Thinking About x86 Emulation on ARM

Thinking About x86 Emulation on ARM

A new report about Microsoft working on x86 emulation for the ARM-based versions of Windows 10 has Windows enthusiasts justifiably excited for the future. Assuming, of course, that Microsoft and ARM can actually pull this off.

This one will of course require a bit of explanation.

To date, most Windows PCs have run on what we still call the Intel x86 architecture, while modern mobile platforms like Android and iOS run on ARM-based architectures. The two systems are incompatible, and are oriented for different workloads. That is, x86 provides far more performance—and performance scalability–than does ARM. But ARM works better on the mobile computing platforms that are the norm today, offering great performance on lightweight mobile OSes with excellent power management capabilities.

It’s easy to emulate ARM on x86: In fact, developers do this all the time when creating mobile apps in Visual Studio, Xamarin Studio, Android Studio, and Apple’s Xcode. Emulating x86 on ARM, alas, is a sort of holy grail: An as-yet unobtainable goal that would help us bridge the legacy PC past with the mobile/ARM future.

There have been many efforts to emulate x86 on less-capable platforms. One many might not remember was the Transmeta Crusoe processor, which came to life in an early HP Tablet PC called the Compaq TC1000. That pretty but woeful device offered decent battery life for the day, but Windows and its desktop applications ran horribly slowly, and it quickly disappeared from the market. As did Transmeta.

Companies like Connectix and Parallels also offered x86 compatibility virtually to PowerPC-based Mac computers years before Apple itself switched to x86. These solutions let enterprising users run Windows in a window on the Mac desktop so that they could access Windows applications that were not available on their own platform. Like the Crusoe processor, these solutions were slow for the most part, though I suspect some will insist to this day that they ran well. It was certainly better than nothing.

In any event, we’re about to try again.

With the personal computing market having moved into the “mobile first, cloud first” era, the PC is being left behind: Each year, the combination of ARM-based devices—mostly smartphones and tablets—has vastly outsold x86-based PCs. And the gap is getting bigger.

The problem, of course, is that many people still need to access their Windows-based applications. So we’ve come up with a variety of ways to make this happen. The most obvious is virtual access to Windows environments and applications in the cloud, similar to what HP is doing with its Workspace solution for the Elite x3 Windows phone. That Windows 10 Mobile device can’t run Windows desktop (x86) natively, of course. But if the apps run in the cloud and are transmitted to the device—which is connected to a keyboard, mouse, and display—the user gets a reasonable desktop PC experience. Assuming they have a great internet connection and are connecting to a physically close datacenter.

If only ARM could run x86 apps. Which it can’t.

But what if ARM could emulate the x86 instruction set, and thus run x86 apps?

What if, indeed.

According to Mary Jo Foley’s sources, this work is now underway. It requires a few things, of course: ARM chips that are powerful enough to emulate x86 and provide a reasonable level of performance. And an ARM-based operating system that would bring this capability to users.

ARM chipset makers such as Qualcomm will need to figure out the first part of that equation. But Foley says that Microsoft is indeed working on the second. It’s due in the “Redstone 3” release of Windows 10, which is currently scheduled for late 2017. (We’re currently testing “Redstone 2,” which will be called the Creator’s Update and is due in the first half of 2017.)

This capability would allow ARM-based versions of Windows 10—currently Windows 10 Mobile, though there’s no reason that Windows 10 Home or Pro couldn’t make the ride too, in something similar to Windows RT—to emulate x86 code and thus run Windows desktop applications locally on the device. It would make the most sense in Continuum situations, where the phone would just run UWP apps, but desktop apps would come to life when the phone was docked and connected to a keyboard, mouse and display.

This is exciting. But it’s also challenging because our experience with past solutions of this kind was lackluster. And there is every reason to believe that x86 apps will run slowly on ARM. It will likely work better for simple apps than it will for big/complex apps like Photoshop, iTunes, or even Chrome.

But what if it worked? Can you imagine such a thing?

Done correctly, ARM support for x86 could be the final nail in the Windows desktop coffin, the thing that moves us past the legacy code that is now holding us back. I literally just wrote about this problem in The Biggest Problem with Windows Today? The Past. Interesting timing, no?

I don’t want to get too far ahead of ourselves here. It may never happen. It may not happen in time for Redstone 3. It may happen so far out in the future that it doesn’t matter anymore. But if you’ve been pining away for Surface phone or something a bit more nebulous—like Microsoft having an in-house mobile strategy that actually makes sense—this is hopeful news. No promises, yet. But still, hopeful.

So we’ll see what happens.


Tagged with

Share post

Conversation 72 comments

  • 5530

    21 November, 2016 - 2:03 pm

    <p>I think this is the thing that could have saved Windows RT – too bad they’re doing it only now.</p>
    <p>My first thought when I read Mary Jo’s article is that,&nbsp;great, in addition to the instability of what windows mobile already is, they’re going to bolt the entire win32 subsystem on it. But then Centennial came to mind. What if only clean installs of win32 centennial apps were allowed? That might work.</p>
    <p>Anyway, this should make cheap Windows tablets viable again. Intel has completely abandoned their low-end tablet processors, so low-end windows tablets, like the ones that were based on Atom, are dead. ARM Windows tablets can only run Windows Mobile – and UWP apps. X86 emulation will make ARM tablets much more viable, and I do hope to see something like Windows RT on those tablets again instead of Windows Mobile.</p>

    • 2371

      21 November, 2016 - 3:03 pm

      <blockquote><em><a href="#26489">In reply to </a><a href="../../../users/FalseAgent">FalseAgent</a><a href="#26489">:</a></em></blockquote>
      <p>Intel only abandoned there Atom processors for phones, not for tablets and other computer form factors.&nbsp; Although, I will not buy that low end of hardware for any of my computers myself.</p>

      • 5234

        21 November, 2016 - 7:32 pm

        <blockquote><em><a href="#26534">In reply to </a><a href="../../../users/RM">RM</a><a href="#26534">:</a></em></blockquote>
        <p>Cherry Trail Z8000 series are dead, except for embedded models.</p>

  • 2371

    21 November, 2016 - 2:23 pm

    <p>Great if it works.&nbsp; However with&nbsp;smartphones having more internal storage, maybe migrating the Win32 API’s (or at least a core subset of them) to ARM and creating a compiler to take allow targeting of x86 programs to compile for ARM instead would be a better solution.&nbsp; Add the Centennial Bridge to the equation and you now have Win32 apps supported by the Windows store and running natively on ARM.&nbsp; Of cause some capabilities of Win32 would not be supported (likely running as a service and system tray functionality would be out; just for a couple of examples).&nbsp; But it would make a more compelling case than emulation and trying to get an emulated program to&nbsp;work with Continuum.&nbsp; Hopefully Microsoft is looking at that as well.</p>

    • 1377

      Premium Member
      21 November, 2016 - 3:27 pm

      <p><em><a href="#26504">In reply to </a><a href="../../../users/RM">RM</a><a href="#26504">:</a></em></p>
      <p>Consider that Windows 8 RT included a desktop Office 2013 version. Desktop software compiled for ARM processors. Yes, it was lacking add-ins, VBA and a few other bits that full Office for Windows provides, but it showed that it is possible to build desktop software for ARM for Windows.</p>

      • 8578

        21 November, 2016 - 3:47 pm

        <blockquote><em><a href="#26546">In reply to </a><a href="../../../users/hrlngrv">hrlngrv</a><a href="#26546">:</a>&nbsp;UWP as supported on ARM is really just a new version of WinRT. The problem was never that you couldn’t do most desktop apps on WinRT but that that unless all your legacy apps ran on it, it had minimal value to PC users. Software companies wisely waited to see if there was a market for WinRT apps before investing a lot of money to develop for it. There wasn’t and so they didn’t. UWP is in a similar&nbsp;situation. With no viable WP market, a question of value of non-game apps on XBOX and uncertain HoloLens market, most companies are unwilling to invest.</em></blockquote>

        • 1377

          Premium Member
          21 November, 2016 - 6:53 pm

          <p><em><a href="#26553">In reply to </a><a href="../../../users/skane2600">skane2600</a><a href="#26553">:</a></em></p>
          <p>Re HoloLens, how many apps intended for HoloLens would be usable or useful on any other type of hardware? How useful would apps meant for other types of devices be on HoloLens?</p>

          • 8578

            21 November, 2016 - 10:33 pm

            <blockquote><em><a href="#26586">In reply to </a><a href="../../../users/hrlngrv">hrlngrv</a><a href="#26586">:</a></em></blockquote>
            <p>I think we are in agreement. Realistically the greatest synergy would have been between WP and the desktop, although even there the ergonomics of the phone would be problematic. But since WP is barely alive, UWP isn’t all that useful anyway. &nbsp;</p>

      • 7112

        22 November, 2016 - 12:20 am

        <blockquote><em><a href="#26546">In reply to </a><a href="../../../users/hrlngrv">hrlngrv</a><a href="#26546">:</a></em></blockquote>
        <p>The Desktop Office in RT was just another example of how Win32 programs were already portable across CPU architectures. The&nbsp;WinRT API and framework was unnecessary. What was needed was a native scalable Win32 API. Of course, MSFT makes zero $$$ off of Win32 program sales … which was the real reason for the WinRT/UWP/whatever-they-call-it-now&nbsp;closed garden. Remind me again, what is 30% of zero?</p>

        • 1377

          Premium Member
          22 November, 2016 - 12:53 pm

          <p><em><a href="#26625">In reply to </a><a href="../../../users/doofus2">doofus2</a><a href="#26625">:</a></em></p>
          <p>30% of zero helped lead to MSFT’s purchase of Nokia’s phone operations and subsequent write-downs. In this case, 30% of zero wound up negative.</p>

  • 5343

    21 November, 2016 - 2:32 pm

    <p>So, does this mean W10M will finally get x64 version? I mean, isn’t 64bit supposed to be the foundation for this to work?</p>

    • 2371

      21 November, 2016 - 3:01 pm

      <blockquote><em><a href="#26508">In reply to </a><a href="../../../users/tomiknez">tomiknez</a><a href="#26508">:</a></em></blockquote>
      <p>The phone hardware is already 64bit.&nbsp; But my understanding is that there already is a 64bit W10M, but no hardware has required it since no phones have needed more than 4 gig.</p>

  • 1143

    21 November, 2016 - 2:47 pm

    <p>"Done correctly, ARM support for x86 could be the final nail in the Windows desktop coffin, the thing that moves us past the legacy code that is now holding us back."</p>
    <p>Paul, you are absolutely correct! No matter how good the emulation is, it will always be slower than native code. This will lead to more UWP apps being developed.&nbsp; the move to UWP apps will speed up!</p>
    <p>I am beginning to see a small glimpse of what a Microsoft Surface small device will be like.</p>

    • 1377

      Premium Member
      21 November, 2016 - 3:37 pm

      <p><em><a href="#26518">In reply to </a><a href="../../../users/slbailey1">slbailey1</a><a href="#26518">:</a></em></p>
      <p>A few things which would need to precede the death of Win32 and Windows desktop.</p>
      <p>If the killer would be Windows phones with Continuum but with ARM processors, where are they? At the moment, Windows phones have less than 3% worldwide smartphone usage. Would Windows phone purchases explode when a slow x86 emulator became available?</p>
      <p>If the killer would be tablets, where are they? There are NO small Windows tablets running Windows 10 Mobile or using ARM processors currently for sale. Where would the initial demand come from?</p>
      <p>Now we get to the most likely candidate: ARM-based laptops and perhaps immobile PCs. MSFT is ready to throw Intel under the bus? OEMs are just dying to make and try to sell ARM-based laptops which would run x86/Win32 softeare MUCH SLOWER than Intel-based laptops?</p>
      <p>But it’s nice to know Windows fans can find pipe dreams everywhere.</p>

  • 1341

    Premium Member
    21 November, 2016 - 3:01 pm

    <p>This is an interesting hedge. By getting it in place now, Microsoft can simply transition to ARM if when arm’s performance surpasses Intels. However, it’s unclear whether ARM can match the performance of x86 and maintain it’s power advantage, and the physics of it seem a little iffy to me. Sure there is some cruft that x86 has to support that might let a ‘purer’ ARM design break ahead, but I am not convinced that’s it enough to really matter.</p>

    • 8578

      21 November, 2016 - 3:26 pm

      <blockquote><em><a href="#26533">In reply to </a><a href="../../../users/asarathy">asarathy</a><a href="#26533">:</a></em></blockquote>
      <p>That’s the way I see it too. They’d essentially have to implement the entire X86 machine code functionality in software. No doubt Intel has spent decades trying to make any X86 "cruft" as efficient as possible, it’s unlikely that MS can do better in 4 or 5 years.</p>

  • 3229

    21 November, 2016 - 3:09 pm

    <p>Or they could just allow .NET apps not signed by MSFT&nbsp;to run on ARM.</p>

  • 8578

    21 November, 2016 - 3:12 pm

    <p>it’s likely that a Win32 application running on ARM through emulation will consume more power than the same application running on an X86 processor for a given level of performance. Given the unsuitability of the ergonomics of cell phones and even small tablets’ to run conventional PC applications, the value of such a capability is difficult to imagine. But contrary to the article, such a capability if it were embraced, &nbsp;would rather bolster windows desktop applications rather than contribute to their demise. in any case, the Windows desktop isn’t holding anything back for MS. They chose to get into the touch-based mobile business late and then chose a doomed hybrid approach instead of creating a "best of breed" mobile OS.</p>

  • 1377

    Premium Member
    21 November, 2016 - 3:23 pm

    <p>How would a Windows x86 emulator for ARM processors be the final nail in the Windows desktop coffin? The emulator would magically fix software designed for lower res monitors when appearing on higher res monitors? Can Continuum display multiple desktop application windows? If not, there’s a huge pent up demand for systems which can’t display multiple overlapping windows which Windows 8 didn’t capture?</p>
    <p>The big problem with an x86 emulator is how hard it’d push an ARM chip. I’m especially interested in floating point calculation performance. Intel processors don’t run hot just because they were designed to be space heaters. How much slower would ARM x86 emulators be?</p>

    • 5539

      Premium Member
      21 November, 2016 - 4:49 pm

      <blockquote><em><a href="#26543">In reply to </a><a href="../../../users/hrlngrv">hrlngrv</a><a href="#26543">:</a>&nbsp;</em>"Can Continuum display multiple desktop application windows? If not, there’s a huge pent up demand for systems which can’t display multiple overlapping windows which Windows 8 didn’t capture?"</blockquote>
      <blockquote>No Continuum doesn’t support&nbsp;multiple overlapping windows, or even split screen at this point. Windows 8/8.1 did, in the desktop mode, and the same is true for Win 10. In tablet mode they both allowed for multiple windows, based on screen resolution, though not overlapping. I’ve had up to 4 apps running in snapped tablet mode. That includes x86 apps full screened in their columns. They are unfortunately side-by-side, rather than in quadrants, but they work.</blockquote>
      <blockquote>That isn’t particularly a problem though if you are running your X86 apps in a VM though, is it. It could be like parallels running on a MAC. Or, as currently available, a full Win 10 PC accessed via RDP and displayed on a big monitor via Continuum. You are running a virtual Windows desktop environment. With this idea that VM could be running on your device, rather than on a remote piece of hardware or in Azure. This may not be the plan, and I’d be surprised if it could be pulled off anyway. If ARM got powerful enough to run an X86 VM, and X86 apps within it, without significant performance hits, why not just recompile the X86 apps for ARM and let them run natively. If you are running a Win 10 VM, that supports external monitors already, so Continuum isn’t really a factor.&nbsp;</blockquote>

      • 1377

        Premium Member
        21 November, 2016 - 6:47 pm

        <p><em><a href="#26567">In reply to </a><a href="../../../users/SvenJ">SvenJ</a><a href="#26567">:</a></em></p>
        <p>My point about Continuum is that it’s a compromise compared to full Windows 10.</p>
        <p>Citrix and VMWare have had products which provide local UIs for remotely running Windows (and other) software for more than a decade. Clients exist for iOS, Android, Chrome OS, and Linux in addition to Windows and macOS. Continuum RDP would be BFD. Not a major selling point. Same goes for HP’s service.</p>
        <p>VM running on your local device is fine, but if the local device uses an ARM processor, a VM running an x86 OS would require an x86 emulator, and it’d likely be quite slow compared to real x86 devices. Can ARM gain a reasonably fast x86 emulator? Remains to be seen.</p>

  • 8579

    21 November, 2016 - 3:29 pm

    <p>I can recall working for a short while with FX!32 – x86 emulator to run existing x86-compiled Win32 apps on the DEC Alpha processor.</p>
    <p>For the most part,&nbsp;it provided&nbsp;abysmal performance&nbsp;compared to&nbsp;the same apps&nbsp;running on&nbsp;even affordable x86 hardware.</p>
    <p>That experiment didn’t last long, and we gave up trying to run "legacy" apps (native Alpha-compiled apps were seen as&nbsp;the future then) on the mega-expensive Alpha-server, and just relegated them to standard x86 hardware.</p>
    <p>I can’t help but think that emulation is still not going to be acceptable, and that ultimately some type of "co-processor" implementation may be the solution for supporting "legacy" x86 apps.</p>

  • 217

    21 November, 2016 - 3:41 pm

    <p>It’s a cool feature set to say the least, but without mobile traction I don’t see the value proposition – you’ll still need a mobile with all your apps and such anyway.</p>

  • 920

    Premium Member
    21 November, 2016 - 3:43 pm

    <p>Thinking&nbsp;of powerful ARM based tablets here.&nbsp; A phone sized device would be good also, but tablets have the real estate to support a large battery underneath.</p>
    <p>For the road warrior, an ARM tablet would be plenty of power for most daily activities.&nbsp; Then the ability to use the same device for a specific business line of applications is a win-win.&nbsp; No one should expect to be running Adobe Photoshop unless ARM processors make some massive performance leaps.</p>
    <p>For phone sized devices, the X86 could be limited to Continuum use only.&nbsp; This would negate the argument that small touch targets on a phone would make X86 useless.&nbsp; This could be a place where screen size could trigger use cases.</p>
    <p>Looks interesting.&nbsp; I can see why HP is working closely with MS.</p>

  • 5394

    21 November, 2016 - 3:50 pm

    <p>I really don’t think so. They need to develop ARM versions of the same applications. Don’t they develop Mac and Windows versions of the same application. They should also do cross processor versions. Some already do like Office, games, and Evernote. Emulators only delay the inevitable. Microsoft needs to bridge the gap by porting their Win32 apps to UWP apps. Then move on from there.</p>

    • 1377

      Premium Member
      21 November, 2016 - 6:35 pm

      <p><em><a href="#26554">In reply to </a><a href="../../../users/glenn8878">glenn8878</a><a href="#26554">:</a></em></p>
      <blockquote><em>. . . Microsoft needs to bridge the gap by porting their Win32 apps to UWP apps. . .</em></blockquote>
      <p>How long will it take MSFT to make full Office and Visual Studio UWP apps?</p>

  • 5587

    Premium Member
    21 November, 2016 - 4:01 pm

    <p>Screw x86 on ARM, just let me have android apps on my Windows phone without having to put up with a horrid Android handset.</p>

    • 5539

      Premium Member
      21 November, 2016 - 4:55 pm

      <blockquote><em><a href="#26556">In reply to </a><a href="../../../users/DaveNestor">DaveNestor</a><a href="#26556">:</a>&nbsp;</em>Depends on whether you are clamoring for Pokeman Go or Visio.&nbsp;</blockquote>

  • 7063

    21 November, 2016 - 4:25 pm

    <p><em>"when the phone was docked and connected to a keyboard, mouse and display."</em></p>
    <p>On a slightly different topic. When is someone like Microsoft or Google going to invent and push for a wireless docking system with Display/Keyboard/Mouse/Audio all built in?</p>
    <p>Have short range, high bandwidth, dock (maybe at 60 Ghz?) with a single pairing point using RFID or confirmation code. &nbsp;I could see walking up to a 24" display terminal at work or in a library, pairing my phone, sticking it back in my pocket, and then getting a full desktop experience. When I’m done, either hit an "unpair" button or just walk more than 15&nbsp;ft away and it disconnects automatically.</p>

    • 5611

      Premium Member
      21 November, 2016 - 4:32 pm

      <blockquote><em><a href="#26558">In reply to </a><a href="../../../users/DataMeister">DataMeister</a><a href="#26558">:</a></em></blockquote>
      <p>It already works wirelessly if you wish. Continuum works fine via Miracast.</p>

    • 5539

      Premium Member
      21 November, 2016 - 4:54 pm

      <blockquote><em><a href="#26558">In reply to </a><a href="../../../users/DataMeister">DataMeister</a><a href="#26558">:</a>&nbsp;</em>As WP7Mango says, that does already work fine with various Miracast capable displays. My LG TV has it built in, so all I have to do is open Continuum and say connect to the LG and we are off. &nbsp;Now, the caveat is that I can’t do that very long. My Lumia 950 connected to Wifi, both the internet access and the Miracast Point to Point, plus the BT to the keyboard and mouse, does use some battery. Plugged into a single USB-C cable though, I get the same experience and the phone gets charged.</blockquote>

  • 8582

    21 November, 2016 - 5:07 pm

    <p>Take a look at Eltechs and their x86 emulation for ARM:</p&gt;

  • 5592

    21 November, 2016 - 5:36 pm

    <p>’Done correctly, ARM support for x86 could be the final nail in the Windows desktop coffin, the thing that moves us past the legacy code that is now holding us back."</p>
    <p>Wow. Even Apple gave up on thinking that we were in the "Post PC" era.</p>
    <p>Desktop level apps require, oh, an actual desktop (or laptop) style device.</p>
    <p>It isn’t about the processor or we’d see programs as powerful as Office (x86 and x64) and Adobe Creative Suite (x64) developed for tablet and phone targets.</p>
    <p>Satya got it absolutely right. We’re about a new generation of computing where the user and their data are centric and the devices and interfaces connecting the user with that data vary based on the use needed.</p>
    <p>Oh, and we don’t have a processor architecture we "<em>still</em> call the Intel x86 architecture". We call it the Intel x86 (or more accurately now the x64 architecture) because the instruction sets and processor architecture WERE created by Intel and ARE the same as the 80×86 family of chips that started with the 80386 in 1985. Despite attempts by multiple chip makers including Intel and IBM and Apple and quite a few others and Microsoft offering versions of Windows that supported those other architectures&nbsp;NOBODY has come up with an architecture that’s been good enough to replace it.</p>

    • 1377

      Premium Member
      21 November, 2016 - 6:31 pm

      <p><em><a href="#26576">In reply to </a><a href="../../../users/MikeGalos">MikeGalos</a><a href="#26576">:</a></em></p>
      <blockquote><em>. . . NOBODY has come up with an architecture that’s been good enough to replace it. . .</em></blockquote>
      <p>Dunno. Were all the top 100 commercial applications in the early 1990s rebuilt for MIPS, PowerPC or Alpha processors in the late 1990s when MSFT was selling versions of Windows NT 4 for all those processors (in addition to Intel processors)? If not, maybe the architecture/OS doesn’t matter nearly as much as the selection of 3rd party software available for that architecture/OS.</p>
      <p>Next question: would there ever be enough systems using ARM processors and running some version of Windows for ISVs to believe there’d be some economic rationale for porting their WinTel software to ARM?</p>

    • 5234

      21 November, 2016 - 7:27 pm

      <blockquote><em><a href="#26576">In reply to </a><a href="../../../users/MikeGalos">MikeGalos</a><a href="#26576">:</a></em></blockquote>
      <p>Actually, the proper name is x86-64. &nbsp;x64 is just shorthand.</p>

      • 54

        Premium Member
        21 November, 2016 - 8:53 pm

        <blockquote><em><a href="#26587">In reply to </a><a href="../../../users/Waethorn">Waethorn</a><a href="#26587">:</a></em></blockquote>
        <p>It’s also called AMD64, since it was AMD that came up with the 64-bit extensions to the Intel x86 instruction set.</p>

  • 4370

    Premium Member
    21 November, 2016 - 5:45 pm

    <p>Not gonna happen, at least not before many many years. Intel struggles with phones but not in tablets and it’s king in PCs. Emulators are never going to compete with native systems. And recompiling millions of professional softwares for Arm is simply not going to happen. There is nothing MS can do. There will be no Windows Mobile/RT competing with Android/IOS/Chrome. They already made Office for Windows RT and offered it for free… it was not enough…. Paul keeps repeating Windows is less and less relevant… This is only true for basic needs and media consumption. But you forget professionals and businesses. For any professional, any company, many students and high end gamers, X86/X64 is a must and it’s not becoming less relevant at all. The bulk of professional software is developed for Windows and to a lesser extent for MacOS. This is not going to change.</p>

  • 5234

    21 November, 2016 - 7:33 pm

    <p>Paul, your sources are wrong on this.</p>
    <p>The x86 stuff is for Server.</p>

  • 7066

    21 November, 2016 - 7:59 pm

    <p>The problem I see here is why does a regular user need x86 applications or even Continuum for that matter on mobile? I haven’t seen a great use scenario. Sure it sounds and looks cool but DO regular users need it? I’d say no.</p>

    • 5496

      21 November, 2016 - 9:51 pm

      <blockquote><em><a href="#26594">In reply to </a><a href="../../../users/Illusive_Man">Illusive_Man</a><a href="#26594">:</a></em></blockquote>
      <p>They don’t. The only thing people say is that the phone would then have more apps.&nbsp;&nbsp; So what.</p>
      <p>And they say people can then hook it up to a monitor and us it on continuum. But all businesses have desktops or laptops for their workers.</p>
      <p>Phones won’t be powerful enough to run x86 apps. It will probably be slow also.</p>

    • 5641

      22 November, 2016 - 1:55 am

      <blockquote><em><a href="#26594">In reply to </a><a href="../../../users/Illusive_Man">Illusive_Man</a><a href="#26594">:</a>&nbsp;People bought iPads because they thought they could replace PC’s. Same idea for Continuum. Those people (and there are lots of non-techies out there) can have a phone and then dock them for a big screen experience without having to buy / manage a PC.</em></blockquote>

      • 1377

        Premium Member
        22 November, 2016 - 12:43 pm

        <p><em><a href="#26632">In reply to </a><a href="../../../users/James_Wilson">James_Wilson</a><a href="#26632">:</a></em></p>
        <p>Phones running Win32 software, so having a Win32 subsystem, won’t need any management?</p>
        <p>What management drudgery do you believe is common on PCs?</p>

  • 6447

    21 November, 2016 - 8:24 pm

    <p>anything necessary to run on x86 windows is very "legacy", and not going to be rewritten from win32 to the flavor of the month universal platform. "legacy’ is the whole reason windows is still used. I just can’t see how if ARM is 10x slower emulated on intel (ever hear about the horribly slow android emulator? that only changed when they started using x86 images by default instead of ARM) that its not going to be multiply worse in reverse. I believe there are decent intel based phones and tablets already, at least with android, and they just didn’t catch on for whatever reason. What happened to those 8" full intel windows tablets and who exactly is waiting for such&nbsp;devices to have ARM CPUs emulating x86 instead?</p>

  • 5496

    21 November, 2016 - 9:44 pm

    <p>They made a device about 4 years ago, called Windows RT.</p>

  • 7211

    21 November, 2016 - 9:48 pm

    <p>There is one aspect of this that I’d love to have more info on. I’ve always thought that the second GPU&nbsp;in the Surface Book could be a pointer to the thinking of&nbsp;Panos and the Surface&nbsp;team when it comes to Continuum type systems. Namely that plugged in systems get a performance boost&nbsp;in addition to a form factor change.</p>
    <p>Of course the holy grail of that idea is hot-pluggable CPUs.&nbsp;Making any&nbsp;extra CPUs x64 has huge implications for the NT kernel and/or mobility. But if they have working, performant x64-on-Arm emulation, a whole range of simpler solutions become viable, all of which are probably already supported by the Kernel.</p>
    <p>So lets say in Jan 2018 you have a HP Elite x4 or Surface Phone sporting the a&nbsp;new Snapdragon 830 or better and you dock it in a lapdock or deskdock. Whereupon Windows identifies 3&nbsp;or 4 extra&nbsp;identical&nbsp;CPUs and a discrete GPU. You now have a CPU and GPU powerhouse of a system with 4 CPU/32 core SMP and discrete graphics.</p>
    <p>How does that convert into real world&nbsp;performance or practicality? I’m not sure. There is bound to be a diminishing return as more CPUs are added and I&nbsp;don’t know how or if they would improve the emulation performance.&nbsp;Is there a market for such a product? Who knows. But something like this is&nbsp;the sort of envelope pushing I could imagine coming from&nbsp;the Surface team in particular.</p>

    • 1567

      22 November, 2016 - 2:57 am

      <blockquote><em><a href="#26617">In reply to </a><a href="../../../users/Neville Bagnall">Neville Bagnall</a><a href="#26617">:</a></em></blockquote>
      <p>if the dock has CPU’s of it’s own then why bother adding the phone’s CPU to the equation? 😛 wouldn’t such a dock have already been a laptop?</p>

  • 1567

    21 November, 2016 - 11:20 pm

    <p>x86 emulated on ARM always leaves me thinking that in execution it would increase an ARM processor’s power usage to the level of making it pointless. Not sure something headed partly by Microsoft could change that.</p>

  • 7751

    22 November, 2016 - 2:11 am

    <p>What if what they are working on is not an x86 emulation but a hybrid system for a SoC where you have an ARM core and an X86core and then you use the x86 core just to run the&nbsp;win32 applications, in this case you would need some kind of hybrid operating system.</p>
    <p>Just thinking out loud</p>

  • 8593

    22 November, 2016 - 3:34 am

    <blockquote><em><em><a href="#26638">In reply to </a><a href="../../../users/nbates66">nbates66</a><a href="#26638">:</a></em></em>
    <p>Enterprises can instead of purchasing for their employees: a phone ($700) + a laptop ($1500) + a laptop dock ($200) just give them a phone ($700) + a phone dock ($200). It&rsquo;s half a price and needs to manage just one device operating system. It&rsquo;s massive saving in both purchasing/leasing cost and cost of ownership.</p>
    <p>BTW, such dock can be implemented as Neville said or as a device running firmware allowing to run apps using remote desktop or virtual technology. In both cases app installations and files are just in one place.</p>

    • 4370

      Premium Member
      22 November, 2016 - 6:28 am

      <blockquote><em><a href="#26639">In reply to </a><a href="../../../users/Meir Pletinsky">Meir Pletinsky</a><a href="#26639">:</a></em></blockquote>
      <p>Your example does not work, a phone is far from replacing a 1500$ laptop&nbsp; (or even a 1000$ laptop for that matter) and you are supposing that a phone dock (with screen etc) cost the same as a laptop dock… </p>

    • 8578

      22 November, 2016 - 12:08 pm

      <blockquote><em><a href="#26639">In reply to </a><a href="../../../users/Meir Pletinsky">Meir Pletinsky</a><a href="#26639">:</a></em></blockquote>
      <p>Most companies don’t provide a cell phone to employees except executives and sometimes people who do 24 hr support. You also left out the cost of the monitor and assumed that laptops require a dock. When you consider performance a better comparison would be a $700 phone + dock and monitor vs a $500 laptop.</p>

  • 442

    22 November, 2016 - 7:54 am

    <p>To me, articles like this only confirm the limited appeal of RISC architecture outside of the engineering aspects.&nbsp; Yes, limited.&nbsp; Not useless and definitely not in-capable.&nbsp; But limited in that it usually doesn’t bring answers to&nbsp;as many questions and needs that many&nbsp;of folks would like.&nbsp; ARM has grown a LOT of CISC like features as each iteration of the design appears.&nbsp; And, i’d bet that Microsoft has had a hand at a lot of those changes (with of course Google and Apple influence too…)&nbsp; But to attempt to fully emulate a CISC architecture on a predominantly RISC design?&nbsp; Hats off to the programmers that might pull this off, the work will not be easy.</p>

    • 2

      Premium Member
      22 November, 2016 - 8:36 am

      <blockquote><em><a href="#26656">In reply to </a><a href="../../../users/Narg">Narg</a><a href="#26656">:</a></em></blockquote>
      <p>Are we really talking RISC/CISC in 2016? :)</p>

      • 514

        22 November, 2016 - 3:58 pm

        <blockquote><em><a href="#26688">In reply to </a><a href="../../../users/paul-thurrott">Paul Thurrott</a><a href="#26688">:</a></em></blockquote>
        <p>Yes the X86 is a classic CISC ISA with some RISC features added (via microcode) whereas ARM is RISC with some CISC capabilities added in recent designs.&nbsp; However, the underlying CISC/RISC architectures will always offer differential advantages to different kinds of applications.&nbsp; Apps targeted at CISC will never run well on RISC without significant recoding.</p>
        <p>Sometimes with the right kinds of multi-ISA&nbsp;compilers these problems can be partially alleviated without the app developer having to worry overmuch about the target ISA.</p>

  • 5553

    22 November, 2016 - 8:19 am

    <p>25% off Premium eh ? Make it 100% and I’ll bite ?</p>

  • 8607

    22 November, 2016 - 8:45 am

    <p>Here’s a crazy idea… How about a co-processor?</p>
    <p>There are going to be some instruction set that will be easy to port to ARM cpu’s, but there are obviously some that will be a real b!tch. Why not identify which instructions are the hold up, then get with some chip designers to come up with a co-processor design that can be easily added to existing ARM platforms to facilitate rapid x86 execution? You wouldn’t even need to call it an emulator at that point, maybe an x86 expediter.</p>
    <p>Maybe even make the chip so that it could be installed in a sim card, an sd card, or even a USB dongle to retrofit older Arm devices? Maybe make the enabling software open source so that even Android or iOS devices could run x86 apps?</p>
    <p>I mean, if you’re gonna hypothesize, go big or go home!</p>

    • 1377

      Premium Member
      22 November, 2016 - 12:36 pm

      <p><em><a href="#26691">In reply to </a><a href="../../../users/FrankinReno">FrankinReno</a><a href="#26691">:</a></em></p>
      <p>Have you compared the number of discrete connectors just a 32-bit processor has to the connectors USB drives, SD cards or SIM cards have?</p>
      <p>Talk about difficult computer science problems!</p>

  • 593

    22 November, 2016 - 9:44 am

    <p>I actually mentioned this concept on a comment thread recently. Glad to know that it actually being worked on. I’m likely going to wait to upgrade my 950xl until next fall when the newer handsets that can truly take advantage of this are out. I only need to run 1, maybe 2 x86 apps that have been out of development for years and obviously won’t be updated, so this is good news for me!</p>

  • 399

    Premium Member
    22 November, 2016 - 3:26 pm

    <p style="padding-left: 30px;"><em>It may never happen.</em></p>
    <p>Never? It already has happened, back in the 90s on the Archimedes (the original ARM based computers). Performance sucked so an add on X86 board was developed later on.</p>

  • 514

    22 November, 2016 - 3:50 pm

    <p>ISA emulation has never worked — performance is always abysmal.&nbsp; Years back Sun Microsystems tried to do X86 emulation on its SPARC processors (arguably at the time far more powerful than the X86 of the day), and still performance was terrible.&nbsp; Applications are usually extensively performance tuned for the target H/W ISA.&nbsp; Emulation can’t address these optimizations short of tuning the application code.</p>
    <p>Sad to say it’s going to be the case (as far forward that I can see) that the only way to provide decent access to legacy apps on mobile devices is to in fact have an X86 mobile SOC.&nbsp; That doesn’t currently exist, and none of the major X86 chip foundries appear to be pursuing such an animal :-(.</p>

  • 1065

    22 November, 2016 - 5:03 pm

    <p>&nbsp;As I recall Micrisoft did a pretty good job of emulating x86 on the Xbox 360. It’s way beyond my pay grade and my skill level but I believe&nbsp;given enough time and money It can be done. Whether it’s worth doing is a completely other matter.</p>

  • 165

    Premium Member
    22 November, 2016 - 6:56 pm

    <p>What are the chances that Intel is still working on a mobile solution for x86?</p>

  • 677

    Premium Member
    23 November, 2016 - 12:02 am

    <p>Will this save Windows Phone/Mobile. NOT A CHANCE.</p>
    <p>I love WP. Have had one since launch, have converted many to WP. But is it really about x86 apps, NO.</p>
    <p>People want the mobile apps that are on IOS and Andriod. So I don’t really see how this will really help.</p>
    <p>I would love it too but even someone like me that is a WP fanboy sees that I will need to change soon. My only option is Andriod as I will never have anything of Apple.</p>
    <p>Perhaps Microsoft should have left that code in WP that allowed Andriod apps to run and just tried to make it stay secure.</p>
    <p>It will be a sad day for me to give up my WP but I see that day coming sooner rather than later.</p>

  • 8182

    24 November, 2016 - 4:22 pm

    <p>So this is my take:</p>
    <p>If it happens, it will be x86 based UWP apps that will be able to run on both ARM and x86. Apps converted with Project Centennial today can take advantage of certain UWP features as well as be sold through the Windows Store, but they won’t run on ARM. But this would enable them to. And now you have an x86 app, running on ARM, able to act a lot like a UWP app, including running in Continuum.</p>
    <p>You also just opened up to the potential of adding more apps to the Windows Store, removed some of the confusion around which apps can run on which platform, enabled legacy x86 apps to be brought forward to take advantage of live tiles, etc</p>


Stay up to date with the latest tech news from!

"*" indicates required fields

This field is for validation purposes and should be left unchanged.

Thurrott © 2023 BWW Media Group