Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×
Internet Explorer Microsoft The Internet

IE11 To Support WebGL 111

Posted by Soulskill from the better-late-than-never dept.
mikejuk writes "The biggest problem with IE10 as far as modern web apps go is its lack of WebGL support. Now we have strong evidence that IE11 will support WebGL. A leaked build of Windows 'Blue,' aka Windows 8.1, also contained an early version of IE11. Web developer François Remy decided to see what it was hiding and found that there were WebGL APIs, but they were non-functional. Rafael Rivera, who writes the Within Windows blog, dug a little deeper and discovered the registry keys that have to be changed to enable WebGL support. Apparently the API works so well that you can take existing WebGL programs (with OpenGL shaders) and just run them. As the implementation also supports DirectX HLSL shaders, it seems reasonable to guess that the implementation maps OpenGL to DirectX, thus avoiding Microsoft having to endorse OpenGL use."
This discussion has been archived. No new comments can be posted.

IE11 To Support WebGL More

IE11 To Support WebGL

Comments Filter:

  • April Fool's... (Score:5, Funny)

    by sstamps ( 39313 ) on Tuesday April 02, 2013 @04:31PM (#43342385) Homepage

    ..was yesterday.

    Just like Microsoft.. a day late and an API short. :P

  • Will I have to upgrade to Windows Blue for this Internet Explorer 11, or will Windows 8 be enough? Somehow I suspect it won't be the latter.

  • IE11 is getting pretty awesome. I'm not a fan of WebGL but I do want my browser to support as many open standards as possible.
    • IE10 has already quite nice HTML5 video playback, also. Much less resource-intensive than of other browsers.

      • Re: (Score:2, Redundant)

        by account_deleted ( 4530225 )
        Comment removed based on user account deletion
      • Does IE 10 support the WebM plug-in like IE 9 did, or is its <video> element MP4-only? I haven't been able to find solid evidence either way, nor do I own a Windows 8 box on which to try it myself.
    • As a 3d graphics dev (primarily OpenGL these days) I reckon it's great that IE is supporting WebGL, it might not be a standard of any sort yet but obviously some sort of access to 3d graphics hardware from the browser is inevitable and the ubiquity of OpenGL makes it - or at least an API based closely on it - the obvious choice, even if on some platforms it's a wrapper for whatever native 3d API that platform uses. I'm still concerned about the security and stability implications of exposing the most volati
      • It is a standard. At least in name. The Khronos group manages it here [khronos.org]. The "the security and stability implications of exposing the most volatile piece of computing hardware through the browser" is exactly why every browser exposes WebGL through a wrapper/translator library that acts as a validator to prevent bad behaviour. WebGL-based exploits have been shown in the past.

        • It is a standard. At least in name. The Khronos group manages it here [khronos.org].

          It's a spec...but it seems to be turning into at least a defacto standard.

          The "the security and stability implications of exposing the most volatile piece of computing hardware through the browser" is exactly why every browser exposes WebGL through a wrapper/translator library that acts as a validator to prevent bad behaviour. WebGL-based exploits have been shown in the past.

          The issue is that video driver crashes are one of the most common causes of system crashes, so driver stability is a major issue and not going to be fixed by a translation layer. The other is security, especially given you are writing to GPU memory, having that sort of access to hardware through the browser is potentially a huge security issue with the ability to exploit driver bugs, cross domain image hacks using fragment shaders, de

    • It could be the awesomest ever, but in my Windows 7 shop, it's irrelevant.

      Thank goodness there's Chrome, Opera and Firefox.

      • IE10 is out for Win7. Have you seen any documented evidence that IE11 won't be available for Win7? I mean, I have no evidence either way, but since they're now releasing browser versions on a more accelerated schedule it seems likely they'll support them on the current generation most-popular Windows variant.

      • Re: (Score:2)

        by Merk42 ( 1906718 )
        If IE11 is released before January 2015 it should be available for Windows 7. That's what differentiates "Mainstream Support" from "Extended Support"; new features. http://support.microsoft.com/lifecycle/?p1=14482 [microsoft.com]

  • What a silly statement (Score:5, Interesting)

    by Sycraft-fu ( 314770 ) on Tuesday April 02, 2013 @04:36PM (#43342425)

    "it seems reasonable to guess that the implementation maps OpenGL to DirectX, thus avoiding Microsoft having to endorse OpenGL use."

    No, more likely MS doesn't want to have to rely on vendors providing a working OpenGL driver, since that can be problematic (looking at you here ATi). If you have an accelerated Windows driver, a WDDM driver, it has DirectX support. That is how it works, just part of the spec. OpenGL, however, is an addon. Vendors can provide an OpenGL driver, or any other API they like, if they wish but it isn't an inherent part of the driver. They can choose not to provide them, or can provide broken ones.

    So, would make sense for WebGl support to have something that does translation, so it works as long as you have a WDDM driver installed.

    • Re:What a silly statement (Score:5, Informative)

      by gigaherz ( 2653757 ) on Tuesday April 02, 2013 @04:42PM (#43342485)

      Even Firefox uses Google's ANGLE to translate WebGL to Direct3D.

      From the ANGLE site [google.com]: "The goal of ANGLE is to allow Windows users to seamlessly run WebGL and other OpenGL ES 2.0 content by translating OpenGL ES 2.0 API calls to DirectX 9 API calls. "

      • You don't want to rely on a host OpenGL driver since OpenGL isn't the native interface for Windows.

        Heck translation might be good even on a GL system, since ES isn't directly compatible with normal OpenGL unless you have a 4.1 or better setup which requires a fairly new card (GeForce 400 or newer in nVidia's case). I don't know of any Intel GPUs that do GL 4.1 yet, even Ivy Bridge is still 3.1.

        So regardless of platform, it could make a lot of sense to implement it as a translation system, and then just choo

      • > Even Firefox uses Google's ANGLE to translate WebGL to Direct3D.
        Whether that's a smart thing to do is another thing, because then they have a different code base on Linux and Mac and OpenGL was designed to be cross platform.

        • Not really. The codebase is OpenGL ES for all of them, they just need to change the translation backend of libEGL in each platform, which would be happening either way, ANGLE or not.

          • OpenGL ES is a subset of OpenGL, in other words it's fairly trivial to get an OpenGL ES program to work on OpenGL. EGL is a complementary library, but not part of the OpenGL standard

    • No, more likely MS doesn't want to have to rely on vendors providing a working OpenGL driver, since that can be problematic (looking at you here ATi).

      Particularly since it's an OpenGL ES driver, Chrome and Firefox do the same thing i believe.

      • Comment removed based on user account deletion

        • Re: (Score:3)

          by ais523 ( 1172701 )
          Basically the problem is that OpenGL has a lot of old cruft in that people have been trying to get rid of for a while, that made sense at the time but nowadays only exists for backwards compatibility. OpenGL ES is gaining in prominence because it looks like it might actually be a chance to make a clean break with OpenGL's past.
          • Comment removed based on user account deletion

            • OpenGL is not an opensource project. It's an open standard managed by a group that's composed of experts from hardware and software corporations, and sponsored by those (and maybe other) companies. The group's name is Khronos. and yes, THEY decide what OpenGL looks like, and who can use the OpenGL logo. Of course you could create a LibreGL, that bases itself on OpenGL, but flows in a different direction, but then you will NOT be sponsored by all those large corporations, who will most probably ignore your e

            • Re: (Score:2)

              by mwvdlee ( 775178 )

              Forking is fine for code, slightly less so for standards.

              Car analogy;
              Forking code is like copying a car design and changing it slightly; it may look and work differently but can still drive on all roads.
              Forking standards is like copying the road design and changing it slightly; no existing car can drive on the new roads and future cars can't drive on the old roads.

              (Actually a train analogy would have been a better fit, since they actually have forked the trainrails design and have ended up with those proble

        • Now correct me if I'm wrong but isn't OpenGL ES a much wimpier subset of OpenGL cooked up for cellphones and consoles NOT for desktops? That was the way I had always heard it explained which if so makes all the hubub just one more "ZOMFG we can be like the iPhone!" as far as I'm concerned.

          Kind of, ES doesn't have a lot of the - now deprecated or removed - cruft in OpenGL like the Begin/End calls in favor of the more modern techniques. It isn't quite as cutting edge as the full OpenGL desktop version that you get in the latest desktop hardware, there are some features like geometry shaders that it omits - i don't think there is support for bindless graphics either - but it has pulled in features like multiple render targets and a lot of new texture format options. Generally ES gets the most u

    • So, how is it that Rage and Quake, and most 3D creation apps use OpenGL without any problems?

  • Another ASP debacle (Score:1)

    by Anonymous Coward

    Yes, let's include another gaping security hole in Internet Explorer that allows direct access to a system's hardware from the browser!

    • Re:Another ASP debacle (Score:4, Insightful)

      by Millennium ( 2451 ) on Tuesday April 02, 2013 @04:51PM (#43342557)

      You mean like the alleged hole that supposedly left Microsoft with no choice but to remove NPAPI plugin support from IE back in the 1990s?

      Frankly, this is huge. Direct3D is probably Microsoft's second most effective tool for locking-in users (behind MSOffice) and the single most effective tool for locking-in developers. To officially support its open competitor -and in a way that would allow apps (read: games) to actually be played on other platforms, no less- is uncharacteristic of them, to put it mildly. Are they so afraid of WebGL's potential that they see simply supporting it as less risky than embrace-extend-extinguish?

      • Re:Another ASP debacle (Score:4, Insightful)

        by NatasRevol ( 731260 ) on Tuesday April 02, 2013 @05:11PM (#43342717) Journal

        The other option is that they're in step one of the embrace-extend-extinguish dance.

        • I don't get why no one else has called this... They're actually already at step 2

          1. Embrace WebGL
          2. Extend WebGL by supporting HLSL shaders
          3. ...

      • Comment removed based on user account deletion

        • Uhhh it OpenGL ES and its being supported in the BROWSER which unless they changed things by default is running in low rights mode anyway...yeah I really don't think anybody is gonna be seeing performance that rivals a DirectX game installed on the hardware Hoss.

          It will be a WebGL wrapper around DirectX, just like Firefox and Chrome do on Windows, so I don't expect there will be anything to hinder performance there.

          • Comment removed based on user account deletion

            • Uhhh...any time you add layers to something performance is gonna suffer friend, if it didn't everything from your browser to your video player would just be throwaway VMs so bugs would be a thing of the past.

              The only thing that matters is how much it suffers in the particular case we're talking about, the answer is that it is negligible. I'm not sure why you're comparing to VMs, a VM is a lot more heavyweight than a simple API wrapper. Even around 15 years ago we had 3d API wrappers (most notably Glide OpenGL wrappers) that had negligible performance impact and these days with so much work done in shader code that is pre-compiled the performance impact is even less.

            • Re: (Score:2)

              by mwvdlee ( 775178 )

              Uhhh...any time you add layers to something performance is gonna suffer friend, if it didn't everything from your browser to your video player would just be throwaway VMs so bugs would be a thing of the past.

              Not necessarily.

              Additional layers hide complexity. In some cases, they hide a sort of complexity that is too specialized for most developers to be able to handle well.
              Stuff like hardware drivers, the TCP/IP stack, runtimes and even the BIOS are abstraction layers too.

              In theory, a smart programmer can get much better performance out of coding his entire 3D hardware accelerated game directly on the BIOS in assembler.
              In practice, I dare say no programmer is smart enough to actually outperform all those abstrac

    • Yeah, looks like they might not be so worried about those security issues [duckduckgo.com] after all. Or maybe they only come in play when you turn your free non-WebGL'd Windows Blue into Windows Blue Super-Clouditized And Also Actually Plays Games Edition with monthly subscription.

      ...but seriously--long side rant follows--their Windows 8 insolence motivated me to get a laptop and make it Arch Linux-only (I have a desktop with Windows 7 and Arch in some crude tandem dualboot way, but I wanted to try non-secure-boot EFI and

      • ...and whaddya know, I open Steam juuuuust after the comment to check, and as if to vindicate me I see the GUTS editor and other fun stuff for TL2 were released [steampowered.com]. (Sorry my comments ended up becoming a whole big TL2 ad, just saying.)

      • There's a huge difference between being concerned about security issues and delaying the inclusion of potentially security-risky features, and omitting them entirely even after having enough time to thoroughly examine the risks, develop mitigations, test extensively, and so forth.

        Microsoft did the former (they were concerned - and probably rightly so - about allowing incredibly untrusted code [anything on the web] to interact with incredibly delicate system components [video drivers]).
        Microsoft is not doing

  • May I assume that Windows 8.11 for Workgroups will be out soon?

  • IE's problem isn't particular support in one area or another, it's that they take years to release new versions, thereby falling behind other browsers. They are never that bad on release, but fall behind other browsers that strive to remain current. But IE8 still has ~10% marketshare, and that was released 4 years ago. IE6 finally became negligible 2 years ago, after 10 years.

    But they're starting to get better, IE10 was the least bad version and it only took them ~17 months to release it after IE9.

  • " LOOK how FAST ie10 is. You can play a GAME on it. TOUCHSCREENS!!!!. We are FINALLY SECURITY!"
    Man all those ads were quite annoying, and rather false.

  • WebGL support? (Score:1)

    by Anonymous Coward

    I'm sure both users of IE11 will be thrilled to hear it.

  • I mean, really, Microsoft has been so discredited in the Internet and mobile space, they should be embarrassed to even continue trying. And don't even get me started about Windows 8....

    .
    Give it up, Microsoft. Your time was Windows 95, and that time has passed.

    • If being discredited was a reason for anyone to stop trying, then no one would come out of their tiny holes. Fall, learn from your mistake, stand up, and try again.

  • The biggest problem with IE10 as far as modern web apps go is its lack of WebGL support

    wrong. the biggest problem with ie10 is ie10.

    what's the meme for crap like this, "first sentence made no sense whatsoever;dr"?

  • Although I do hope that IE11 gets released to Windows 7.

    Here is my thinking - I love WebGL, but I don't see it really taking off unless IE supports it (granted, IE is loosing market share, but that's another topic). However, Windows 8 seems to be a bigger bust for Microsoft than Vista and ME was. So, if IE11 is exclusive to Windows 8, that still means that the default webbrowser used by a good portion of the web users won't support it.

    Probably why many webpages still look like they did 10 years ago, website

  • .. which means that as usual IE will be holding back web development for another 5 years. I am being serious. This is an ongoing problem for anyone who developers client-facing sites especially when long-term support is part of the requirement. Most companies simply can't justify having one group of engineers working on WebGL and then another group working on some other IE-only implementation... they do not have resources like this. You have limited resources and need to choose one solution that works acros

Slashdot Top Deals

Don't tell me how hard you work. Tell me how much you get done. -- James J. Ling

Close