Slashdot Mirror
NVIDIA's Drivers Caused 28.8% Of Vista Crashes In 2007
PaisteUser tips us to an Ars Technica report discussing how 28.8% of Vista's crashes over a period in 2007 were due to faulty NVIDIA drivers. The information comes out of the 158 pages of Microsoft emails that were handed over at the request of a judge in the Vista-capable lawsuit. NVIDIA has already faced a class-action lawsuit over the drivers. From Ars Technica:
"NVIDIA had significant problems when it came time to transition its shiny, new G80 architecture from Windows XP to Windows Vista. The company's first G80-compatible Vista driver ended up being delayed from December to the end of January, and even then was available only as a beta download. In this case, full compatibility and stability did not come quickly, and the Internet is scattered with reports detailing graphics driver issues when using G80 processors for the entirely of 2007. There was always a question, however, of whether or not the problems were really that bad, or if reporting bias was painting a more negative picture of the current situation than what was actually occurring."
Just sayin
I really hope there's some way I can use those same drivers under linux!
Oh....wait.
There are shills on slashdot. Apparently, I'm one of them.
The linux drivers for nvidia suck too, nvidia clearly take a long time to get up to speed on new operating systems, it's one reason I no longer use them. Having said that, they're pretty damn solid, so its most likely becuase vistas so mucked up when it comes to drivers.
For a hardware manufacturer to build h/w for the Windows PC? Is there some SDK or some specified method by which the co. can write device drivers? Or is it done by guesswork and hacking, and paying Microsoft for the honour?
I'm seriously puzzled why and how device drivers can cause such major issues in Windows but seldom in Linux (identical hardware, mind).
If you keep throwing chairs, one day you'll break windows....
What about the other 62.2%? ATI. ;)
Quit jabbering on the phone while driving. You are not that important.
Well, this wouldn't be the first time Nvidia drivers are responsible for instability.
I remember when the first nForce3 drivers came out that had those IDE problems. And the continuing problem with the SW drivers. Man, I thought something was seriously wrong with my new rig. Nope, just the drivers....
Do not downmod posts "overrated" simply because you disagree with them.
This is descending below lawsuit territory, I'm starting to think that the whole PC hardware industry should be taken out back and shot. They supported MS in the release of an OS with crap under-powered hardware with smiles and big adverts, in full knowledge that these systems would never work or just were not ready for Vista.
"The Wow Starts, oh around 2009 if you'll just let us fix this, upgrade that and force you to buy some new stuff" Should have been the tagline for Vista.
If he's the Walrus then can I be a penguin please?
ATI was 9%
I can verify this problem. I bought a new Dell XPS 720 with the fancy shiny new Nvidia 8800 GTX, and let me tell you it crashed all the time. I tried playing Crysis, and found my screen would go totally black. Next thing I know I'd get the windows message that my video driver has crashed and was handled by Vista. For the longest time just watching DVD's through Windows Media Player I'd get the same thing, working with VS 2008, or anything that required using a monitor I'd sporadically get video card crashes. My blog is full of rants from those times periods when I had to swich over to my XP laptop just to do anything other than get pissed off at Vista. So now I can be pissed off that I didn't buy an ATI. Oh well, it's a giant quagmire either way you look at it.
Parent links to shock site - do not click. This is much more amusing, if you want to click on something ;)
Is crushing a suspect's child's testicles illegal?
John Yoo: "No, [if] the President thinks he needs to do that."
First, given the popularity of some of their chip sets, this probably isn't bad. Quite a few systems out there with the 6100 and 6150 UMA chipsets. And what about the other 71.2 %
Could be the UMA in Vista is unstable? I am using a 8500 GT and I haven't crashed once. No UMA in use though. I question those running UMA for Video on Vista, Vista needs a beefy video.
I do have slow disks, slow network I/O and slow... but no video issues. And the best part is that it also works with Linux/Solaris. (8500GT).
http://www.google.com/search?q=vista+driver+problems
boycott slashdot February 10th - 17th check out: altSlashdot.org
Nvidia have a shamefully lax attitude to the stability of their drivers even under XP. Try searching google for NV4Disp.dll and you'll see that there is an issue that still causes BSOD's years after it was first reported, ironically the latest drivers only make the issue worse. This latest news will only make sure that my next card will not be from Nvidia.
Quite often hardware overheating issues will cause the kernel to crash in the nVidia driver. This will probably be the first thing nVidia claims and it's pretty true.
Perhaps there is some blame on nVidia, but there would have to be equal blame to Microsoft for not being more compatible with windows XP drivers and spending far too long in development.
28.8% of Vista's crashes over a period in 2007 were due to faulty NVIDIA drivers
Well then it's a good thing their driver support is so crappy with Linux!
Oh wait...
More seriously, I rag on Nvidea for poor Linux support, and this is more of a chance to bash them, but their drivers work fine under XP. If Microsoft provided better documentation of their APIs, as the EU has been demanding, perhaps writing drivers wouldn't be such a pain in the ass?
I also wonder why closed source vendors don't open their code. They don't have to release it under the GPL, they can reatain all their copyrights, just publish the source. How could it hurt them? They retain copyrights and presumably patents so it's not like anyone could copy them.
Is closed source closed so that nobody will realise just how abysmally shitty their kludges are?
If your OS crashes, your OS is crap. Microsoft, fix your OS and publish the code. Nvidea, fix your shitty drivers and open the code. Don't give up any rights, just open it.
I'd like to see copyright law changed so that executables can't be copyrighted unless the source is also provided. How can IBM tell what parts of their code they stole from SCO? Of course the answer was "none". Time to reboot copyright law!
-mcgrew
mcgrew's razor: Never attribute to stupidity that which can be explained by greedy self-interest
I tried Vista on two machines running ATI cards- a desktop and laptop. They crashed an average of 2-3 times a day (BSOD). In all cases- Microsoft blamed the ATI video drivers, which I kept updated from ATI and Microsoft's own updates. I got fed up with it after a month.
I dropped Windows completely and went with Ubuntu Linux. It has issues with video cards too, but aside from not being able to enable some eye-candy- it almost never crashes. (Usually the only time it does is when I try to tweak video settings or try new drivers.)
Video card drivers are probably the number one problem with computers right now, in ANY operating system. It wouldn't surprise me if they are responsible for a lot of game console crashes too.
Did MS certify they drivers? If so, it's still _their_ fault
I assume the drivers for such a critical component were officially 'certified' by Microsoft. In that case, it's not NVidia's fault alone and Microsoft should also be jointly accountable for the problem - since such certified drivers are supposed to be thoroughly tested by MS.
/That or Windows should just stop warning users while installing uncertified drivers, since it doesn't really mean anything either ways.
There was a time back in the Detonator days that Nvidia had the best drivers going. Each generation has gotten a bit buggier. My first Vista machine has a 7900GS, not even an nv80 series chip and it had horrible problems with rendering and stability. It never brought the whole OS down but the number of times the driver itself would crash and restart (a neat feature in Vista by the way) was beyond count for the first few months.
They've gotten far better now at least, but they (Nvidia) really dropped the ball on that one.
That's nothing. ATI's driver caused 100% of my crashes on Sunday. I picked up an AGP 2600XT on clearance ($60) for my old MCE machine. The machine successfully booted once with the new driver installed. After it crashed I couldn't get it to fully boot again (only safe mode worked). Spent about 4 hours screwing around with it. Back to the store it went. I think I've only had one ATI card that worked well, and that was a 9800Pro.
As a developer, I was forced to upgrade my work environment to vista in order to support users. During pretty much from when vista shipped till about a month ago, Vista would BSOD at least once or twice a day, very frustrating. Since I don't play games and the 3D hardware's only use was to display video (certain video rendering modes in windows use the 3D hardware) it was quite frustrating. However, after several updates from both NVIDIA and Microsoft, Vista has been very stable for me over the last month. However! NVIDIA's Vista driver are no where near complete compared to their XP drivers. There's a lot of missing functionality, especially when dealing with analog (S-Video) output. Considering vista was released over a year ago, this is very frustrating.
Zoom Player Lead Dev.
Intel has highest marketshare of video adapters probably around 40%. Their video is mainly used on laptops and the most intensive thing they do is check email and surf the net. Comsidering that number would include a lot of wireless drivers though it's probably a good figure for them.
.05% performance, i doubt anybody bothers to upgrade their intel drivers.
Nvidia has about 35%, Ati has about 20%. So obviously Nvidia will get more crashes because they have higher market share. People who buy NVidia and ATI probably use their computers a lot harder, play games which are always likely to crash something and may overclock too. People who own Nvidia probably upgrade their drivers 3x more often hoping to eek out that extra
The point is the stats mean nothing when you take usage, market share and all the other factors into consideration. Give me a graph detailing the of number of crashes per cpu/gpu instruction.
..given that drivers have always been windows's leading cause of instability. The only performance-impacting difference between datacenter and other editions of windows servers (a difference of two 9s) is that datacenter only ships on OEM hardware with signed drivers.
Frankly, given that Microsoft are effectively the OEM for the majority of drivers that ship with windows, I'm amazed the figures for crashes involving their drivers are so low. I think this demonstrates clearly the amount of pain consumers suffer through vendors (nvidia) not adequately investing time and money in training and R&D for driver development on a new platform. I know I've been feeling the same pain on linux for years.
Maybe MS could contribute some developers to Nouveau project and then insert hooks into it for their specific kernel?
839*929
.. but what percentage of Vista PC's have Nvidia cards (and what percentage of those pc's reported a crash?) You can't really compare it to the percentage of crashes reportedly caused by other drivers without knowing that.
NVidia has been really good in the past for me, for drivers, especially compared to ATI. Until recently. I run Vista Ultimate 64 with twin 750GB Seagate HDDs running in RAID 0 via an NVidia raid controller. I used to run with windows update on and set to automatically update at 3AM each day. Until NVidia released new RAID drivers a week or so ago.
I would wake up each morning to find my computer constantly rebooting. It would blue screen and I couldn't even make out the error before it was off the screen. I managed to get Vista running with Last Known Good configuration (amazing to me, I never see this work) and the first time I checked the last update and it was the NVidia RAID drivers. I figured what the hell, maybe it was a fluke.
Well, the third day in a row of finding the computer in this state in the morning and I finally cancelled the RAID driver install. The next morning, the computer was fine.
It was the NVidia drivers... and possibly Vista.
I do not respond to cowards. Especially anonymous ones.
We're a dell hardware shop. We buy on a 4 year cycle, every machine gets replaced every 4 years with the latest latitude line shipping model of laptop. In this past few cycles they've been NVidia based. They all have 2 gigs of ram, sata hard drives, dual core higher end processors and of course, NVidia Mobile chipsets. So, all 800 people at my company with nvidia chipsets cannot deploy vista until a) the drivers are fixed. b) the hardware cycle comes up in 4 years. All the people getting new machines right now are perfectly happy because the hardware is supported, but just those purchased 6 months ago and before (D820's) are not capable of running vista with dual monitors without gambling on whether or not they will be alive after a weekend on screensaver.
Speak for yourself.
This is why i use my computer without graphics card. Nor screen. I am guided purely by instinct.
Nvidia drivers have been working perfectly on my Sony VIAO SZ460NC laptop with Vista right from the start. In fact, I even have an Nvidia driver that works on my FreeBSD partition.
When I had Acer laptops with their crappy ATI graphics, OpenGL never worked. I had 3-d modeling programs that only worked in 2 dimensions. It didn't matter what ATI driver you had. You could only spin the objects in the x and y dimensions. Ubuntu's ATI drivers never worked either on the Acer laptops.
Now with Nvidia, I get true 3-d on all the OS's on this laptop.
I only wish it had worked that well for me. I did something very similar - MythTV machine (AMD64, dual-core) connected to LCD TV via HDMI. Huge pain in the ass to get working. I ended up having to dig deep into the nVidia documentation to find the various override switches to tell the card/driver that, yes, really, you can send a 1080p signal to this device, honest, I swear! Once I got it working it was a dream, but I spent at least two days dicking with it.
What's comical for Microsoft is that they would go and change the driver models for everyone for their new OS, and then blamed the resultant bugfest from the imposed change over on all of its business partners. Way to go Microsoft! You guys are a bunch of class acts!
This is my sig.
I'm running the 169 series on Ubuntu Gutsy and I haven't had many problems. Sure, every once in a while (maybe every couple of weeks) X will crash (but not the whole OS). That's happened since I started using the nVidia drivers. Other than that, they've worked great for me. I mostly wanted 3D acceleration so I could use Google Earth and possibly experiment with compiz at some point (which I still haven't). Maybe if I was running compiz on my desktop I'd see more frequent problems.
Alright listen up you numbnuts. The percentages basically equate to the market share. ATI causes just as many crashes. The total percent is lower because more people have nVidia cards.
By the way, I had absolutely not a single problem with my G92 based card in December/January running Vista 64-bit.
...or are the Vista users on here beginning to sound almost as rabidly fanboi-ish as the Apple ones?
...Microsoft told us that this wouldn't happen on Vista.
Something about moving graphics drivers out of kernel space or something like that.
I tried vista when it was new. After installing nvidia's driver the system would crash whenever I tried to change resolution or play a game. Apperantly this was a common problem with the Geforce 7900. I tried some workarounds but none worked very well. I ended up removing vista (not only because of the crashes).
A few weeks ago I tried installing vista again. I hoped that the problem would've been fixed with SP1 and nvidia's latest driver. But no, the problem still remains after over a year.
I also submited a bug report to nvidia a year back but never got any response.
Since apparently you can have a class action lawsuit for drivers not working, lets open up the floodgates and punish the manufacturers for not having compatible software! And why stop at video drivers? Lets sue all the makers of legacy hardware. And wifi hardware. Have an OLD 5 1/4 floppy? Sue! Have one of those old HP video-now PCMCIA cards? Sue! Sue sue sue!!!
My only serious problems with Vista have been related to NVIDIA drivers. The latest driver update crashed my machine on a regular basis. I reverted to the previous version (that was in early January) and haven't had a crash since. My laptop has been more lucky - the latest NVIDIA driver update hasn't affected it (different card, of course.)
Most of these driver incompatibilities were actually caused because microsoft changed the driver structure at the last minute which basically shot a lot of the manufacturers in the foot at the starting line. If this class action lawsuit goes through... how likely do you think NVidia and ATI are going to be to jump on the bandwagon for Windows 7? I mean, fool me once, shame on you. Fool me twice, shame on me. I can't imagine being the victim of a multimillion dollar class action lawsuit because of microsoft's incompetance is going to make them the best buddies. Then again, I wonder if nvidia and ati have the right to sue microsoft in response should this current class action lawsuit go through? They developed to the specs microsoft had given them, so if microsoft changed those specs at the last minute... seems kind of uncool to me.
Most of my company's servers and desktops from the 1995-2000 era (some 600+ machines total) had Trident video chipsets either onboard or on a discrete video card. They just worked for the most part, as I recall.
Of course, there were many "inexplicable" crashes in our environment during the Win9x / NT4 era (blue screen or freeze without a corresponding crash dump file), so who knows. We assumed those were from faulty IDE/SCSI/RAID drivers or hardware, since no crash dump was written to disk. But that was really just guesswork - we had no other tools available to gather the required data for analysis (users just restarted their machines before IT could investigate). It was something we just lived with until Windows 2000 came out, and everything magically improved.
yeah right... its a driver issue of course :P
maybe some kind of strange virus or malware
I have some weird problem with my 7900GT with Vista where it goes nuts if I plug in the supplemental power cable. Without it, the card at least outputs correct video, but it dials back performance if it can't draw enough current. When I plug in the power cable, it boots up fine, but when the nVidia drivers load, my screen goes nuts like it's not syncing properly or something.
The real issue is still being missed here. With all of the promises that have been PRed our way and the years of available technology and programming technique to back it up:
It shouldn't be possible to crash the kernel because of a driver. (well maybe a disk driver...)
I can see the driver going bonk and having to reset itself but it shouldn't be possible to take down the whole system. It is supposed to be the end of such things..
And how often would recent stable builds of the Linux kernel run without crashing if you discounted problems caused by video card drivers?
Where are these crashing Linux systems you mention? Is that what happens if you use Nvidia's non free drivers? I have used them and the do lower system stability but not to Windows levels. Nearly 100% of my crashes were due to Nvidia and they vanished when I moved to nv, along with pretty special effects. The crashes were just Xwindows restarts not full system stopping BSoD like you get in Windows. What's more, Intel graphics chips offer effects that are almost as good with free drivers that have no stability problems at all.
http://slashdot.org/comments.pl?sid=216934&cid=17629948
Losing time on nvidia reverse engineering... all this time could have been used to work on the new Linux graphic stack. Thank you AMD (need full R7XX specs).
I never really had problems with the Trident drivers crashing or anything like that. My problems with Trident are mostly related to personal experience. In the late 90's, I had several Trident SVGA cards that either worked like crap or didn't work at all on Xfree86. I bought a laptop back in 2001 that had a Trident chipset. In Linux, I could never get more than 640x480, 16-bit color, or 800x600 (the LCD's native resolution) in 8-bit color. It was horrible. A few years later I had the opportunity to get a good deal on a laptop at a bankruptcy auction, and once again, I get onboard Trident video! ARRGH!! The 2D acceleration is some of the worst I've ever seen. Video playback is marginal at best, since the CPU all gets eaten up pushing bits around the screen (I benchmarked video using mplayer and a null video output device, so I know the CPU/codec are plenty fast enough). It does OK in Windows XP, but that just makes it more frustrating, as I know the damned thing can do better 2D acceleration, but no matter what I tweak in xorg.conf, I end up with two choices - "bad" or "hideous."
Someday I'll buy a laptop that's not from the late Triassic period - one with on-board wireless, Bluetooth, and nVidia or Intel graphics. Until then, I'm cursed with Trident.
NVIDIA vs ATI drivers - I don't really care.
"It worked for me" - I don't really care.
Statistics on the cause of crashes - I don't really care.
Anybody running unsigned drivers and experiencing crashes - I don't really care
Hang on. Let me explain.
The fact that you can STILL crash a Windows machine with a dodgy driver - that I care about. I thought everything was supposed to be userspace. I thought the error-handling was supposed to be better. I thought that Windows was supposed to be more stable and secure. I thought people who were using signed drivers were supposed to be "approved" and relatively crash-free.
Unsigned drivers? You can't support that no matter who you are, unless you're confident they are PURE userspace - they could be doing anything (like the 3DFX drivers that used to open access to all sorts of things it shouldn't in order for a primitive user-space part to actual drive the hardware). That's why you have to click that "CONTINUE Anyway" button with the dire warning. That's the Windows equivalent of kernel tainting. Once you've done that, nobody cares. The fact that most XP drivers are still using uncertified drivers is a bit of a problem but I can understand the reasons why. But you can't blame MS for crashes in uncertified drivers under XP. I thought Vista was supposed to be different, though.
If a certified driver is crashing that often, then you have an entirely different matter. The certification effectively becomes worthless. Nobody trusts it. Therefore every driver manufacturer ignores certification and just tells users to click "Continue". Then you will have nothing BUT uncertified drivers. Catch-22.
Blue screens should not happen. They certainly shouldn't happen often enough that people have coined the term "blue-screen" or BSOD to mean a crash. When they DO happen, when the driver goes absolutely nuts and starts stomping memory, aren't things like DEP and the user-space driver model supposed to STOP that happening and recover in some half-decent fashion? Or shouldn't the machine at least what the cause was and provide the user with some hint of what went wrong (i.e. "You installed an uncertified driver. Tough.").
Let's compare for a second - Linux kernels crash too. They crash much more often if third-party drivers are installed and nobody really cares about that except the third-party and their users. When they do crash, there's not much you can do but most of the time you'll get all sorts of debugging information and usually you can carry on. You might lose X, which may or may not load up again - I have a laptop that likes to crash X if I run more than one copy of Xine at a time but the worst that happens is X dies and restarts and then carries on working for hours/days/weeks as if nothing had happened (and yes, I need to update the kernel/X on that machine!) but things keep on working as best they can. You can do pretty much what you like in terms of software but the worst that'll happen if you're not actually loading a kernel module or patching a kernel or playing with kernel-level features is a software crash and be chucked back to the command-line. Sometimes you might even end up taking out X, like my example above.
You can rip out the harddrive and *make* the kernel crash but most of the time things will carry on, just without the component you ripped out (i.e. the IDE layer may die, but it'll still keep running as best it can without it). Even when Linux comes to a complete halt and freezes, you have debugging information and logs with which to narrow down the cause yourself, without needing to consult Linus himself.
When Windows crashes (even with certified drivers and clean installs), there's bugger all to go on. Half the time the event log doesn't show anything at all. The second you see a blue screen, the computer is down and there's little arguing. There's zero information to go on. You have no idea what caused the crash at all because usually all you get is a generic STOP error and a
My first vista machine was using a nvidia agp 128mb 5200FX. An old card. It worked never had crashes or stability issues. Also never gamed in vista on it. This was at work so no gaming. Alli was trying to do was to see what all the fuss was about. Shiny stuff in windows, and you got a second way to switch between running apps. Oh, and a few new screen savers. Vista is still XP with a face left to me. The only other thing is old apps do not run. Newer things work, jus the old stuff (pre year 2000 for us anyway) has issues.
Nvidia already dishes out specs and programmer time for 2D drivers. It's second rate treatment but it's a start. Nouveau and both Intel and ATI's full bore release of documentation will light a fire under Nvidia that should produce more.
http://slashdot.org/comments.pl?sid=216934&cid=17629948
In soviet russia drivers blames you!
"entirely" should be "entirety".
ATI has had the worst drivers ever. I can not imagine them being any worse. OpenGL they would just crash, even on Open GL screen savers. You would get errors in the event logs that ATI.drv had went into an infinite loop. One Open GL application would would and the other one not, 2 weeks later they put out a patch and the non working one now worked and the working one now would crash. Finally after one of my more critical Open GL apps where I earned my living from would crash on any Open GL operation I finally switched to NVidia. Then I found out how crappy they really were. Try uninstalling the driver and the ATI apps. You would get the error that it could not uninstal because it could not find the ATI hardware. Well Duh I removed the piece of crap.
I will never ever buy a ATI card again
The linux drivers for nvidia suck too, nvidia clearly take a long time to get up to speed on new operating systems, it's one reason I no longer use them. Having said that, they're pretty damn solid, so its most likely becuase vistas so mucked up when it comes to drivers.
Are computers fast enough these days to allow for protected memory for drivers? Given that most crashes are from drivers it would be nice, but how this is realistically?
Jumpstart the tartan drive.
These theories for why drivers are closed source keep coming up. Dave Airlie's Kitten Killing article which states some (but not all) of your points:
1. The licensed parts may simply have to be rewritten by the community or go unimplemented. It's tricky but not impossible (if Sun can do it with Java...).
2. These folks allegedly employ the smartest people out there. They are probably already capable of reverse engineering binary drivers. If your competition can reverse engineer hardware as complicated as today's graphics card from a driver, well you had better be really concerned...
3. If it were to be released under and open source license then it wouldn't be illegal to redistribute the code. People will already solder cards to get more speed out of their cards so I would think depending on software lockouts would be too risky.
4. Certainly this is becoming more of an issue.
5. If you are making guarantees to someone like military maybe you should make sure that you can really fulfill your obligations?
1 can be worked around and 4 is turning into a growing problem. If you've done 5, well you are always on rocky ground.
Microsoft have changed their driver model because of all the DRM that has gone into Vista. One part of that driver model kills any driver that can even remotely result in giving up secured content. Most of the stability problems are a direct result of this killswitch and other DRM idiocy. The reason the certified drivers work much worse than the drivers from the vendor is that the vendor drivers have many of theese things turned off or worked around. Anyone interested can read about this and other crap that DRM in Vista is a direct result of, like file copying. Just look it up at microsoft.com, its a pretty horrid reading.
DRM suck in any way possible and evidently also brings endless troubles for all the parties involved.
HTTP/1.1 400
I don't get it. I understand why faulty code in a graphics driver might make the screen go black, or blue, or display garbage, or whatever, but I don't see why any operating system should allow it to anything that would crash the system.
The essential relation between a graphics device and an operating system is unidirectional. The graphics subsystem acquires data from the OS. The OS can and should be read-only from the point of view of the graphics subsystem.
If it wants to, the OS might well allocate some RAM for the sole use of the graphics driver, and the graphics driver might choose to do some local processing in that space, but that should all be a walled-of area and the OS ought to be able to enforce that.
Undoubtedly the graphics card will want to do high-speed data transfers using DMA or whatever specialized capabilities the hardware provides, but again, I don't see why an OS shouldn't be able to manage this in a safe way, so that if the graphics code malfunctions it only impacts the graphic operations themselves and doesn't affect or bring down the rest of the system.
Obviously, an OS can't protect itself entirely from faults in a driver that handles operations that the OS requires for its own operation... the driver that access the disk drives, for example. But a graphics driver? Surely it can and it should.
For decades now, Windows has been crashing and Microsoft has been blaming the drivers. Microsoft ought to write the critical drivers (like those for disk drive access) itself, and it ought to design the OS so that non-critical drivers (e.g. graphics, sound output) simply can't damage the OS.
"How to Do Nothing," kids activities, back in print!
From Wikipedia:
Source article
This is the only way the "trusted path" will work and it would be convenient for Microsoft if people and institutions did not realize that this is an unacceptable way of doing things.
No calls now, I'm
My experience is the exact opposite of this study. Most people I know that are having major issues with Vista have ATI cards, and myself and others that have had no issues at all with Vista all have nVidia cards.
Nice troll, but I wasn't attempting to make some clandestine endorsement of Linux. I was simply mentioning that fact because I've seen bad video drivers do both - take down X, or take down the whole OS. It's a distinction that probably only matters to those of us who hang out on Slashdot for a bit of geek talk, rather than to troll.
Don't the ATI/nVidia Omega drivers work in Vista? Assuming they do, it seems most of the crashes were due to people being ill informed or giving up rather than the fault of either manufacturer. Personally, I would place blame on Microsoft before any manufacturer as I am sure they have *something* to do with the driver design process and making sure nVidia and ATI are properly informed. The ultimate blame, of course, rests on the users for daring to install a Microsoft product before SP2... http://www.omegadrivers.net/
That graphic drivers are tied so closely to the OS bothers me. Vista still doesn't seem like an OS to me as much as a resource hogging application that I MUST have running in the background at all times.
Hi, I Boris. Hear fix bear, yes?
It's the only time with this computer I've had a complete lock up that wouldn't even respond to Alt-SysRq-B. Had to hard power-off.
I've got a Winfast Leadtek 7600GT running in an Acer Aspire with Athlon 64 X2 4000+ and 3G RAM - crashes after about 10 mins of gameplay on my Vista, also appears to crash when running 3D apps on Linux.
It'd be interesting to hear some NVIDIA programmers take on the situational details. Anyone who has developed commercial software for Windows, at any layer from ring 0 hardware interrupt handlers to supposedly platform-independent scripts that process nothing but high level abstracted messages, knows from experience that Microsoft breaking you is the norm, not only across Windows product changes, but even with service packs. Microsoft does not care about third party software reliability in the slightest, and will not only change critical APIs knowing that they will break third party software, but will sometimes deliberately change critical APIs for the purpose of breaking third party software. It's quite possible that the class actioners need to sue Microsoft, not NVIDIA.
Many people blame Vista DRM for the video driver problem across the board (not just NVIDIA). Whether that's true or not, we should all point the finger at DRM until MS can prove otherwise. The more we can associate MS with DRM, the better life will be for everyone else. In a perfect world, MS dumps DRM, but not before conceding a huge chunk of market share and restoring competitiveness to the industry. Given all of the corporate love for DRM, it will take the loss of a few billion dollars in losses to get the corporate golfers back into the office to think about what went wrong.
All of your applications probably still crashed.
I'm rather surprised that I didn't see a lot more mention of this in this discussion.
I'm an American. I love this country and the freedoms that we used to have.
I've not tried the Intel cards, but I've always had good luck with Matrox.
These statistics were calculated using Excel.
Have gnu, will travel.
Trident was ok, but I always had better luck with S3.
How many people around Slashdot actually link to twitter's journal? Exactly five, including twitter. Isn't that just the biggest coincidence?
Do you realize that your first sentence conflicts logically with your second?
..then it's not secure. Period. (Whether it happens to work or not, you don't know how well it works, and therefore it can't be trusted.)
As copyright owner of this comment, I authorize everyone to defeat any technological measure which limits access to it.
That's nothing, the ATI driver caused 100% of the crashes on my Ubuntu box. Until I installed an nVidia card and 100% of the crashes were caused by the nVidia driver.
Well, this wouldn't be the first time Nvidia drivers are responsible for instability.
Which begs the question: Why the hell hasn't MSFT done its QC/QA job and taken NVidia to task? Why can't it use its dominant, near-monopoly status in the industry for GOOD for a change, instead of for whipping the competition?
These "root cause analysis" numbers were trotted out by MSFT in its defense argument in the "Vista Capable" lawsuit. How does this absolve MSFT of anything here? Wasn't the "Vista capable" label supposed to MEAN something? Don't they do thorough regression tests of systems to make sure that they really ARE capable of running Vista? Why were machines equipped with NVidia chipsets, which relied on drivers responsible for nearly 30 percent of crashes, allowed to be sold as "Vista capable"? In my opinion, it just bolsters the argument that MSFT was negligent and misleading in it's Vista Capable campaign when machines with such crappy drivers are allowed to be pushed out the door on vague assurances that by the time Vista was in wide release the bugs would be squished.
Apple's computers "just work" because they have complete control over the system software and hardware stack and can ensure total quality control. Linux and BSD offer superior reliability and generally better performance because the software is developed and scrutinised by a huge community, and hardware is closely studied or specs are fully disclosed (interestingly the worst Linux drivers happen to be those where hardware specs are most closely guarded). MSFT needs one or the other. They aren't in a position do specify exact hardware as Apple is and Ballmer would be throwing chairs left and right if source code was ever opened up, so they must insist on more control over the drivers supported with their OS and demand more access to hardware specifications so they can assist in making quality drivers. If MSFT ever hopes to have a TRULY quality OS, and they are not willing to move towards a Free software development model, they absolutely MUST exert as much control as possible over the development of system-level software for thrid-party hardware vendors' products (drivers in particular).
His entire "paper" was based on anecdotal evidence gleaned from message boards and willful "misinterpretations" of Microsoft's documentation.
And when he was called on it, Peter Gutmann hemmed and hawed, shuffled some things around on his website, but never defended his results from a single challenge.
But you already knew that, Twitter.
Lesson learned: Don't run unstable OS builds. Ever.
I have used numerous nVidia cards with many games and have never seen an nVidia driver crash on me, in Windows or Linux.
Maybe some of these crashes are caused by the flaky motherboards and memory that the drivers run on, or power supplies, and it's just that code in those drivers is what pushes the hardware to the max and makes it crash.
[an error occurred while processing this directive]
Actually Microsoft had been talking to the graphics IHVs about the new Longhorn "Advanced Driver Model" as early as spring 2005. Both ATI and nVidia had representatives (i.e., developers) working closely with Redmond during that time. The Longhorn/Vista display model became known as "WDDM" and was more or less locked down, from what I understand, by late 2005. By the time of WinHEC 2006 (April), they were already talking about WDDM 2.0, as you can see from this presentation. If you take a look at the slide deck, ATI's Tim Kelley actually delivered part of the presentation on WDDM 2.0.
Frankly, I don't think nVidia invested enough energy in making high-quality Vista drivers in time for launch. They had approximately a full year of Betas, the same time that ATI and Intel had. The Vista Beta and RC programs had hundreds of thousands of users around the world, for which Microsoft collected crash dump data (which is the same type of data mentioned in this article, collected BEFORE launch). Yet even with this time, and the user crash dump reports, clearly by launch in January 2007 nVidia still wasn't ready with robust drivers.
The evidence here really does point at nVidia, no matter how much you want it to point at Microsoft.
What's that "oldid" part of that URL? I don't see that in most wikipedia links I follow... you wouldn't be trying to pull a fast one by simply noting that Wikipedia in the past has been defaced (and quickly fixed, a whole 15 minutes it was up), yet trot it out as an operational fact, would you? Because that would be dishonest, and I wouldn't think an AC would be dishonest.
My blog. Good stuff (when I remember to update it). Read it.
I switched from a shitty Intel integrated graphics card to a nVidia 8600GT a few months ago, no crashes yet! I'm running Windows Vista and Linux on this computer (Vista I only use for gaming), no game has ever crashed (except for one, but I overloaded it by accident), and no OS issue has ever happened caused by the graphics card in Windows or Linux. However, you know how many times the drivers for the shitty Intel 845G integrated graphics card my computer came with caused the computer to crash? Too many to count! When I would try to perform simple and specific tasks (for instance, load a specific level or play on a specific online level in games), the program would crash, and it would spit back a c0000005 Access Violation error (this has NEVER happened since I got the nVidia card by the way), or a message that says "Graphics driver intel845g had to be restarted". And then there was the time Vista completely fucked up because of the driver (One day, I could NOT get on Windows because of a c0000005 Access Violation which caused a BSoD). I had to REINSTALL VISTA! The sad part: Almost ALL casual user PCs have Intel graphics cards in them! Most Dells, Gateways, and HP PCs in the $500-700 range have these cards that can't run any games! There's the real problem!
One can always count on any thread on /. about Windows Vista to generate posts full of verve, passion, and technical ignorance. So far the responses have included long rants about NVIDIA's (and other video vendor's) policy towards Linux drivers, the usual anti-Vista FUD from posters who are proud of their lack of actual experience with the OS, and even the old "DRM sux" chestnut.
XPDM (the current nomenclature for the Windows Driver Model for Windows XP video) was the end-point of a decade of side-by-side work between the DirectX graphics technology (DirectDraw/Direct3D) and the core Graphics Device Interface (GDI). It worked, but only when the OS itself never used 3D features at all. In fact, 3D applications were extremely limited in their ability to multi-task the GPU, and writing XPDM drivers had become extremely complex. After years of experience writing & optimizing drivers for Direct3D9, it was obvious that the bottlenecks for future performance growth were in the driver stack itself. Many years of work lead to the development of WDDM (the nomenclature for the Windows Display Driver Model for Windows Vista video) to address GPU-sharing, 'small-batch' performance overhead, and to unify the GDI/Direct3D APIs to simplify driver development.
Drivers have always been a major source of crash reports. This is fairly obvious for any OS that has a 3rd party plug-in model: MSFT can test the OS with the drivers it ships with, but it cannot test every possible combination of hardware in the market. Windows Error Reporting (aka WATSON) provides numerous data points that help MSFT find these problems, and video drivers crashing in kernel-mode remains a key source of crashes on XPDM. Part of this comes from the complexity of supporting both GDI and Direct3D DDI at the same time, but a lot of it comes from the problem space inherent in programmable shader GPUs. XPDM drivers include shader compilers, and these code bases are like most compilers for non-trivial cases: difficult to get right 100% of the time. Therefore, one of the design features of WDDM was segregating the video driver into a kernel-mode piece and a user-mode piece, so that crashes in complex shader compilers would result in user-mode applications crashes, not BSOD. I note that the original article cited in this thread doesn't state if these are application crashes or BSOD crashes. WDDM didn't just change for the sake of change, but to invest in the needs of the next 5-10 years of video graphics performance, stability, and security.
While MSFT, the hardware vendors, and end-users would all like to have seen 100% rock-solid WDDM drivers and full performance optimizations across Direct3D9 (XP-era games), Direct3D9Ex (the new Windows Vista Shell), and Direct3D 10 the day Windows Vista shipped, the work involved was immense and the timing very tight. AMD/ATI & NVIDIA were developing major revs of their hardware which market realities demanded worked well on XP (and all the benchmarks at the time would judge them on XPDM), at the same time they were supporting an entirely new driver model and new API, had to support both 32-bit x86 and 64-bit x64 native kernel-mode drivers to get full coverage for the transition to 64-bit mandated by the Windows driver logo programs, and deal with the technical challenges 512 MB+ VRAMs created for SLI/Crossfire and PC BIOS compatibility. All that while investing in their own initiatives (CUDA, OpenGL, Linux, etc.) and dealing with things like the AMD / ATI merger.
The transition from XPDM to WDDM is no more messy than the transition from Windows 9x/ME was to NT/2000. There are a lot of moving pieces, a lot of actors to coordinate, and a great deal of technical challenges to overcome and new optimization expertise to develop. As with the previous transitions, it took a year or so to get the kinks worked out, and today the latest WDDM drivers are in pretty good shape overall.
Now back to your regularly scheduled Schadenfreude...
ook. ook ook, ook. uhuhuh ooooooook. oook oook, ahhha, ook ook.
(trans: if enough of us work at it, we're bound to make the perfect driver)
int main()
{
int *i;
while (1)
i = new int;
return 0;
}
And just watch them kick their keyboard across the room shouting "Freakin LAG!" he he.
The point of this post is to demonstrate that crashing a computer is not as hard for a programmer to do as people might think.
Censorship is the opposite of education. If neo-darwinism were defensible, people would not need to try and censor ID.
The main reason I switched to linux 10 years ago was some CLI geeks showed me all the advantages of running things like startx -- :1. Not to mention multiple users logged onto my hardware with their own x sessions running. but if I have my FTP transfer running on VT1, and wget doing something somewhere else, and my webserver running too, yeah I guess I would rather restart X than reboot my machine.
So while I understand your situation, I disagree with your assessment. Maybe you should start taking advantage of the coolness of the penguin.
p.s. I never struggle with shitty hardware or software, I am quite happy being 6 months behind the bleeding edge.
About a year ago, my college had an alumni breakfast in Silicon Valley. One of my fellow alumni proudly exclaimed that he worked for NVidea writing drivers.
When asked about Vista, he told us how Microsoft was "sooo understanding" about letting them ship drivers before they were complete. I bit my tounge and decided to stay away from Vista.
No, I will not work for your startup
Its clearly cause vista sucks....