Microsoft: No Windows 8 ARM support for x86 apps Despite prior hints – and a Redmond developer conference that was all about app compatibility – Microsoft’s Steven Sinofsky has said that software for x86 Windows 8 systems will not run on ARM architecture. For months, Redmond has promised that any code that ran on Windows 7 would run on Windows 8, with the expectation for …
Yes, it's relatively easy to compile Linux stuff to run on an ARM processor. I've got various programs running on both ARM and x86. It's a bit slower on the ARM, but then a multi-core x86 chip consuming almost 100W is going to outperform an ARM running under 5W any day.
If Windows has a well-defined API fully implemented on both platforms and programmers stick to the API then they'll find their programs work on both platforms. I wonder what mark-up there will be for adding the ARM compilation ability to the existing x86 tools? If that's anything above a nominal amount then people won't bother.
I guess that well-defined API is Metro (for GUI) and WinRT (anything else), and of course .NET. Well that's the problem with Microsoft - they have plenty of APIs and majority are "legacy" in one way or another. So they replace them with even more APIs.
Good point about pricing you made there, though.
"I wonder what mark-up there will be for adding the ARM compilation ability to the existing x86 tools? If that's anything above a nominal amount then people won't bother."
Good news. Microsoft have supported both x86 and ARM processors for their embedded OS for many years and it *is* just flicking a compiler switch.
Before everyone gets their knickers in a twist, lets be clear what's being said here:
Windows 8 on ARM systems will not run x86 apps out of the box.
Is that surprising? No. Would Apple ever try to get Mac software to run directly on the iPad? No chance! If MS did provide out of the box support for old software, would most reviewers and users complain about how poor performance is and battery life is for Windows 8 on ARM systems? Definitely - even though it's in no way MS fault.
There will be ways round this:
1) Cloud based virtualisation.
2) Server based virtualisation.
3) Hyper-V (when it's ported, which it probably will be) virtualisation.
4) Cloud-based replacements (e.g. Office 365).
Not only this, but lets face it... how many ARM computers are going to have the grunt needed to run x86 apps properly anyway? Anything with any real hardware demands will need to be re-written. CAD software on what is essentially a netbook?
Businesses and home users need to get real - if you expect to run old software designed for a single-core P3 or P4 PC with permanent mains supply and dedicated graphics processor etc on a 1GHz quad-core tablet / budget laptop smoothly without sucking the battery dead in five mins you're deluding yourself. Look at Motorola's experience with Android - 7/10 returns for Motorola handsets are (apparently) down to the user installing poorly written apps that make the phone unstable and kill the battery. And those apps are deigned to work on the hardware in question!
And let's get off this whole app thing. The big selling point of W8 is that it is universal user experience, not that it has loads of apps. Most home users won't want to run old applications on their W8 tablets any more than they'd want to run them on their smartphone. What MS are offering is the same way of accessing all your cloud-based stuff from all devices, be it Xbox, phone, PC, netbook, smartphone, virtual server session etc. Yes, lack of apps will slow sales for tablets - but I'd suggest this is short-term only and in the business sector is already more than compensated for by native support for all the MS products that most businesses already use.
So, realism is called for. MS don't help themselves, that is true, but posters on El Reg should know better. Will Windows 8 become the dominant OS on tablets? Not for a while. Will ARM become the dominant architecture on computers? Also, not for a while. Remember, the RISC / CISC debate back at the turn of the millennium? The only way ARM will ever seriously take off for productivity computing is when all those old x86 apps are rewritten or moved into the cloud - irrespective of what Windows 8 does or doesn't run.
This is not a Windows 8 issue, it's an ARM issue.
> The big selling point of W8 is that it is universal user experience
What are you saying ? That it has every different GUI: XP style WIMP legacy, ribbon style, WP7 touch, Metro; so it must therefore be 'universal' ?
Or are you saying that _everyone_ will be forced into the 'one true way' of metro and therefore it will be 'universal' in that way ?
W8 is a mish-mash of several incompatible UIs tossed together as MS tries to catch up to the way the market has moved over the last few years.
> Yes, lack of apps will slow sales for tablets
Only of Microsoft tablets. iPads, Androids and others will still sell well. One reason that WP7 failed to sell because WM6.x apps could not run and there was a disjoint between those that used WM6 (corporates) and the market target of WP7 (consumer, twitterers).
W8 falls in the same boat. There are no substantial apps for W8 tablets and no point in using Desktop apps on tablets, or for using Metro touch on desktops. XP and Win7 users will continue to use those, tablet buyers will buy iPads and Androids.
"lets face it... how many ARM computers are going to have the grunt needed to run x86 apps properly anyway?"
This year's ARM tablets can wipe the floor with the sort of "heavy" desktop machines that were available 5 years ago. There were *plenty* of Windows apps that managed with those "limited" resources. Most of them are still in use. Most of the horsepower added by Intel and AMD in the last five years has gone into running the OS and the desktop, not the apps.
This post has been deleted by its author
Trying to support all the legacy applications has hobbled innovation at microsoft and given us crappy operating systems (especially compared to OSX).
Now they're doing the right thing and everyone's jumping on them for not making the kind of mistakes we used to slag them off for.
Seriously, what is the value of porting the entire windows Ecosystem to ARM by means of some lame intermediate layer that will effectively kill the "Windows on ARM" concept by making slow and cumbersome?
M$ seems to be taking a page from Apple's strategy book here:
Applications that are really relevant will get ported. People who need to use legacy x86 applications (i.e. primarily corporate users) will choose to run windows 8 on x86 PC's and tablets.
As Steve Jobs would say "it's not that big of a deal"...
> what is the value of porting the entire windows Ecosystem to ARM
The value is in regaining control of the OEMs. The ARM version of W8 (Wait? Weight? Windows 8 my homework?) is not about being sold to users it is about control of the OEMs and killing of ARM tablets.
When the netbooks first came out they were cheap machines with DVD screens, SD card 'disk' and zero cost Linux OS. MS 'persuaded' the OEMs that if they did not install Windows on them then they would lose their 'discounts' and 'advertising partnerships'. ie would have to pay MS $millions more. MS even revived XP for this.
Now the OEMs are making ARM netbooks and tablets with Linux (Android, WebOS) installed. MS can't wave loss of discounts yet, not until a version of Windows _could_ be run on them. When W8 ARM is shipped then the OEMs will fall into line and will put W8 on them, or payup millions to MS. These won't sell because of
lack of legacy apps. The OEMs will switch to Atom tablets and laptops with W8 so legacy will work. MS will regain control of the market. Or such is the plan.
You're being very silly.
No one with a clue is suggesting that. Not yet anyway.
What people with a clue can see quite clearly is that 95% of what 95% of PC users do with PCs uses a lot less "grunt" than the PC vendors like to sell. But that's been OK so far because people have been convincable that they need a Window box else it's not really a PC.
Unfortunately for the old ways, the genie is now out of the bottle and Joe Public can quite clearly see courtesy of Apple, Android, and others that what Joe Public mostly needs, and even what Joe IT User mostly needs, is not a deskbound box or even a laptop, to be incompatibly replaced every two or four years, but something far more basic (and less profitable for the suppliers, but hey...).
And when Jo Finance Director at Enterprise Customer 1 susses this too, based on the stuff her friends and family use at home, I wouldn't like to be the account manager at the desktop PC supplier, or their IT department contact...
The times, they are a'changin.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
