 10.4 ~10.7, which may have made things easier all around, and then SJ passed...
Snow Leopard + Rosetta quite literally runs 10 versions (2001~2013) of the main vertical application software on this studio's macs.
A 2010 Pro Mac can literally run 20 years worth of the above software across two macOS (Snow + High Sierra)
Windows support lifecycles have been much longer, with Windows 7 spanning 2009 ~ 2020, offering business customers more choice in upgrade timing, training and migration efficiencies.
Current macs will only run TWO versions (2018~2019) of the same vertical application mentioned above. How this is 'better' for this long time Apple customer and users generally I simply don't understand...
ARM is great for power saving, and in some tasks they exceed the slower Intel CPUs. But there are other tasks they're much slower at, and generally even the best ARM CPU is miles off the mid range Intel ones. I don't really see the point in switching architecture again, people aren't complaining about the battery life on MacBooks, which is really the only advantage x86 in a laptop has.
Moving from x86 not only means switchers to the Mac won't have the "safety net" of running windows, whether natively or in a VM. The vast numbers of utilities for x86 Linux would also become incompatible, people who want to dabble in the occasional game can't reboot to Windows either. We used to dual boot Macs at the school network I ran for various Windows apps. Switching would mean developers would need fat binaries again (apart from MAS distribution) and no doubt it would be another chance for Apple to apply even more OS restrictions. There are a lot of downsides for essentially no upsides. Don't get me wrong - x86 is a crap architecture and if it wasn't for AMD bodging 64-bit support on we'd probably be back to a form of RISC architecture now, like Itanium, but ARM's disadvantages way outweigh the advantages imo.
You're making many false assumptions.
ARM is an ISA, just as x64 is. Neither of these have anything to do with ultimate performance. That has more to do with hardware; silicon design and implementation. ARM is great for "power" saving only because it can be designed to be so, more so than Intel's CPUs can. That does not mean in anyway, that performance cannot be scaled up. Most implementations of ARM-based chips are power-efficient because that was their biggest advantage over anything Intel could produce (while maintaining backwards compatibility). And there's the fact that Windows only ran on Intel, so why bother creating extremely powerful ARM-based CPUs when the world's #1 desktop OS couldn't run on those systems?
You're also making assumptions about the performance capabilities of RISC vs. CISC. While ARM is RISC, there are many non-power-efficienct RISC based CPUs that are extremely powerful (and power hungry) including IBM's POWER architecture and Oracle's SPARC CPUs. In fact, Intel's own Core CPUs have RISC cores with a CISC ISA front-end. This is what allowed them to maintain their high performance while being able to keep the power requirements down and remain compatible.
OS support for ARM is also not an issue today. Microsoft is pushing Windows on ARM fairly aggressively and Linux-arm has almost complete parity with Linux-x64.
Another thing to consider is that most consumer implementations of ARM-based silicon comes in the form of an SoC (System on a Chip). This type of chip prevents the CPU cores from scaling. Breaking out the CPUs cores onto their own chip could dramatically increase performance levels simply from having more space for more cores.
All this conjecture rides on a ‘performance’ narrative - however, there is a lot more to a purchasing strategy than a linear processor/software agenda.
Business looks at stability first. Apple arguably the worst player in the market in this aspect. With no stated roadmap, ultra-secrecy, 7 years between updates, an unending train of failed proprietary standards, spats with software and hardware developers, platform switching, and ‘bold decisions’ that upend workflows and handicap users - there is a reason Apple isn’t a player in the business universe outside of niche.
Their track record from a stability standpoint is not good - anymore. While general consumers can roll with these punches on 1-off purchases - business can’t and won’t.
If recent history is any indication, I don’t believe Apple even cares - as they become almost solely consumer driven in all their offerings. A platform switch doesn’t mean much to the general consumer. Professionals are stampeding out the door anyways after the MacPro debacle - so I don’t see the switch being that difficult for anyone.
To me this move just seems to be made to generate more profit. Arm has a place yes, but not in the Pro market. It makes sense in mobile devices but not a desktop. It makes little sense and Apple should instead research AMD’s professors that are just seemingly getting better and better superseding Intels chips in performance.
If I owned a Mac desktop with an ARM processor I wouldn’t be able stop thinking to myself, it’s got the same processor as my phone has...
And then you have the persuasion of developers to make their X86 platform programmes work on ARM, and no running mobile apps doesn’t cut it on a desktop costing 2 grand or more.
Comments
I did read your post. Like I said, Apple ARM CPUs are outperforming Macbooks, it’s already happened. Way back with the iPhone 6s over the MacBook:
https://thenextweb.com/apple/2015/09/24/the-iphone-6s-outperforms-the-2015-macbook-in-some-tests-which-says-a-lot-about-the-ipad-pro/
Then the iPhone 8 over a MacBook Pro:
https://9to5mac.com/2017/09/22/iphone-8-geekbench-test-scores/
And the X’s A11 out performed then-new MacBook Pros in multi-core:
https://bgr.com/2017/09/14/iphone-x-vs-iphone-8-a11-bionic-benchmarks-macbook-pro/
The iPhone X’s processor is more powerful than the newest MacBook Pro
“the iPhone X’s brand new chip is now more powerful that the Intel chips powering some of Apple’s 2017 MacBook Pros”And the iPhone 11’s A13 is deemed by AnandTech to match Intel and AMD desktop CPUs:
https://www.mactrast.com/2019/10/anandtech-testing-finds-apples-a13-chip-essentially-matches-intel-and-amd-desktop-performance/
“This year, the A13 has essentially matched best that AMD and Intel have to offer.”
...mobile device form factor aside, it is clear that Apple ARM CPUs offer more powerful for less energy. Obviously one day they will hit a wall — that’s a universal truth about any computing bus so it’s pointless to say. The point is today, and in the relevant future, ARM is a better option for Apple’s needs.
The writing has been on the wall.
10-15 years of security and compatibility support? With macOS? Well, good luck.
ARM processors have no such translations because the instructions are already RISC.
Internal hardware translation is of course very inefficient (I used to choose assembly instructions that executed in 1 cycle when programming in assembly for such processors) and apparently done by Intel to be able to change processor backend hardware at will (designers flexibility at the cost of processor complexity and performance).
Compilers that target x86 assembly have a problem choosing specific instructions because Intels hardware backend has a fuzzy execution of specific instruction combinations and cannot be predicted performance wise (logically it should pan out, but I'm not 100% sure about that).
Intels x86's try to do way too much predicting code and scheduling instruction parts, confusing compilers which reduces the efficiency even more.
ARM doesn't need al this.
As a consequence Intels x86 processors burn power while ARM sips it.
Another important performance difference has to do with chip fab production. Intel has its own fabs (which cost billions of dollars) and have long be the feature size champion (size in nm of transistors on chip). They thus where able to produce reasonable fast complex processors because others produced chips with a larger feature size (and chip speed, density and watts all go up, up and down quadratically inversely proportionally to the feature size).
A few years ago TSMC became feature size king, because ASML sells them the best UV waver stepper fabs, thus ending Intels reign.
Regarding this point:
I wonder what fraction of Mac Pro users are using the Mac Pro primarily to run Apple software, especially FCPX and LPX. My hunch is that it's a big fraction -- maybe 80% or more. If so, then Apple can make sure that those apps are ready to go on an ARM-based Mac Pro.
The reason I suspect that FCPX and LPX are the main uses for the Mac Pro is that it seems custom built specifically for that software. The hardware that makes the Mac Pro special isn't the Xeon -- instead it's Afterburner and the GPU modules. In a lot of ways, the current Mac Pro has more in common with those old highly specialized SGI workstations than with other previous Macs. If you just want a lot of CPU power to run non-Apple software, you're probably not going to buy a Mac Pro --- there are FAR more economical (and powerful!) options available (for example, https://system76.com/desktops/thelio-major-r2/configure ).
Just as iOS app developers upload bitcode versions and Apple delivers a version of the app optimized for the processor and device on which it's being downloaded.
Any differences will end up being most certainly bugs attributed to Apple's technology and come down to devs awaiting fixes and providing workarounds to their customers until Apple supplies bug fixes.
I see Zen based CPU's as the transition and then the hybrid. If makes no sense to go ARM in the Pro market systems.
Pancakes anyone?
...macOS releases were every 2 years from 10.4 ~10.7, which may have made things easier all around, and then SJ passed...
https://en.wikipedia.org/wiki/MacOS_version_history#Releases
Windows support lifecycles have been much longer, with Windows 7 spanning 2009 ~ 2020, offering business customers more choice in upgrade timing, training and migration efficiencies.
https://support.microsoft.com/en-us/help/13853/windows-lifecycle-fact-sheet
Current macs will only run TWO versions (2018~2019) of the same vertical application mentioned above.
How this is 'better' for this long time Apple customer and users generally I simply don't understand...
Has insanely great become greatly insane...?
You're making many false assumptions.
ARM is an ISA, just as x64 is. Neither of these have anything to do with ultimate performance. That has more to do with hardware; silicon design and implementation. ARM is great for "power" saving only because it can be designed to be so, more so than Intel's CPUs can. That does not mean in anyway, that performance cannot be scaled up. Most implementations of ARM-based chips are power-efficient because that was their biggest advantage over anything Intel could produce (while maintaining backwards compatibility). And there's the fact that Windows only ran on Intel, so why bother creating extremely powerful ARM-based CPUs when the world's #1 desktop OS couldn't run on those systems?
You're also making assumptions about the performance capabilities of RISC vs. CISC. While ARM is RISC, there are many non-power-efficienct RISC based CPUs that are extremely powerful (and power hungry) including IBM's POWER architecture and Oracle's SPARC CPUs. In fact, Intel's own Core CPUs have RISC cores with a CISC ISA front-end. This is what allowed them to maintain their high performance while being able to keep the power requirements down and remain compatible.
OS support for ARM is also not an issue today. Microsoft is pushing Windows on ARM fairly aggressively and Linux-arm has almost complete parity with Linux-x64.
Another thing to consider is that most consumer implementations of ARM-based silicon comes in the form of an SoC (System on a Chip). This type of chip prevents the CPU cores from scaling. Breaking out the CPUs cores onto their own chip could dramatically increase performance levels simply from having more space for more cores.
With no stated roadmap, ultra-secrecy, 7 years between updates, an unending train of failed proprietary standards, spats with software and hardware developers, platform switching, and ‘bold decisions’ that upend workflows and handicap users - there is a reason Apple isn’t a player in the business universe outside of niche.
If recent history is any indication, I don’t believe Apple even cares - as they become almost solely consumer driven in all their offerings. A platform switch doesn’t mean much to the general consumer. Professionals are stampeding out the door anyways after the MacPro debacle - so I don’t see the switch being that difficult for anyone.
It makes little sense and Apple should instead research AMD’s professors that are just seemingly getting better and better superseding Intels chips in performance.
If I owned a Mac desktop with an ARM processor I wouldn’t be able stop thinking to myself, it’s got the same processor as my phone has...
And then you have the persuasion of developers to make their X86 platform programmes work on ARM, and no running mobile apps doesn’t cut it on a desktop costing 2 grand or more.
I’m not a fan of this idea.