×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

IE 11 Getting WebGL, SPDY/3, New Dev Tools

Soulskill posted about 10 months ago | from the growing-up dept.

Internet Explorer 119

rescendent writes sends this report about new features in Internet Explorer 11: "Microsoft released Windows Server ("Blue") to MSDN subscribers today, ahead of the BUILD conference later this week in San Francisco. The build provides us a number of clues as to what we will see in the official Windows 8.1 (Blue) preview. The server build number is 9341, the Windows 8.1 preview build will be: 6.3.9431.winmain_bluemp.130615-1214. IE11 scores 351/500 + 2 bonus point, and 25/25 for WebGL. Since this is a server build, the score may be a little higher than IE11 on Win 8.1, but this confirms WebGL for IE11. IE11 WebGL Conformance Test Results: 14,748 of 20,509 tests pass (71.9%). Many things seen in the Server 2012 R2 preview will also show up in the Windows 8.1 preview."

cancel ×
This is a preview of your comment

No Comment Title Entered

Anonymous Coward 1 minute ago

No Comment Entered

119 comments

WebRTC? (0)

Anonymous Coward | about 10 months ago | (#44109441)

Still no mention of WebRTC?

Re:WebRTC? (0)

Anonymous Coward | about 10 months ago | (#44110233)

Of course not, because Microsoft have come up with their own alternative solution, which they're still pushing for.

Frankly, I'm surprised WebGL made it in to IE before they had a chance to implement a WebDirectX.

Re:WebRTC? (1)

Lennie (16154) | about 10 months ago | (#44112187)

Maybe why they aren't bothered to much, as far I know all WelGL implementation on Windows use DirectX.

Re:WebRTC? (1)

binarylarry (1338699) | about 10 months ago | (#44112257)

OpenGL and Direct3D do the same things, the same way, on the same hardware.

So it shouldn't really matter how the OpenGL API is implemented, as long as it's done competently.

Given Microsoft's past of half implementing things and blaming the specification for their problems, I suspect IE 11 WebGL may be botched purposely.

Re:WebRTC? (0)

Anonymous Coward | about 10 months ago | (#44114317)

Most standards-compliant browser available. AC still complains.

server build? (1)

Anonymous Coward | about 10 months ago | (#44109467)

why would that make a significant difference?

Re:server build? (2, Funny)

gl4ss (559668) | about 10 months ago | (#44109487)

why would that make a significant difference?

well, because it's SERVER it has a much stabler kernel and supertuned internals and therefore is better for running an irc client... that's what one ms fanboi once told me anyhow.

seriously though, dunno, maybe it boots straight to desktop and has couple of flags so it will run ms servers with higher connection counts..

Re:server build? (-1)

Anonymous Coward | about 10 months ago | (#44109499)

seriously, if you believe the incoherent babble you just wrote.....you are one dumb mutherfucker!

Re:server build? (0)

Anonymous Coward | about 10 months ago | (#44110115)

Reading comprehension is not your strongest point, is it.

Re:server build? (5, Interesting)

_merlin (160982) | about 10 months ago | (#44109503)

Server installations by default have the scheduler configured to prioritise services over UI applications, and to provide more deterministic scheduling at the expense of responsiveness to user input.

Re:server build? (0)

Anonymous Coward | about 10 months ago | (#44110105)

As I understand it though the kernel of the Windows client and Windows Server are basically identical.

The settings can be changed on one to match the defaults of the other.

The real difference is in the bundled userland and functionality.

Re:server build? (3, Informative)

_merlin (160982) | about 10 months ago | (#44110185)

You are correct - you can get the same results by adjusting settings and policies to match on pro and server. It's only the defaults that differ. Also, yes the included userland functionality is the big difference.

Re:server build? (2)

KiloByte (825081) | about 10 months ago | (#44110481)

Kind of like /proc/sys/vm/swappiness, which decides that if you copy over 10GB junk to a backup drive, all your running processes get swapped off to make space for file cache because "it improves throughput".

Re:server build? (4, Insightful)

drinkypoo (153816) | about 10 months ago | (#44111097)

Swapped? What is this swap? I have 8GB of RAM, I neither need nor have swap. For people with too little memory to run anything without it, it's mandatory. To everyone else, it's just a way to slow down your computer.

Re:server build? (-1)

Anonymous Coward | about 10 months ago | (#44111321)

Lol. 8gb. You must not do much of anything. What a faggot.

Re:server build? (0)

Anonymous Coward | about 10 months ago | (#44111913)

Swapped? What is this swap? I have 8GB of RAM, I neither need nor have swap. For people with too little memory to run anything without it, it's mandatory. To everyone else, it's just a way to slow down your computer.

The only valid reasons to disable swap are for security or to reduce wear on SSDs.

Disabling swap causes less unused memory to be allocated to buffers and storage caches which only hurts performance.

Re:server build? (3, Insightful)

drinkypoo (153816) | about 10 months ago | (#44111987)

The only valid reasons to disable swap are for security or to reduce wear on SSDs.

Bullshit, and also bullshit.

Disabling swap causes less unused memory to be allocated to buffers and storage caches which only hurts performance.

It would hurt performance, but I have more RAM than I actually need, so it doesn't. Meanwhile, swapping out so that you can have more room for buffers hurts performance, especially if you wind up having to swap back in.

Re:server build? (1)

Krojack (575051) | about 10 months ago | (#44112669)

I have 12 gigs of RAM and still get about 500 megs of swap used after a long day of normal usage. Sure it's reserving 12 gigs of space for swap but never uses more than 500 megs. Long idle processes will still get pushed off to swap. It's so little usage that you won't ever notice assuming you have your overall system setup correctly.

That said, disabling swap if you run an SSD (or move the swap to a non-SSD drive) is wise because like I said, it will still get used no matter how much RAM you have. The laptop I'm using right now has 4 gigs of RAM. 2.12gigs is used yet 3.8% of 4,054 MB swap is in use. I just booted my system up about an hour ago so it's still low usage for now.

Re:server build? (2)

drinkypoo (153816) | about 10 months ago | (#44116469)

If you have SSD but you still need swap you can set swappiness to 0, and in practice you will rarely swap. But you don't need swap unless you're short on RAM, and RAM is pretty cheap these days and has many uses. I built this system when the Phenom II was still a pretty new chip, so 8GB was actually pretty good for a desktop.

Re:server build? (1)

flyingfsck (986395) | about 10 months ago | (#44111983)

LOL. How will swap slow your machine down if swap isn't used? Maybe you should get a real computer.

Re:server build? (1)

Agent ME (1411269) | about 10 months ago | (#44114557)

The only time my machine ever uses swap is when a program has a memory leak and takes much more memory than it needs. Usually this is garbage data and I want the program killed. Without swap, the out-of-memory killer would do exactly that for me. With swap, my machine desperately tries to keep it running by swapping out everything I'm trying to use, and wasting 5 minutes of my time.

Re:server build? (1)

drinkypoo (153816) | about 10 months ago | (#44116261)

It's amazing how many people don't get this, but posted (mostly anonymously) in response to my comment anyway. If you have more RAM than you're using then the only time you're going to swap is when something goes badly wrong, and then it just leads to a lot of disk thrashing before either the OOM killer comes into play or the kernel panics, whichever comes first. Either way, I'd rather not sit there trying to gain meaning from the HDD LED. 8GB may not be a "lot" but it's more than I need to run several sizable applications at once without running out.

Re:server build? (0)

Anonymous Coward | about 10 months ago | (#44116743)

You're amazed because you're the one who "doesn't get it", so the people who know what they're talking about all seem wrong to you.

You can have 128G of memory and swap will still be used. It will page out unused memory so that more of your system RAM can be used to actively speed up the system. By disabling swap, you're preventing your kernel from using its memory fully because it needs to keep all of the inactive memory in RAM instead of paging some of it to disk.

It matters more on low memory systems because when you're running on 512MB even a few MBs paged out to disk can make a difference, but on big systems there's still no reason to disable swap other than to reap the disk space. Even if swap is slow you will not notice performance degradation because the paging is done when the area of memory is inactive.

Re:server build? (1)

drinkypoo (153816) | about 10 months ago | (#44117003)

It matters more on low memory systems

No, it only matters on low memory systems. It doesn't make a big difference whether I have 2GB of disk buffers or 4GB of disk buffers if I have a SSD. And meanwhile, swap will get used when from my point of view it really shouldn't, and then when I switch tasks I'll wind up swapping again. I'm glad to be rid of all that nonsense, and you're imagining things if you think that letting your kernel swap out a couple hundred megs of crap that never should have been loaded is going to make any big difference in performance on a modern system, unless its impact occurs right when you don't want it to.

Re:server build? (0)

Anonymous Coward | about 10 months ago | (#44112139)

Another moron who doesn't know how swap works and acts like tough shit because he has an insignificant amount of memory.

Off yourself.

Re:server build? (0)

Anonymous Coward | about 10 months ago | (#44115117)

I have 32GB of memory, and 18GB of swap. I'm using only 5MB of swap at the moment, so I'm not sure how swap is slowing down my computer?

It sure is handy though. If a program starts using too much memory, it doesn't just crash (which is good, because by the time that happens the program might have been running for several days, and it would be a shame to have to restart it).

Re:server build? (-1, Troll)

Anonymous Coward | about 10 months ago | (#44109625)

Real pros compile their kernels with the /server flag of gcc. It makes their internet faster on the localhost.

Re:server build? (2, Informative)

jrumney (197329) | about 10 months ago | (#44110165)

It used to be the case that non-server versions of Windows had limits on simultaneous connections in the TCPIP stack, which could affect web browser benchmark performance.

Re:server build? (0)

Anonymous Coward | about 10 months ago | (#44110067)

If the server is configured as a terminal server then users need to be able to run desktop applications.

Terminal Server CALs; WebGL over RDP (1)

tepples (727027) | about 10 months ago | (#44112373)

If the server is configured as a terminal server then users need to be able to run desktop applications.

For one thing, I'm not sure how much use the terminal server gets in practice other than a single administrator remotely logging in at once. Microsoft's standard operating procedure has been to charge excessive rates for the client access licenses needed for terminal server operation. For another, I'm not sure how well WebGL will pass through an RDP connection.

Obligatory (-1)

Anonymous Coward | about 10 months ago | (#44109517)

But they didn't listen to us, they didn't bring back the start menu!

There, it's done, it comes up in every Windows 8-related thread.

Re:Obligatory (0, Troll)

Vanderhoth (1582661) | about 10 months ago | (#44110409)

How about, windows 8 is an even stupider idea on a server than it is for a desktop?

You know for a little variety.

Amazing feat. (-1)

Anonymous Coward | about 10 months ago | (#44109585)

That's amazing given that the opengl API distributed with windows is stuck at v1.1 last I checked. #gottafetchthempointers

Re:Amazing feat. (0)

Anonymous Coward | about 10 months ago | (#44110265)

GPU drivers however upgrade that. Current NVIDIA+AMD drivers will raise that to 4.3.

Re:Amazing feat. (0)

Anonymous Coward | about 10 months ago | (#44110373)

And that is irrelevant. All that matters is what the drivers themselves implement. You can program to any OpenGL version as long as the drivers support it (which they do).

Re:Amazing feat. (0)

Anonymous Coward | about 10 months ago | (#44110463)

Not really. Everyone and their mothers implement WebGL with DirectX on Windows, anyway (using ANGLE) – due to the shotty state of OpenGL support on Windows.

351 +2 (1, Flamebait)

narcc (412956) | about 10 months ago | (#44109731)

IE11 scores 351/500 + 2 bonus point

Wow, that's ... pitiful.

Re:351 +2 (1)

Horshu (2754893) | about 10 months ago | (#44109753)

It's also a preview build. Get over it.

Re:351 +2 (1, Flamebait)

narcc (412956) | about 10 months ago | (#44109955)

Considering IE 10's score, I seriously doubt that the fact that it's a "preview build" matters. It's not likely to improve much, if at all.

Hell, give it another 10 points and it's still behind a number of televisions.

Get over it.

Re:351 +2 (3, Insightful)

RulerOf (975607) | about 10 months ago | (#44110123)

To be fair, those TV's are probably all running Webkit.

I still prefer Chrome over Internet Explorer, but IE 10 (the "Metro" version anyway) isn't a mind-numbingly terrible piece of software in comparison to the competition. It's good to know that, however ironic it may be, Microsoft, Mozilla, and Opera are all working opposite Google to keep the web away from just a different monoculture.

Re:351 +2 (0)

Anonymous Coward | about 10 months ago | (#44110187)

Newest version of Opera uses Blink, so it's basically chrome now.

Re:351 +2 (2)

neonmonk (467567) | about 10 months ago | (#44111375)

Actually there's nothing wrong with an open source "monoculture" - maybe if everyone used their own fork of webkit web development would be simpler without any of the players being able to stop innovating without falling behind.

Re:351 +2 (0)

Anonymous Coward | about 10 months ago | (#44114239)

This is utter rubbish. Next you'll be telling me that it would be an improvement if the only OSS operating system kernel out there was Linux. Forks do not spur on innovation or competition as much as people seem to think they do.

Re:351 +2 (1)

LordThyGod (1465887) | about 10 months ago | (#44114487)

To be fair, those TV's are probably all running Webkit. I still prefer Chrome over Internet Explorer, but IE 10 (the "Metro" version anyway) isn't a mind-numbingly terrible piece of software in comparison to the competition. It's good to know that, however ironic it may be, Microsoft, Mozilla, and Opera are all working opposite Google to keep the web away from just a different monoculture.

I've not had the personal "pleasure" of using IE10, since my long experience with previous MS products led me down a different path. But it is kinda of good to see MS is not shitting all over their own customers any more, just because they can and their customers don't know any better.

Re:351 +2 (1)

styrotech (136124) | about 10 months ago | (#44117299)

It's good to know that, however ironic it may be, Microsoft, Mozilla, and Opera are all working opposite Google to keep the web away from just a different monoculture.

Opera? Opera is now a Chrome skin.

Html5 Test at http://html5test.com/ (2)

tuppe666 (904118) | about 10 months ago | (#44111635)

IE11 scores 351/500 + 2 bonus point

Again a reasonable post marked flamebait. The quote *directly* from the summery presented the score in a preview browser as great when in reality it is pitiful. The figures are from http://html5test.com/ [html5test.com] and the Browser I am currently rocking is Firefox 22 (A released browser) which scores 409/500 + 10 bonus points.

Re:Html5 Test at http://html5test.com/ (0)

Anonymous Coward | about 10 months ago | (#44112011)

Chrome 27 on Windows XP gets 463+13

Re:Html5 Test is webkit, not W3C html 5 (1)

Billly Gates (198444) | about 10 months ago | (#44112819)

With rapid releasing the new IE 6 is webkit. MS is being conservative as it does not want another box model scandal for 10 years as developers write 2 different standards. I know its cool to hate IE but it does have 90% of Firefoxes features and is updated annually now. W3C already changed standards which hurt older phones as they recognize webkit css for Android 2.3.

Re:Html5 Test at http://html5test.com/ (1)

Krojack (575051) | about 10 months ago | (#44112821)

Weird, I get 410 + 14 with my FF 22. It's biggest failing points are the various FORM INPUT TYPE= test. I can live with that.

Still I like Chrome more. Seems faster on my laptop. I also like the sync options to keep me laptop and desktop settings in sync. The FF sync system would keep failing on and off for months and they seemed to never get it fixed for good.

WebGL? Nice (0)

Anonymous Coward | about 10 months ago | (#44109811)

Wasn't sure if they were ever going to support this, but I guess it was just a matter of time

Holy confusion Batman! (1, Funny)

Anonymous Coward | about 10 months ago | (#44110089)

Is the name of the submitter rescendent writes?
Is this article about a web browser or a web server?

"IE11 scores 351/500 + 2 bonus point, and 25/25 for WebGL."
So 25 for WebGL and 351+2 for ??

And why the fuck are we even reporting some M$ marketing FUD to begin with?

Re:Holy confusion Batman! (0)

Anonymous Coward | about 10 months ago | (#44110441)

this!

Is the name of the submitter rescendent writes?
Is this article about a web browser or a web server?

"IE11 scores 351/500 + 2 bonus point, and 25/25 for WebGL."
So 25 for WebGL and 351+2 for ??

And why the fuck are we even reporting some M$ marketing FUD to begin with?

Re:Holy confusion Batman! (1)

CrashandDie (1114135) | about 10 months ago | (#44110495)

Is the name of the submitter rescendent writes?

Yes.

Is this article about a web browser or a web server?

Yes.

So 25 for WebGL and 351+2 for ??

Yes.

And why the fuck are we even reporting some M$ marketing FUD to begin with?

Yes!

Quite advanced for Microsoft (0, Troll)

antifoidulus (807088) | about 10 months ago | (#44110181)

By 2014 they will be caught up to where other browsers were in 2011, for Microsoft that's pretty impressive. Usually IE is about half a decade behind real browsers.

Lag in features from Microsoft by 2 years (2)

tuppe666 (904118) | about 10 months ago | (#44111741)

By 2014 they will be caught up to where other browsers were in 2011, for Microsoft that's pretty impressive. Usually IE is about half a decade behind real browsers.

Firefox 14 released July 17, 2012 scored 252/500 so realistically its more like 2 years of Non-Microsoft released browsers vs Microsoft unreleased browsers. there is a graph at the bottom of the page here http://html5test.com/results/desktop.html [html5test.com] that shows the lag in Microsofts development.

Re:Lag in features from Microsoft by 2 years (1)

Billly Gates (198444) | about 10 months ago | (#44112859)

Fud

I have firefox esr 17 which just came out. Not even e year old. 2 years ago it was Firefox 4

Server build (1, Troll)

Bert64 (520050) | about 10 months ago | (#44110317)

Why would the score be higher for a server build? What use would a server have for a web browser? Surely they should be making more effort on the client version of the browser, where it might actually be used?

Re:Server build (1)

benjymouse (756774) | about 10 months ago | (#44111627)

Why would the score be higher for a server build? What use would a server have for a web browser? Surely they should be making more effort on the client version of the browser, where it might actually be used?

It is the same version, but running on the server it may disable some features because of security settings. Actually I'm surprised that WebGL is enabled on a server build, given that WebGL has the potential to inject and run code on your GPU.

However, there needs to be a server version for 1) those who install and maintain a server through it's own GUI and 2) servers used as "terminal servers". Yes, 1) is bad practice but it actually allows small shops to run their own server. Microsoft recommends installing servers as "server core" - i.e. with no GUI.

Defending their reputation (0)

Anonymous Coward | about 10 months ago | (#44110333)

At least there's one organization that tries to save IEs reputation as a horribly bloated browser. For a while, I thought Firefox was going to win that race...

but what about security (2)

gbjbaanb (229885) | about 10 months ago | (#44110525)

ok, so it is still considered harmful [technet.com] or was that the usual guff about how DirectX (or whatever brand-name they're pushing today) is vastly superior to anything standard.

Have they done anything with WebGL to "fix" the mentioned problems, or have they just realized no-one listens to their FUD anymore?

Re:but what about security (1)

DrXym (126579) | about 10 months ago | (#44110667)

The main requirement for WebGL is the driver supports OpenGL ES 2.0 and an extension which says it's "robust", i.e. it does additional range checking and recovery so it can't be abused by a site and cause a BSOD. Browsers may also implement a whitelist / blacklist of drivers which are known to be good or bad and act accordingly.

But it's obviously an additional attack surface which is largely outside of a browser's control and it requires the driver author to declare robustness but how does it know? It's a bit like the safe for scripting bit on ActiveX controls. Therefore while I think WebGL is a good feature to have I would hope that browsers either disable it by default, disable cross frame support and ask on a site by site basis whenever JS attempts to access the API via the canvas.

On the flipside moving functionality like graphics and audio which used to be executed in a Flash plugin into the browser is probably a good idea in the long term.

You moved your mouse. Cancel or Allow? (1)

tepples (727027) | about 10 months ago | (#44113395)

But it's obviously an additional attack surface which is largely outside of a browser's control

A poorly written 2D graphics driver is likewise. What makes a 3D graphics driver substantially more so?

ask on a site by site basis whenever JS attempts to access the API via the canvas.

Which runs the risk of creating a "Cancel or Allow" type scenario where every site wants to use it, and you end up having to click "Allow" for every site you visit.

Re:You moved your mouse. Cancel or Allow? (0)

Anonymous Coward | about 10 months ago | (#44114671)

A poorly written 2D graphics driver is likewise. What makes a 3D graphics driver substantially more so?

The 2D graphics driver isn't usable by web content.

Re:You moved your mouse. Cancel or Allow? (1)

spongman (182339) | about 10 months ago | (#44115267)

A poorly written 2D graphics driver is likewise. What makes a 3D graphics driver substantially more so?

The 2D graphics driver isn't usable by web content.

The 3D graphics driver isn't usable by web content

Re:You moved your mouse. Cancel or Allow? (1)

Tailhook (98486) | about 10 months ago | (#44115133)

A poorly written 2D graphics driver is likewise. What makes a 3D graphics driver substantially more so?

Traditional web pages don't directly use a 2D API in the browser so even when vulnerabilities exist they are often beyond the reach of a web application. A web application is limited to manipulating the DOM. No direct access to blitters, compositors, video modes or other hardware features. Even with only this high level exposure there have been exploits, often in image format interpreters.

WebGL directly exposes the complete OpenGL 2.0 ES API (soon 3.0) to Javascript. WebGL apps can probe hardware for features (memory capacity, optional features, extensions, etc.) and identify implementations, each harboring distinct vulnerabilities. WebGL does things like compile shader code, directly manipulate texture memory and transfer large vertex buffers to video hardware. Traditional DHTML can't do anything like that. WebGL truly is an exposure of GPU hardware and driver software directly to web applications.

Not that I am adverse to WebGL. It think WebGL is great and I'm thrilled that Microsoft has capitulated. I'm just failing to pretend WebGL doesn't create new vulnerabilities. It does and we'll muddle through until the stack is hardened because it's worth it. A few years from now the moral equivalent of Photoshop and Solidworks will be in browsers because of WebGL.

Re:You moved your mouse. Cancel or Allow? (1)

spongman (182339) | about 10 months ago | (#44115375)

WebGL directly exposes the complete OpenGL 2.0 ES API (soon 3.0) to Javascript

where did you get this? this is crap.

WebGL truly is an exposure of GPU hardware and driver software directly to web applications.

this is just complete FUD. WebGL is just an API.

an implementation might blindly forward everything to an apparently compatible hardware device. it would be fucking stupid if it did. especially since it's trivial to statically determine such things beforehand.

it would equivalent to having a "asm(...);" function in your browser's javascript JIT that allowed you to inject arbitrary x86 code into the function, or allowing you to write outside the bounds of arrays. this is perfectly valid javascript, but if your implementation allowed this you would be at fault - not the language.

Compare to HTML formatting in Slashdot comments (1)

tepples (727027) | about 10 months ago | (#44115621)

WebGL does things like compile shader code, directly manipulate texture memory and transfer large vertex buffers to video hardware. Traditional DHTML can't do anything like that. WebGL truly is an exposure of GPU hardware and driver software directly to web applications.

WebGL doesn't necessarily "expose" the GPU to the web application any more than a site that allows HTML comments "exposes" the viewer's browser to the user posting a comment. Slashdot and several other web sites allow users to post comments with a subset of HTML. Arbitrary HTML can perform cross-site scripting using <script> elements, attributes whose name starts with "on", and URLs using the "javascript:" scheme. To prevent this, forum software used by these sites parses and sanitizes [wikipedia.org] the provided HTML before passing it to the web browser. Likewise, a web browser should sanitize WebGL shader code before passing it to the host OpenGL implementation.

Re:but what about security (1)

spongman (182339) | about 10 months ago | (#44115221)

The main requirement for WebGL is the driver supports OpenGL ES 2.0

that's bollocks. there's no requirement that the driver support any variant of OpenGL. the requirement is that the browser can perform the operations requested of it by the WebGL program and fragments. the browser is free to ignore them, send them to an OpenGL ES 2.0 driver, transform them into DirectX calls, or send them by carrier pigeon to your grandma.

somewhere along the long the line you might want to check your inputs, but this certainly doesn't need to be a function of any driver.

it's the same with JITing javascript - if you don't check your bounds and send crap code to the CPU then you're probably going to be vulnerable to remote code injection. but that's not the fault of javascript, just of your shitty implementation.

there's nothing about WebGL that's unsafe - just bad implementations. and marketing FUD.

Re:but what about security (0)

Anonymous Coward | about 10 months ago | (#44115641)

WebGL security is a problem.

Even if the static security guarantees of WebGL are strong enough to prevent (for example) infinite looping in a fragment shader, out-of-bounds access to vertex buffers, etc, and even assuming that the underlying 3D driver that WebGL forwards to is bug-free (hah!), any WebGL application can trivially hang your GPU by simply rendering a large number of large blended quads in a single draw call. Since current GPUs have no preemption, that will hang your GPU. There’s no way to detect statically all the variants of draw calls that might cause arbitrarily long execution — that’s asking to solve the halting problem.

In an ideal world, your graphics driver will detect the hang after a few seconds, reset the GPU, and kill the offending process. In practice, GPU reset doesn’t seem to work all that well, and even if it did, it is not without consequence — Any other process which was using the GPU for any other purpose may see interrupted computations, corrupted images, etc. and realistically, probably wasn’t coded to handle those kinds of situations, leading to knock-on problems.

In the future, if/when GPUs have protected memory & preemption, WebGL will be safe enough. But now? Chrome and Firefox certainly seem to think that having a web page hang your computer is fine. I’m not so sure.

Windows XP still 20%...and again unsupported (0, Troll)

tuppe666 (904118) | about 10 months ago | (#44110989)

I know supporting XP is against two (different?)major strategies; Selling people on Microsoft's stupid everything is a tablet so we are winning(the new strategy of copying Apple), and the old we are Monopoly, buy a new version by crippling(discontinue) the old version so we can can roll around in cash(The old strategy when they were called Micro$oft).

Ignoring problems from fragmentation, and support...and it holding back the web for many years, or how Microsoft basically won against the United States by building IE into the OS (a partially successful strategy for them). Google has started separating its first party Applications from the underlying OS(in Android) by giving its users a great internet experience.

I notice that chrome continues to rise in usage as Internet Explorer continues to Dive(Firefox too, but for different reasons)

These graphs say it all http://www.w3counter.com/trends [w3counter.com].

Re:Windows XP still 20%...and again unsupported (0)

Anonymous Coward | about 10 months ago | (#44111261)

Every company discontinues support for older products. Microsoft has a well defined support lifecycle for all of it's products. They extended the support for Windows XP by quite a few years. Do you think they should backport everything they do across their entire product line? Do we need to be able to run IE11 on MS-DOS?

As for "building the browser into the OS", the only thing notable about MS is that they did it before everyone else. Making web browsing, rendering and common Internet networking functions (and not just sockets) common libraries to the platform has been par-for-the-course for a decade now. The Safari story on MacOSX is identical to that of IE on Windows, you can "hide" the browser but you cannot remove the rendering engine and supporting libraries.

Is Internet Explorer relevant (0)

tuppe666 (904118) | about 10 months ago | (#44111963)

Every company discontinues support for older products

Except Microsoft's competitors haven't; which is the point Firefox and Chrome latest versions are still available for XP, and unlike Microsoft's business Model there business models demands it. You should reread my post. Its about questioning Microsoft's Business Models, not about Microsoft's unsavoury business practices and continued Monopolistic abuse.

In answer to your question "Do we need to run IE11 on MS-DOS?" the answer is yes if the costs of supporting it are less than the potential revenue generated from *converting* a customer to Internet Explorer and its defaults to its Bing search engine (and other web services), as well as a whole host of intangibles(Its online office and Windows Phone looking more attractive). In reality IE should work on Linux.

Maybe your arguing that Internet explorer cannot compete on its own merits(and never will) and has to use monopolistic abuse as a means of staying relevant, and I think I maybe agree with you.

Re:Is Internet Explorer relevant (2)

KingMotley (944240) | about 10 months ago | (#44112969)

Except Microsoft's competitors haven't

Oh? Where can I find the repository for security patches for linux 1.0? Or how do you install safari 6.0 on OS/X 10.6 or Windows anything? I'd like to install iOS 6.1 on my iPhone 3. How about the latest firefox on Windows 2000?

Which competitor were you talking about?

Where...I everywhere. (0)

tuppe666 (904118) | about 10 months ago | (#44113551)

Oh? Where can I find the repository for security patches for linux 1.0?

https://www.kernel.org/ [kernel.org].

The reality though by compitors I unsurprisingly meant (but not limited to :) Firefox http://www.mozilla.org/en-US/firefox/new/ [mozilla.org] and Chrome https://www.google.com/intl/en/chrome/browser/ [google.com]

XP exists on about 20% of computers...or about 220,000,000 which is why the point is about XP :)

Re:Is Internet Explorer relevant (1)

Ravaldy (2621787) | about 10 months ago | (#44115137)

KingMotley, you can't argue with people who are sold to hate MS. It's like people who hate rich people until they become rich themselves. Then all of sudden poor people are the problem. Hypocrites I say.

Re:Windows XP still 20%...and again unsupported (0)

Anonymous Coward | about 10 months ago | (#44111783)

I know. IE11 doesn't run on my C64's MS-developed BASIC, either. I therefore conclude that Microsoft are a bunch of mindless jerks.

It also doesn't run on iOS ro Android (0)

tuppe666 (904118) | about 10 months ago | (#44112093)

I know. IE11 doesn't run on my C64's MS-developed BASIC, either. I therefore conclude that Microsoft are a bunch of mindless jerks.

Except that Later versions of IE does not run on competing platforms or earlier iterations of itself still in use. I don't think it makes them Mindless Jerks...losers perhaps, it means that Chrome experiences continued to grow market share on the Desktop while IE continues to plummet...even without throwing in Mobile browsing where its presence continues to be non existent.

Stop complaining about XP's EoL (3, Insightful)

Bacon Bits (926911) | about 10 months ago | (#44112591)

Windows XP was released in October of 2001. That's also the same month Red Hat 7.2 was released. I guess you could say that was a good month for operating systems.

You know when Red Hat 7.2 was end-of-lifed? December 2003 [redhat.com]. Ten years ago.

Nothing to do with EOL (2, Informative)

tuppe666 (904118) | about 10 months ago | (#44113009)

Windows XP was released in October of 2001.

Yet was not replaced for till Vista was RTM as of January 30, 2007...although XP was still the only viable alternative till Windows 7 in July 22, 2009, although still massively too bloated an OS for most existing XP machines. You measure from the last sale not the first one :)

But that has nothing to do with my comment which is Browser market share...and the benefits from having one. Personally I love the fact your defending Microsoft treating its customers with second class versions of its software. I am running Firefox 22 on my preferred GNU/Linux right now :)

Re:Windows XP still 20%...and again unsupported (2)

PhrostyMcByte (589271) | about 10 months ago | (#44112699)

Supporting XP is especially complicated for IE, because starting with IE9 they've rewritten the rendering engine to use Direct2D and DirectWrite (Vista only), which depend on Direct3D 10 (Vista only), which depends on WDDM 1.1 (Vista only). Should they be backporting all of these APIs and kernel-mode code as well? Or maybe just maintain a large GDI fallback just for XP?

XP is 12 years old, and its last major update is 5 years old. How long should they continue to develop new software and major updates for a platform?

It's not like they're holding people back: they can switch to Chrome or Firefox if they want a modern experience, and the IE they have is the one they need for compatibility with IE-only websites.

Get over it. (0)

tuppe666 (904118) | about 10 months ago | (#44113457)

Supporting XP is especially complicated for IE,

Then perhaps they should have designed IE explorer to be cross platform :). The fact that IE does not run on Android soon to be the most dominant platform is an example of what a failure it is. The fact that is supplies its own customers with a second class experience.

Re:Windows XP still 20%...and again unsupported (2)

LocalH (28506) | about 10 months ago | (#44112729)

How come we never hear people bitching that the newest Safari doesn't run on anything older than OS X 10.7? It doesn't look like Safari 7 will be available on anything other than 10.9, as well. Why is that any different to MS not supporting XP, which is much older than 10.6?

Re:Windows XP still 20%...and again unsupported (1)

Billly Gates (198444) | about 10 months ago | (#44112963)

Oh cry me a river.

I work for a dirt cheap client. Even we have been moving off XP for a year now. That number is rapidly declining and non biased sites like g.statcounter show XP equal to MacOSX in the US. Only china uses it as 97% is all pirated and Windows 7 is harder to steal it. By 2014 it will be single digit marketshare in the western world.

Not a Linux user today ;) (0)

tuppe666 (904118) | about 10 months ago | (#44113619)

Oh cry me a river.

I work for a dirt cheap client. Even we have been moving off XP for a year now.

Not pretending to be a Linux user today. I like you more as a straight Microsoft Apologist. XP exists on 220,000,000 computers. I personally love the fact that you use an Apple excuse to defend Microsoft it, shows how overpriced and under-supported Windows is.

Re:Not a Linux user today ;) (1)

Billly Gates (198444) | about 10 months ago | (#44119191)

I frankly dont know know where you stand? Statcounter shows its death accelerating. XP in my mind is like IE 6. Different era and on its way out. In 2011 when IE 9 came out yes it was debatable as XP loyalists were huge then. Ms is heading the direcrion though Metro is poorly executed. IE is growimg up

Re:Windows XP still 20%...and again unsupported (1)

Ravaldy (2621787) | about 10 months ago | (#44115087)

People are way too serious about their browser selection and take it far too personal. Some of you Slashdotters are starting to sound like broken records or a bunch of red necks arguing whether Dodge or Ford has the best trucks.

Fact is that IE still fills a number of needs for enterprise and corporate customers. It has features Chrome and Firefox don't offer and don't plan on offering. Chrome and Firefox cater to WEB USERS and they do a very good job at it. IE is still a popular browser and will continue to remain one of the top 3 browsers for a long time so crying about it won't make them disappear.

FYI, I use both Chrome and IE. Whatever is on the box I work with is what I'll use because I just don't see a difference.

Fact is...IE does not make the cut(or does it) (1)

tuppe666 (904118) | about 10 months ago | (#44117509)

Fact is that IE still fills a number of needs for enterprise and corporate customers.

Its not a fact. In reality any firm wanting a cutting edge advantage (as well as others) will avoid IE like a plague. The reality is targeting a platform dependant in today BYOD rather than a cutting edge browser specification, for internal...or mobile deployment is incredibly foolish.

Less like comparing brands...more like hiring a serial killer as a baby sitter.

This is perfect, now support OpenGL ES 2.0 (1)

goruka (1721094) | about 10 months ago | (#44111183)

WebGL is basically almost the same as OpenGL ES 2.0, which is missing on Windows Phone, Windows RT and Windows 8 modern.
Nowadays most mobile apps use this API for 3D, which makes porting to their platform a big hassle (Only D3D in 9_3 profile is supported, which is even more limited than GL ES 2.0).
Im sure that this is done with a wrapper over their D3D driver, so I hope they make it available for the C++ APIs.

Re:This is perfect, now support OpenGL ES 2.0 (1)

Lennie (16154) | about 10 months ago | (#44112335)

Yes, I believe the WebGL implementations even in Chrome and Firefox on Windows are implemented in Direct3D/DirectX/whatever.

Don't count on it the wrapper is probably in IE itself.

Use a WebView and you can use WebGL, that solves your problem. ;-)

Re:This is perfect, now support OpenGL ES 2.0 (1)

Derek Pomery (2028) | about 10 months ago | (#44115319)

Firefox and Chrome use ANGLE, yes.
http://code.google.com/p/angleproject/ [google.com]

You can use native opengl though.
In Firefox, about:config and search for webgl.
Set:
webgl.prefer-native-gl;true

In past I've needed to do this on some windows machines to get some WebGL to work. Shader issue or somesuch.

There's also:
webgl.force-enabled;true

While you're in that section, btw, if you feel you know your card/driver combo better than their blacklist does.

Oh, and:
gfx.direct2d.force-enabled;true
or even
gfx.direct2d.disabled;true

If you're on a blacklist for that, or just want to investigate some direct2d rendering issue.

Re:This is perfect, now support OpenGL ES 2.0 (1)

spongman (182339) | about 10 months ago | (#44115287)

you can't use WebGL in web safari on the iPhone because that would allow you to write compelling content on the web that would cut into Apple's 30% of everything.

Can't Wait (1, Funny)

Jason Levine (196982) | about 10 months ago | (#44111447)

I can't wait to support the wonderful features in IE11. Going by my site stats for IE over the 2 past years and W3Counter's graphs (http://www.w3counter.com/trends), I look forward to 10% of my IE users (themselves only 15% of my visitors) using it in 2 or 3 years. (Or, I can support wonderful new features in Chrome, FireFox, and Safari and cover over 75% of my users.)

Re:Can't Wait (0)

Anonymous Coward | about 10 months ago | (#44111857)

I can't wait to support the wonderful features in IE11. Going by my site stats for IE over the 2 past years and W3Counter's graphs (http://www.w3counter.com/trends), I look forward to 10% of my IE users (themselves only 15% of my visitors) using it in 2 or 3 years. (Or, I can support wonderful new features in Chrome, FireFox, and Safari and cover over 75% of my users.)

Totally!!!!11!11!!one!!!! OMFG M$ should just give up and stay on the same browser for five years, that would totally solve your problem wouldn't it!?!?!one!!!

Strict Transport Security (HSTS) (0)

Anonymous Coward | about 10 months ago | (#44111793)

All other major browsers have had this no-brainer security feature for years now and still crickets from MS land. Not cool.

w3counter? (1)

gtirloni (1531285) | about 10 months ago | (#44112279)

I wonder where they are getting this statistics from, who is their sources, if there is any bias, etc.
Load More Comments
Slashdot Account

Need an Account?

Forgot your password?

Don't worry, we never post anything without your permission.

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>
Sign up for Slashdot Newsletters
Create a Slashdot Account

Loading...