×

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!

Despite Game-Related Glitches, AMD Discontinues Monthly Driver Updates

timothy posted about 2 years ago | from the oh-the-humanity dept.

AMD 213

MojoKid writes "Recently AMD announced that it would cease offering monthly graphics driver updates, and instead issue Catalyst versions only 'when it makes sense.' That statement would be a good deal more comforting if it didn't 'make sense' to upgrade AMD's drivers nearly every single month. From 2010 through 2011, AMD released a new Catalyst driver every month like clockwork. Starting last summer, however, AMD began having trouble with high-profile game releases that performed badly or had visual artifacts. Rage was one high-profile example, but there have been launch-day issues with a number of other titles, including Skyrim, Assassin's Creed, Bat Man: Arkham City, and Battlefield 3. The company responded to these problems by quickly releasing out-of-band driver updates. In addition, AMD's recent Catalyst 12.6 beta driver also fixes random BSODs on the desktop, poor Crossfire scaling in Skyrim and random hangs in Crysis 2 in DX9. In other words, AMD is still working to resolve important problems in games that launched more than six months ago. It's hard to put a positive spin on slower driver releases given just how often those releases are necessary."

cancel ×
This is a preview of your comment

No Comment Title Entered

Anonymous Coward 1 minute ago

No Comment Entered

213 comments

They didn't say slower (5, Insightful)

macemoneta (154740) | about 2 years ago | (#40201315)

They didn't say slower, they said as needed. Since they are already releasing 'out of band' they are just normalizing that process. They will release when they have fixes / function instead of on an arbitrary timeline. It seems to make perfect sense.

Re:They didn't say slower (0)

Anonymous Coward | about 2 years ago | (#40201465)

It makes even more sense if you realise that an out-of-band update means doing the release engineering process twice that month. But it disappoints people that assume, not entirely unjustifiably, that it indeed will mean less updates. A regular schedule at least ensures that even minor fixes will make it out in reasonable time.

I'd suggest they commit to releasing again at the latest two months after the most recent. That way you can aim for one release a month but be reasonably flexible about "out-of-band" releases.

Re:They didn't say slower (1)

im_thatoneguy (819432) | about 2 years ago | (#40201817)

Not only that, but releasing 'out of band' patches on a regular basis was probably a huge drag on their responsiveness. If you have a deadline for a new driver patch EVERY SINGLE MONTH regardless if the patch is ready then you are either testing inconsequential releases and wasting your QA resources on releases that aren't important in order to meet some arbitrary deadline or you are releasing insufficiently tested patches to meet some arbitrary quota. Neither is a recipe for efficient allocation of your resources. It's either a distraction from what's important (fixing critical bugs) or it's forcing you to release bug fixes that are insufficiently tested.

An artificial release schedule may slow things (1)

perpenso (1613749) | about 2 years ago | (#40201983)

An artificial release schedule, one tied to the calendar rather than bug fixes, can actually slow things down. It can cause a certain amount of disruption when a team is in the middle of taking care of a bug. It seems somewhat similar to having to put together a demo when you are in the middle implementing a feature. I'd say try for a monthly release but don't necessarily let that goal interrupt fixes underway, let in progress fixes delay the release when it makes sense to do so.

Re:An artificial release schedule may slow things (0)

Anonymous Coward | about 2 years ago | (#40202647)

The Catalyst releases where not really time boxed as the actual release date could slide arbitrarily within the month. At least one time they failed to land inside the month. Perhaps they should have separated the development of product improvements and the actual productizing of the driver. Now the users have to suffer from the random control center bugs for longer periods time. It will be painful.

Is this nvidia spin? (2, Interesting)

Anonymous Coward | about 2 years ago | (#40201319)

I mean, of course frequent updates are desirable. On the other hand, every release produces overhead which could be used to fix the problems at hand. In my experience, monthly update schedules are a terrible waste of valuable time.

Personally, I'm an nvidia user, since I hate the driver issues of AMD... but this news sounds like nvidia spin to me.

Which is worse, AMD or nVidia? (0, Flamebait)

Futurepower(R) (558542) | about 2 years ago | (#40201545)

LaptopVideo2Go.com [laptopvideo2go.com] is a very active web site entirely devoted to making nVidia graphics devices work correctly. nVidia tried to avoid doing anything about defective chips in HP laptops. [pcworld.com]

When you download AMD's ATI drivers, the web site tries to sell violent video games. The new drivers often have serious bugs.

If there is a competition, which CEO will be voted the worst? nVidia does not seem honest, and AMD seems to be trying to drive itself out of business.

Re:Which is worse, AMD or nVidia? (1)

drinkypoo (153816) | about 2 years ago | (#40201683)

The argument between HP and nVidia over defective GPUs is between HP and nVidia, not between me and nVidia. My HP laptop with QuadroFX1500M had a known problem. I had to fight with HP for more than 24 total hours on the phone to get them to admit it and issue me a replacement, but I did so. I had no problem with nVidia. HP had a problem with nVidia. I had a problem with HP. Replacement had a newer GPU (and everything else) and I sold it and bought some netbooks.

One of the netbooks I bought has AMD Athlon 64 L110 and R690M chipset. It works properly only under Vista. There are no platform drivers even for Windows 7. Under Windows 7 suspend/resume works once. Under Vista everything works... verrrry sloowwwwwwly. Under Linux I have graphics corruption even with RenderAccel disabled. fglrx doesn't support it and didn't even when it was shipping. ati causes trashing. Power management is essentially nonworking.

I hope one day the ati driver works properly on here, and that I can get coreboot working, because AMD also let Gateway disable AMD-V on the machine and still call it an Athlon 64. In my book when you disable features that differentiate one product from another, you should use a more honest name. That's Gateway more than AMD, except it was wearing the "Athlon 64" sticker, so I knew the CPU would have AMD-V. Indeed, it does, but I can't use it. AMD should not permit the use of the sticker when features are disabled.

I'm over AMD for video. Still happy with my Phenom II X3 720, plan to upgrade to an X6 for encoding. Works nice with my 240GT.

Considering nVidia's actions, do you feel safe? (3, Interesting)

Futurepower(R) (558542) | about 2 years ago | (#40201991)

"The argument between HP and nVidia over defective GPUs is between HP and nVidia, not between me and nVidia."

The way nVidia has acted in the past is an indication of how it may act in the future. See one of the many articles, for example: Dell and HP balk at replacing bad Nvidia chip. [windowssecrets.com]

If you buy something with an nVidia product in it, you may get involved with enormous hassles like that. People who weren't following the sneakiness and dishonesty closely didn't get their computers replaced because there was a very limited period in which customers needed to act.

Both AMD and nVidia need better management, in my opinion.

Re:Considering nVidia's actions, do you feel safe? (1)

Anonymous Coward | about 2 years ago | (#40202825)

That wasn't even a problem with the Nvidia chips themselves. The problem was solely HP's due to poor design, the motherboards would warp from thermal stress and the BGA would eventually peel away from the GPU.

Re:Which is worse, AMD or nVidia? (-1)

Anonymous Coward | about 2 years ago | (#40201773)

Which is worse I can tell you... Neither, how about shit game developers that can't test or make their crap work. Instead they like to blame the video drivers. Take Id for example, carmacks crap programming created all sorts of glitches and problems and it was Nvidia and Amd that had to clean it up for him.

Re:Which is worse, AMD or nVidia? (5, Insightful)

Cute Fuzzy Bunny (2234232) | about 2 years ago | (#40202051)

If as a programmer I can do something that crashes your driver or blows up your machine, then the problem was with the driver, not the application programmer.

I was a systems programmer for 30 years. I wrote a ton of OS and driver code, especially drivers. If you could break the machine or cause stupid things to happen by having your app do something improper with the driver, then that was my fault.

Re:Which is worse, AMD or nVidia? (-1)

Anonymous Coward | about 2 years ago | (#40202489)

Moron-like typing detected

Re:Which is worse, AMD or nVidia? (1)

Cute Fuzzy Bunny (2234232) | about 2 years ago | (#40202005)

I don't think they're trying to drive themselves out of business, they just aren't that competent but their competitors are.

This smells like a cost savings move to me, I guess the positive spin is that they might take longer to go out of business if they cut costs.

What'll actually happen is they'll get a bad rep for having problems that sit around too long. Seems they have OCZ disease...too excited to release new products and start making money and not smart or thorough enough to validation test their products. I got tired of being someone elses unpaid validation tester. I bought a 6670 and found that out of the box with the latest drivers it couldn't even play a blu-ray without tearing. I'm sure it was just a setting or I just needed another driver, but I'm not going to go through that trouble. A frickin HD video card made and sold in 2011-2012 should play a blu-ray in stunning quality right out of the box. OCZ dropped the vertex 2 and vertex 3 on an unsuspecting customer base, who I'm sure liked the frequent data corruption.

From now on I only buy from companies that do extensive validation testing and get good customer reviews. AMD smells badly in that regard right now. I suspect it'll get worse as they edge closer to chapter 11.

Re:Is this nvidia spin? (1)

Cute Fuzzy Bunny (2234232) | about 2 years ago | (#40202065)

I simplified my entire life by using an xbox 360 and a playstation 3 for gaming, and a blu-ray player for movies. If I have a rip I just burn it and stick it in the blu-ray player.

While there are occasional bugs in the console games, I've rarely experienced them.

So.... (0)

Anonymous Coward | about 2 years ago | (#40201325)

...what their saying is... is that we should purchase NVidia cards because they will actually make their drivers work, and will update the drivers regularly to address these issues? I remember when ATI drivers were notoriously bad, and now it looks like we've returned to those days. Goodbye ATI, wish I could say it was nice knowing you... but that would be a lie.

Let me get this straight: (5, Insightful)

Anonymous Coward | about 2 years ago | (#40201331)

You fix third-party software... by modifying drivers?

How about forcing the game makers to TEST THEIR DAMN GAME before releasing? Is it really so hard to throw together four test-beds with GPUs from different vendors?

Re:Let me get this straight: (5, Insightful)

91degrees (207121) | about 2 years ago | (#40201435)

I'm in two minds over this. Assuming this is an actual glitch in the drivers causing the problems.

On one hand, AMD should fix it. On the other hand, AMD graphics cards are pretty popular. Their game should be designed to work on what they can reasonably expect their users to have.

Re:Let me get this straight: (5, Insightful)

Anonymous Coward | about 2 years ago | (#40201503)

It's easier to just release the game and let gamers and video card manufacturers fight over who is in the wrong. By the time someone figures it out the developers have made their money and run off.

Re:Let me get this straight: (0)

Anonymous Coward | about 2 years ago | (#40201711)

"On the other hand, AMD graphics cards are pretty popular."

They are. Although to what extent that is the result of Bitcoin mining (at which AMD/ATI cards excel, and Nvidia cards suck), I'll leave as an exercise for the reader...

But to address the larger point...

It seems like monthly updates weren't terribly effective, were they?

It also seems - to me, anyway - that if one were to take all the resources AMD devoted to those almost endless upgrades, and instead portion them out to individual *game* updates, it might work out better. It's not like anybody will be inconvenienced by having to download a game-specific driver, rather than getting a generic driver when they purchase a graphics card - they're all downloaded now, as it is. And you've got to have internet access to register all of these games - and to play most of them. Why not just get a game-specific version of Catalyst for each of the (what? two dozen?) major games out there, with a select-button in the Catalyst program to chose the game you'd like to play?

Re:Let me get this straight: (4, Interesting)

murdocj (543661) | about 2 years ago | (#40202451)

Really? Bitcoin mining??? You think people are buying ATI cards to mine bitcoins? And not for gaming? Maybe a few people are reusing their old cards for mining, but the bitcoin fad has pretty much passed... I'd be shocked if even .1% of the AMD graphics cards sold are for bitcoin.

Re:Let me get this straight: (1)

Dogtanian (588974) | about 2 years ago | (#40202977)

On the other hand, AMD graphics cards are pretty popular."

They are. Although to what extent that is the result of Bitcoin mining (at which AMD/ATI cards excel, and Nvidia cards suck), I'll leave as an exercise for the reader...

Is Bitcoin mining *really* that significant a part of the market as a whole? I suspect it probably seems more prominent on Slashdot than it actually is.

Besides which, from what I understand, the increasing difficulty of solving new "problems" to generate Bitcoins meant we'd passed the point where the electricity needed to power the computations outweighed the generally-accepted value of the generated Bitcoins some time back.

Re:Let me get this straight: (0)

Anonymous Coward | about 2 years ago | (#40201489)

Sure, just as soon as we convince all card manufacturers to unify their hardware.

Re:Let me get this straight: (2)

Joce640k (829181) | about 2 years ago | (#40201499)

How about forcing the game makers to TEST THEIR DAMN GAME before releasing?

This.

Unfortunately, though, a forum full of "It doesn't work on AMD cards! OMG!!!" makes AMD look bad, not the game developer. AMD then have to go about emulating NVIDIA's driver bugs.

Re:Let me get this straight: (2)

drinkypoo (153816) | about 2 years ago | (#40201629)

Bahahahha. I'm not denying that nVidia has had driver bugs, but complaining about AMD having to emulate nVidia's driver bugs is like complaining that Intel had to implement AMD64. nVidia is so much better at drivers than AMD that your comment looks like the insane rantings of a madman.

Re:Let me get this straight: (5, Interesting)

Joce640k (829181) | about 2 years ago | (#40201833)

nVidia is so much better at drivers than AMD that your comment looks like the insane rantings of a madman.

Oh, yeah? I program 3D graphics for a living so I have to deal with this stuff on a daily basis. I'm working around a bug right now.

Question: Are occlusion queries supposed to return number of samples or number of pixels in Direct3D?

A certain company's "pro" graphics cards seem to differ from their "consumer" graphics cards over this.

The only way I've found to get my program working is to do a dummy occlusion query when I create the framebuffer and see what happens.

Re:Let me get this straight: (4, Informative)

Shinobi (19308) | about 2 years ago | (#40202153)

Direct3D technically allows for both, the XNA game dev framework specifies number of pixels however, for performance reasons. The number of samples method tends to be more accurate but very slow. It's the same thing on the OpenGL side. CAD, 3D applications such as Maya etc, compositing programs etc tend to use samples over pixels, for more accuracy.

Re:Let me get this straight: (3, Informative)

Anonymous Coward | about 2 years ago | (#40202279)

Question: Are occlusion queries supposed to return number of samples or number of pixels in Direct3D?

Occlusion queries are supposed to return number of pixels in both Direct3D and OpenGL.

A certain company's "pro" graphics cards seem to differ from their "consumer" graphics cards over this.

In both API's or just one? If just one, then the problem is actually within Direct3D and isn't the card at all.

The only way I've found to get my program working is to do a dummy occlusion query when I create the framebuffer and see what happens.

Then you're doing something else wrong and have misidentified the source of your trouble. I won't get into here, but this might prove helpful to you:
http://http.developer.nvidia.com/GPUGems2/gpugems2_chapter06.html
http://http.developer.nvidia.com/GPUGems/gpugems_ch29.html

Re:Let me get this straight: (4, Informative)

sandytaru (1158959) | about 2 years ago | (#40202025)

nVidia has the mother of all driver bugs and they've refused to fix it for years. If you run a DVI to HDMI cable from an nVidia card with no native HDMI support, the driver recognizes the HDMI cable anyway, assumes it can run sound, and attempts to run sound via the nonexistent sound chip on the video card. In essence, it overrides the onboard sound and sometimes even a discrete sound card in the computer. Since native HDMI support was introduced in newer cards, nVidia has felt no need to address this glitch in their older cards. I ended up recycling an otherwise perfectly good GeForce 9800 GT because the computer it was in was hooked up to the 40" television, but any time I had the video card driver installed I had no sound!

Re:Let me get this straight: (0)

Anonymous Coward | about 2 years ago | (#40202059)

You do know that the solution to that problem is as simple as going to Playback devices and setting the default to your sound card instead of HDMI, don't you?

Re:Let me get this straight: (0)

Anonymous Coward | about 2 years ago | (#40202435)

Don't go telling him the solution, he'll never realize that the PEBKAC is in effect.

Re:Let me get this straight: (4, Interesting)

makomk (752139) | about 2 years ago | (#40201951)

Even more unfortunately, NVidia have realised this and have been paying off video game developers not to test their games on AMD graphics cards prior to release and not to allow AMD access to pre-release versions to do it themselves.

Re:Let me get this straight: (4, Informative)

EdZ (755139) | about 2 years ago | (#40201541)

It depends on where the problem lies: If the game is using the directX (or openGL) libraries correctly but the driver is mucking things up, then the game developer should not need to code around driver bugs. Conversely, if the game developer is using a 'clever hack' to eke out some more performance, this creates a headache for the driver developers to keep this hack working in one instance but stop it working for things written to the word of the API in other instances.

Re:Let me get this straight: (5, Interesting)

Svartalf (2997) | about 2 years ago | (#40202207)

It's not always driver bugs. Many of the fixes are things that tapdance around bad, buggy code within the game itself. Oftentimes the studio's devs play fast and loose with shader parameters or API compliance- and NVidia does it differently than AMD, etc.

Any time you see a "MAY" within a standards document, it really ought to be treated as a "SHALL" unless you know you're working on ONLY a target environment that the "MAY" doesn't affect you. Prime example would be something along the lines of VBO mapping to host addressing space. The spec says that it MAY stall the pipeline if you do this while you're in the middle of a rendering pass. Well...NVidia's implementation knows what VBOs are in-flight with a rendering pass and will stall only if it's known to be about to be used by the current pass in progress. AMD's drivers took the other, in fact, sensible approach because it's easier to implement and gains you performance overall if you don't have devs doing stupid things- they stalled ANY time you mapped any VBOs involved with the rendering pass in progress.

A major studio (Who shall not be named, nor shall the game...who knows, maybe you can guess the title...) did this in their GL code- they recycled VBOs, but did it intra -frame instead of inter -frame. The first is realtively safe, producing pretty good performance, the other's very much not so, based on the lead-in I gave just now. I should know, I've used it with some of the games I've done porting work on (Because the studio did the same thing in DirectX...which has the same restrictions here...). When you do it intra-frame, on NVidia, it slows the render pass down, but not unacceptably because it only stalls as long as needed to assure you're not corrupting the render pass. AMD, until they re-worked their VBO implementation would plummet to seconds per frame slide-show renderings on an X1950XTX card when it was THE hottest, fastest card out there- because it would stall the pipeline, taking milliseconds to recover, each and every time they re-mapped the VBO they were re-using to conserve on card memory on the frame's rendering pass.

Was it the driver's fault? Not even remotely close to the truth there. But...people will blame the driver, calling it "buggy". In fact, that's what happend, even.

Re:Let me get this straight: (2)

lexsird (1208192) | about 2 years ago | (#40201551)

Indeed, they should test their damn games. SWTOR has some of the worse issues I have ever encountered from an MMO. It's so sloppy and I am so pissed at the hype. What burns me is the computer game industry is monstrous in scope and size, yet there isn't an iota of gamer rights advocacy at all. If any other industry foisted off such shoddy, broken on purchase products, they would be rotting in prison. Can you imagine how things would be if the other industries had such slacker, shitty standards?

Consumers rule and they need to get their collective shit together and start cracking whips.

Re:Let me get this straight: (1)

Dogtanian (588974) | about 2 years ago | (#40203011)

Consumers rule and they need to get their collective shit together and start cracking whips.

Have you boycotted distributors and/or development teams whose games have had this problem?

Re:Let me get this straight: (2)

Alarash (746254) | about 2 years ago | (#40201603)

I don't know. Since (literally) the start of ATI I've heard about news like this one, or just that some specific (often popular) games not 'working properly.' It's one of the reasons I've never owned a single ATI video card and always went the 3dfx/nvidia route. I'm baffled that some people keep buying ATI, even if they are cheaper on a power:price comparison.

Re:Let me get this straight: (1)

jakobX (132504) | about 2 years ago | (#40202495)

Its cause we dont have these problems. Ive had mostly AMD/ATI cards in my main machine and ive not had many problems. Same when i used nvidia. Funnily enough ive had more crashes and annoyoing bugs with nvidia cards. Nothing major though.

Re:Let me get this straight: (5, Informative)

drinkypoo (153816) | about 2 years ago | (#40201609)

How about forcing the game makers to TEST THEIR DAMN GAME

Games often expose driver bugs. Major game developers are in communication with GPU vendors and when they discover bugs, the ones which turn out to be in the driver or the microcode sometimes get fixed, depending on how new the product is and whether the GPU is from Intel, AMD, or nVidia. nVidia has by far the best record in terms of working drivers, and also in terms of improving support for old hardware in new driver revisions. AMD is by far the worst. They have abandoned whole platforms while they were still shipping, for example R690M. I'm using a subnotebook based on it right now. Only thing it will run without shitting itself is Vista. And fglrx didn't support it when it was brand new, and still doesn't support it, and never will.

Don't be so quick (or anonymous, or cowardly) to assume that it's the game developer's fault when a problem "with the game" is fixed with a driver update.

Re:Let me get this straight: (2)

Splab (574204) | about 2 years ago | (#40202093)

I keep hearing people claim this about ATI/AMD; I must be the luckiest SOB in the world when it comes to buying hardware from ATI, I've never had trouble with any of my cards. Granted I run them under Windows.

Nvidia on the other hand, I have a single GFX sitting in my laptop and that is the crappiest piece of shit I've ever own. GFX driver keeps locking up, keeps crashing and has extremely poor performance compared to its competitors.

Re:Let me get this straight: (1)

St.Creed (853824) | about 2 years ago | (#40202833)

Get another driver version. I've rolled mine back to the one from 2011 and it's pretty stable. It took me 4 new installs to get the one that worked, though...

Note: I could choose between the one supplied by MS through Windows Update for my laptop, the one supplied by HP for my laptop (latest version had lower version than the MS version) and the ones from NVidia. Since the older HP one refused to remove the latest update, I ended up with the older NVidia one. Pretty happy with it, it works okay now.

But anyway: experiment a bit. You may find it helps.

Re:Let me get this straight: (0)

Anonymous Coward | about 2 years ago | (#40202229)

Heh... They also expose bugs within themselves. About 1/2-2/3rds of the bugs worked on when I worked with AMD as a contract dev were working around screwed up shader implementations and other playing fast and loose with the standards type coding within the games.

Re:Let me get this straight: (1)

mwvdlee (775178) | about 2 years ago | (#40201647)

You fix third-party software... by modifying drivers?

How about forcing the game makers to TEST THEIR DAMN GAME before releasing? Is it really so hard to throw together four test-beds with GPUs from different vendors?

Do you mean to tell us that all vendors combined only have four different graphics cards available?

Re:Let me get this straight: (0)

Anonymous Coward | about 2 years ago | (#40201855)

You fix third-party software... by modifying drivers?

How about forcing the game makers to TEST THEIR DAMN GAME before releasing? Is it really so hard to throw together four test-beds with GPUs from different vendors?

Do you mean to tell us that all vendors combined only have four different graphics cards available?

You fix third-party software... by modifying drivers?

How about forcing the game makers to TEST THEIR DAMN GAME before releasing? Is it really so hard to throw together four test-beds with GPUs from different vendors?

Do you mean to tell us that all vendors combined only have four different graphics cards available?

Their is lots of company's that make graphic cards, but only three main gpu's!!! nvida, Intel, and AMD (formally ATI) xfx, sapphire, gigabyte and other card makers use one of the three gpu's, and Intel is usually a built in graphics on the chip that fails at every turn (since they make there chips mainly for industry use like warehouses and private companys that can't afford unix systems). so yes, four test beds is more then enough for game testing.

Re:Let me get this straight: (0)

JamesP (688957) | about 2 years ago | (#40201669)

The real "broken stuff" is probably in : DirectX

Or better, the lack of a precise specification. And different manufacturers implementing different details in different ways.

But it's lots of games depending on several different functionalities. You can certainly test in most games. But then you find out Quake III relies on an old bug or quirk of the spec that you fixed and that broke the game

Game developer makes the game with available cards. Sees that no one does 'item X' correctly, finds a solution that works on everybody. Item X is fixed but breaks old games.

Not to mention AMD (ATI) has a "tradition" of shipping crappy drivers.

Re:Let me get this straight: (3, Informative)

Sir_Sri (199544) | about 2 years ago | (#40201673)

You fix third-party software... by modifying drivers?

How about forcing the game makers to TEST THEIR DAMN GAME before releasing? Is it really so hard to throw together four test-beds with GPUs from different vendors?

Having been on both sides of this.

There are some functions, usually directx functions that just do not behave properly with certain drivers. There is, in many cases, nothing you can do except ask the company to fix it. This is a double problem because a lot of times they won't look at your game until it's finished, so if you finish on friday and release on tuesday guess how much it's been looked at by nVIDIA or AMD.

  While you are writing your game nVIDIA and AMD are writing new drivers and changing how their drivers behave. usually to accommodate someone eleses release, but not necessarily. That's incredibly frustrating, because you may not know whether the bug is your end, or theirs, especially if it behaves differently between driver releases.

For anyone who got the original version of the witcher 2 you could see the problem with 'test their damn game'. There was a problem with how ubersampling the ability to interact with objects. So the game came out with this problem, which is actually rare because almost no one had a card capable of doing ubersampling (even a new gtx680 today has slowdown with it). So AMD and the Witcher devs get onto fixing this problem. I think the problem was actually in how AMD was handling the sampling, but I'm not 100% sure. CD projekt did a hack workaround patch that changed how they did the sampling slightly, and at the same time AMD issued a fix, that wasn't compatible with the workaround. So you ended up in this problem where you're not even sure which solution you should be using as an end user.

Sure, a lot of the releases basically exists to clarify which codepath a particular game should be rendered with, or which SLI/crossfire profile it should use, which is relatively minor on the scale of things. But it really is a problem on the driver end that games are all treated inconsistently, or maybe that's a feature. Depends on your perspective. Treating games differently is a massive pain in the ass for development, but makes the experience much better for players, so take your pick.

Not always or even often the game's fault (2)

Sycraft-fu (314770) | about 2 years ago | (#40201913)

If it has graphical glitches, ya that's probably the game, Poor performance, depends on. The problem with Rage is it is OpenGL and AMD has shitty GL drivers, they have for a long time. nVidia has long had GL and DX drivers that performed equally, AMD has long had GL problems (used to be much worse than now).

If it is BSODs or GPU driver crashes though? No, that is 100% on the graphics drivers. No matter what the program does, it shouldn't bring the system down. Anything running in Ring 3 can't bring the system down without a problem form something in Ring 0, or a piece of hardware. That means the drivers (though they are largely Ring 3 these days) or card.

Drive quality has long been a problem with AMD (formerly ATi) graphics cards. There was a time when they were near unusable for anything but 2D. Some of the old Rage products you wanted to run with the included Windows drivers not the ATi provided ones because they had so many problems. They've gotten a lot better, but they still have more issues.

An example of a recent issue I've run in to was with Sony Vegas. It uses GPGPU to accelerate video effects. For nVidia, it uses CUDA, for AMD it uses OpenCL (since those are what they prefer). I was having all kinds of crashing issues with it on my work system, which had an AMD card. I tried disabling GPU acceleration, no crashes. So I tried an nVidia card in it. Again, no crashes. Not long after Sony released an update disabling a bunch of GPU effects on AMD cards until AMD fixed their driver (which they just did not long ago and Vegas has now reenabled the effects).

Re:Let me get this straight: (2)

RyuuzakiTetsuya (195424) | about 2 years ago | (#40201931)

...aaaaand PC gamers are wondering why I went Mac and console only for games rather than PC.

The games are still buggy now they're CONSISTENTLY buggy!

Re:Let me get this straight: (0)

Anonymous Coward | about 2 years ago | (#40202355)

wait...

Don't all desktops use the same Nvidia / ATI / Intel Graphics cards on x86 hardwar? Won't they have the same driver bugs?

Re:Let me get this straight: (1)

Anonymous Coward | about 2 years ago | (#40201993)

What if it's the games triggering bugs in the drivers? Those drivers are huge, and for them to be completely debugged, is down right impossible

The NVIDIA Transition? (4, Insightful)

deweyhewson (1323623) | about 2 years ago | (#40201333)

As someone who is generally an AMD fan - their processors and video cards generally provide much better performance for much cheaper - their driver support, or lack thereof, is frustrating. NVIDIA consistently has far better driver support, and features, than their AMD counterparts, even if their cards don't provide as much bang for the buck.

If AMD falls even further behind in that game, I may just bite the bullet and switch to NVIDIA just to stop having to worry about driver-related frustrations altogether.

Re:The NVIDIA Transition? (4, Interesting)

DWMorse (1816016) | about 2 years ago | (#40201349)

To date, my nForce motherboard can't hit sleep mode without the network card going full retard. You NEVER go full retard. For shame, Nvidia. It's been over 2 years and they still haven't released a fix. Nvidia has their share of issues too.

This. (1)

Anonymous Coward | about 2 years ago | (#40201389)

Millenia ago, I went with nVidia after having some terribly hilarious problems with ATI drivers (eg, OpenGL not working).

Later, I switched to ATI when it was nVidia's turn to put out worthless drivers.

Presently, I have a lovely situation where my screen goes black for 5-10 seconds before returning to normal. Nothing else is affected during this period - music continues to play, et cetera.

I suspect I'll be picking up an nVidia card soon.

I also suspect that ATI-that-is-AMD and nVidia are in collusion, and purposely orchestrate the alternation of who is putting out shitty drivers at any point. :p

Re:This. (1)

dr. chuck bunsen (762090) | about 2 years ago | (#40201587)

This is a bug in certain apps and games, not your card. I believe there is a driver update that fixes it, but if not the solution is to turn off the factory overclocking in the amd vision software.

Re:This. (0)

Anonymous Coward | about 2 years ago | (#40201677)

i was with nivida from rava tnt through geforce 4, them ati started making decent drivers. i used ati until nvida released the 8800 (dx10 and all) then a gtx260, now a 560ti. nvidia is where i am gonna stay for the time being. cpus on the other hand. i have owned only about 2 intels vs. 7 amds.

Re:The NVIDIA Transition? (0)

Anonymous Coward | about 2 years ago | (#40201417)

nForce have you checked the brand of ethernet controller?
the original chipset drivers will fix it, no nforce had a built-in ethernet controller.

Re:The NVIDIA Transition? (2, Informative)

Anonymous Coward | about 2 years ago | (#40201615)

no nforce had a built-in ethernet controller.

Yes they did. See the forcedeth [wordpress.com] driver (nVidia also provided a binary driver, called "nvnet" in Linux).

Re:The NVIDIA Transition? (2)

game kid (805301) | about 2 years ago | (#40201455)

As someone who very recently switched back to AMD because recent Nvidia cards (including my own) have been giving me and others some annoying and only occasionally recoverable [microsoft.com] Purple Screens of Death*, I can't wait for a decent Company #3** to kick both their asses on driver size and reliability.

*In my case, a GTX 460, after a year of use. After it started interrupting my Terraria games (even with motherboard settings changes) I thought it was time to recheck what others experienced; and after that, time for it to go.

**Intel does not currently count. They need more mana and must drink more booze. [wikipedia.org]

Re:The NVIDIA Transition? (1)

Bacon Bits (926911) | about 2 years ago | (#40201509)

After years of frustration with crap drivers for ATI video cards and crap drivers for AMD chipsets from third parties (any of them) I finally switched to Intel CPU, Intel chipset, and Nvidia graphics cards. I even bit the bullet and got an Intel model motherboard and made sure the RAM I bought was on the list of tested RAM [intel.com] .

I have had zero problems since I bought it in 2009. Intel DP55WG, Intel i7 860, EVGA GeForce GTX 260, 8GB of a supported SKU of Kingston RAM. The biggest problem I've had (knock on wood) is that one of the case fans rattles rarely.

Re:The NVIDIA Transition? (0)

Anonymous Coward | about 2 years ago | (#40201607)

generally it is the job of the mobo maker to fix that, not the chipset designer, unless it is affecting that same chipset across all mobos suing it.

Re:The NVIDIA Transition? (1)

Blaskowicz (634489) | about 2 years ago | (#40201939)

if your ethernet controller is crappy you'd better disable it and use an old 3COM 100Mb card or something.

Re:The NVIDIA Transition? (0)

Anonymous Coward | about 2 years ago | (#40201461)

ATI/AMD has had inferior drivers for years.

NVIDIA had bad hardware for the past few years- as in hardware that was more likely to fail. I'm not sure if this has changed recently.

So pick your poison...

Re:The NVIDIA Transition? (1)

Bieeanda (961632) | about 2 years ago | (#40201515)

If you do, I suggest EVGA. Lifetime warranty on the cards, and if they manage to send a lemon replacement out (mutter) they'll send a Fedex guy to your door to retrieve and replace it.

I got sick and tired of AMD/ATI back when Voodoo was still a pass-through board, and while their hardware has improved I'm never surprised to hear about bullshit with their drivers.

Re:The NVIDIA Transition? (1)

Shinobi (19308) | about 2 years ago | (#40201965)

Funny, EVGA(3 graphics card in succession having huge flaws, such as fan not working properly, capacitors falling off, while the computer hadn't been moved in 6 months, and the third had bad RAM chips) is on my list of "hardware to avoid at all costs", just like Antec PSU's(none of them lasted more than 6 months, unlike the Q-tec, the cheap piece of shit they were meant to replace, is still alive to this day, 11 years after I bought it...), Gigabyte motherboards etc.

Re:The NVIDIA Transition? (4, Interesting)

TheEyes (1686556) | about 2 years ago | (#40202779)

Here's irony for you:

-AMD supposedly releases driver updates on a monthly basis, though they haven't quite managed it for the last couple years, sometimes not making the deadline, sometimes just releasing basically the same driver two months in a row, then releasing out-of-band updates when games break their cards.)
-nVIDIA has always released drivers "as needed'.
-AMD switches to releasing drivers "as needed".
-Everyone complains, and threatens to switch to nVIDIA.

Re:The NVIDIA Transition? (1)

morian97 (1325925) | about 2 years ago | (#40202927)

Don't do that. Do you really want to be associated with company whose CEO shows mockups on stage of products they don't have and has serious production issues? I switched 4 years ago to and never looked back - have over 50 games on steam all (including crysis 2 and rage) running flawless on 6950.

I'd consider buying Nvidia but (2, Interesting)

NotSoHeavyD3 (1400425) | about 2 years ago | (#40201353)

With their constant rebranding of old boards I can never keep straight what the hell I'd be buying. (Is that 600 series a kepler or fermi based board? Who can tell?)

Re:I'd consider buying Nvidia but (1)

Trepidity (597) | about 2 years ago | (#40201407)

While I agree it's pretty annoying, and certainly confusing for many consumers, it's not as if you can't tell what's inside a particular model, since it's pretty easy to find that information by googling.

Re:I'd consider buying Nvidia but (1)

Sycraft-fu (314770) | about 2 years ago | (#40201937)

Well if you really are unable to do a minimal amount of research to find out, ok I guess that's a reason not to buy, but I would think it wouldn't be to hard to just, you know, look shit up. nVidia's site is a good place and not hard to get to.

Also if you are talking desktops, and I assume you are from the use of the term board, then you are talking nonsense. The rebranding has been in the laptop space, not the desktop space. With laptops they do have some mixed naming as there are 600 series parts from their 40nm and 28nm lines. With desktops all 600 series parts are 28nm.

Ultimately it really doesn't matter as what you should check are features and speed, not an arbitrary choice of what technology they use.

But whatever you like to justify your purchase decisions.

Re:I'd consider buying Nvidia but (1)

NotSoHeavyD3 (1400425) | about 2 years ago | (#40202197)

I'm mostly thinking of the NVidia 640 series (Admittedly OEM) where some of them are kepler and some are fermi. (At least that's what I read on the Wikipedia list of all those units.) I mean they have multiple Geforce 640 that are different cards with different chipsets.

Bias much? (5, Insightful)

neokushan (932374) | about 2 years ago | (#40201365)

AMD says that they're moving from a monthly release cycle to a release-when-needed cycle and someone decides to write this piece of trash about it?
It's not a bad thing, it makes sense to do it like this. As the summary points out, AMD currently releases out-of-band updates for when a high-profile title has an issue or launch day performance increases, so it doesn't make sense to make another release that month that doesn't change much. It's just confusing and frankly unnecessary. Doing it "as needed" just means that when a driver release comes out, it's worth updating to. If that means I only have to update my drivers once every few months, I'm fine with that - even if it occasionally means there's 2 or 3 updates in the space of a month because a lot of games happened to come out then. Overall, it's better for everyone.

Article is a big load of FUD and should be ignored.

Disclaimer: I've currently got a Geforce 560 Ti in my desktop and my laptop uses a Geforce 555M chipset - frankly, I'm an nvidia fanboy and this article still disgusts me.

Re:Bias much? (1)

Dyinobal (1427207) | about 2 years ago | (#40201431)

Ya I read it pretty much as the same, this isn't so much as tech journalism as it is an opinion piece about a recent decision by AMD.

Re:Bias much? (1)

cynyr (703126) | about 2 years ago | (#40202149)

so when will linux drivers be needed? It sure won't be to fix windows game under wine.... anyways that is my concern about this statement. I'm still not running xorg-server-1.12 on my one AMD machine because FGLRX doesn't support it yet..

Re:Bias much? (1)

neokushan (932374) | about 2 years ago | (#40202481)

I don't think this statement has any bearing on their linux driver support. Linux driver support from both nvidia and AMD could be a lot better than it currently is, but I don't see how it's going to make support any worse.

Re:Bias much? (2)

DudemanX (44606) | about 2 years ago | (#40202657)

The article at Anandtech is less ominous and explains why this is actually a good thing with video chips and drivers as complicated as they are today.

http://www.anandtech.com/show/5880/amd-discontinues-monthly-driver-updates-releases-catalyst-126-beta [anandtech.com]

What the summary and article from the submitter are missing is the term WHQL. AMD has and always will be releasing beta drivers to fix games as needed just as Nvidia does. What they are stopping with this announcement is halting the monthly WHQL releases. To get WHQL certification from Microsoft the driver needs to be validated by MS for a week or two. By the time the drivers get certified they're already out of date. Cutting edge gamers almost never use the WHQL drivers and will use the "beta" drivers anyway.

The main people concerned with WHQL releases(OEMs) are ok with new releases every 3-6 months like Nvidia does it. The OEMs are only going to support whatever drivers they want to anyway so this really is a non-story.

wtf? (0)

Anonymous Coward | about 2 years ago | (#40201373)

For the life of me, I cannot tell ehat the editorial is trying to say. Is AMD evil for trying to update less often, after recognizing the strain this puts on consumers? Or are thry evil for issuing updates very often, because they want their consumers to be able to. Enjoy the use of their product, and there have been peoblems in the field?

Editors, wtf? Why do you let so much crap through?

As needed != more seldomly (0)

Anonymous Coward | about 2 years ago | (#40201385)

You do realize this could also mean quickly releasing hotfixes for games that need them, right? Of course you don't, everybody loves hating on ATI.

consoles (-1, Redundant)

Anonymous Coward | about 2 years ago | (#40201411)

And people wonder why PC gaming is only a shadow of what it used to be, and most gamers have moved to consoles.

Re:consoles (1)

Mashiki (184564) | about 2 years ago | (#40201679)

Really? Odd that developers seem to be abandoning consoles, and even console gamers seem to be abandoning them in favor of the PC. The last two years have seen a pretty good resurgence. If you want to see some odd stats, look at some of the indie titles that sold so poorly on consoles but sold so well on PC's that they made enough to make a sequel.

Re:consoles (0)

Anonymous Coward | about 2 years ago | (#40201765)

Yes, consoles never have any problems. Save for RRODs, poor performance and bugs in some games, including major releases like Skyrim (especially on the PS3), scratched disks. Minor issues, nothing to worry about.

Irrelevant (1)

Anonymous Coward | about 2 years ago | (#40201429)

The only graphics cards manufacturer that provides a complete free software driver is Intel.

Re:Irrelevant (0)

Anonymous Coward | about 2 years ago | (#40201637)

AFAIK Intel does not make graphics cards..... They just make graphics chips

Maybe they would not have as many issues (1)

wisnoskij (1206448) | about 2 years ago | (#40201569)

If they released software and patches when they were done instead of on an artificial time schedule.

Crysis 2 in DX9 (2)

Osgeld (1900440) | about 2 years ago | (#40201627)

heh ok, wonder how bad the demand for that is...

"check it out I got a i7 extreme fucking overclocked, 32 gigs of ram overclocked, quad ATI's also overclocked, 4 SSD's in RAID, and Windows XP cause DX9 is the shit yo"

cause no one plays crysis for the game, its a epeen ruler.

Re:Crysis 2 in DX9 (1)

Sycraft-fu (314770) | about 2 years ago | (#40202011)

Not the case for Crysis 2. Crytek scaled it back a lot, and actually focused on making a game rather than just a tech demo. When it came out, it was DX9 only. Later they released a patch that introduced DX11 support and had some bigger textures, but at launch it was a DX9 title.

STFU (0)

WillyWanker (1502057) | about 2 years ago | (#40201697)

Oh for Christ's sake shut the fuck up. Removing the need to release a new driver every month frees up time to work on any issues that come up with new releases. It gives them more ability to fix things in a timely manner, not less.

Spoken like an Nvidia fanboi.

Re:STFU (0)

Anonymous Coward | about 2 years ago | (#40202893)

Agree on the release schedule, if it is being done for the right reasons and not because they are laying off half their development team.

But calm down Dorris. This is about video cards being used for games. Not really all that important beyond the expendature of middling amounts of money.

This smells of... (0)

Anonymous Coward | about 2 years ago | (#40201705)

a new Catalyst driver every month like clockwork. Starting last summer, however, AMD began having trouble with high-profile game releases that performed badly or had visual artifacts.

Looks like someone is using Agile!

Window and Linux Drivers Both (0)

Anonymous Coward | about 2 years ago | (#40201759)

The thing that drove me crazy with their scheduled releases was that they often broke the working additions from the month prior. My friend has a 6870 and had to roll back his drivers to the previous month's for every release because the latest releases always introduced some new debilitating issue while the CAP profiles (individual game compatibility drivers) released after the latest drivers fixed the previous month's issues.

In addition to not being able to use proper 3D acceleration even using the proprietary Catalyst drivers for my Juniper series in card Linux, which leaves Gnome 3, non-2D Unity, and most GPU accelerated applications in general in an unstable or unusable state of limbo, there have been no obvious improvements that would have justified their release schedule, nothing that has improved existing stability or performance for general use, and nothing like a roadmap to indicate that they understand what the issues are or what they plan to do about them. If it does exist, I imagine it's a giant list of "Won't Fix"es with few "Critical"s about popular benchmark compatibility.

Even though I really like the idea of the low wattage (75W for ~6950 performance) 7 series, I look at nVidia cards every month waiting for the prices to drop so my graphics card's drivers won't be the limiting factor in determining which Linux distro I use.

subject and post and post time (-1)

Anonymous Coward | about 2 years ago | (#40201969)

WOW what a coincidence...this article speaks of AMD and the original post was posted at @11:40AM . read the time backwards and you get MA04.remove the 0 and you get MA4......the fourth letter in the alphabet is D.....so MAD - AMD :)

all....planned by the system.....the matrix has you!

I want a stable driver (1)

JDG1980 (2438906) | about 2 years ago | (#40202217)

Not everyone is a 3D gamer who wants to be on the absolute cutting edge of everything. Not everyone thinks trading off stability against a few extra FPS is a good deal.

Would it be too much to give us a stable driver, with maybe one update per year? By stable I mean no dodgy hacks, and no game-specific "optimizations". I mean a driver that won't crash, and that isn't afraid to be a little slower in order to do things right. Is there really no one else out there who cares about stability?

Re:I want a stable driver (0)

Anonymous Coward | about 2 years ago | (#40202437)

Nobody gives a S**T about stability nowadays.

It's all about games, since they are selling GAMING CARDS. AMD and Nvidia sells hardware that everyone (press included) tests with GAMES... People out there decides what card is faster testing GAMES, and even if certain games crashes, no problem , since the card is fast and updates will come. Granted, the nvidia GTX 680 was slower in certain OpenCL operations and software, and also they removed some features that games doesn't use from the hardware. But guess what: Almost nobody noticed.

Re:I want a stable driver (0)

Anonymous Coward | about 2 years ago | (#40202445)

Buy a FireGL/Quadro card. They are expensive, but stability is one of their selling points.

I'm all for it... (0)

Anonymous Coward | about 2 years ago | (#40202399)

... if it means fewer nags to update drivers.

It is about how much the driver is being worked on (0)

Anonymous Coward | about 2 years ago | (#40202425)

Frankly, monthly updates is NUTS. You quite simply CANNOT properly QC drivers with that kind of timeframe. Add in the out-of-band bugfix releases and you have an absolute shambles when it comes to control of the code and code quality. The fact that it has gone on this long is the REASON AMD is having so many issues. Spaghetti code on a massive scale, insufficient documentation, and ever more of same being piled on by programmers who don't have time to get a proper handle on the code, so they treat the symptoms instead of the disease.

Hopefully them stepping back will alow them to reign it in and start improving the drivers in a rational manner.

What are the options? (1)

utkonos (2104836) | about 2 years ago | (#40202471)

I've had stability problems with both ATI cards that I've owned recently. The older one was a 4xxx series that was totally inadequate, and my current one is a Radeon HD 6670 which should be adequate for most things, but really doesn't provide a smooth experience in Skyrim under Windows 7 nor is it the best in Linux. Compositing under KDE is not stable with this card. I don't use the closed source driver, however, under Linux, but I don't feel that I should need to use the Catalyst driver just to get KDE's eye candy to work right.

This begs the question: what is a good stable video card that can give modern games under Windows an enjoyable experience and also provides a solid experience under Linux with preferably an open source driver?

What difference does it make (1)

robbo (4388) | about 2 years ago | (#40202581)

if your OEM locks you out of driver updates in the first place? I've had no end of frustration with my Lenovo laptop and the fact that they unlock a new AMD driver about once per year.

Another reason to avoid Awful Macro Devices (-1)

Anonymous Coward | about 2 years ago | (#40202975)

Betweent heir shitty quality, lack of open-source involvement, and shitty tech support is it any wonder Awful Macro Devices is well behind Intel and Nvidia. It is a good thing they are going mobile as hopefully Intel will go that route and totally obliterate AMD off the fucking map. GOOOODBYE AWFUL MACRO DEvICES!!!!!

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...