Display driver has stopped responding and has recovered error

[SickBeast]

SB,can u check the event viewer and post the entire message when it happens?Also if using beta pls rollback to the last whql driver.

I don't know how to do that.

Even with the drivers rolled back to the February version I'm still getting the crash.

I will RMA the card but I need to find a hot deal on a new one first.

 

[SickBeast]

Even with Aero turned off the display driver keeps crashing.

 

[iluvdeal]

Try dis: http://support.microsoft.com/kb/2665946

 

[SickBeast]

Try dis: http://support.microsoft.com/kb/2665946

Thank you for that, iluvdeal.

I've tried the registry setting and will report back with the results.

I have a feeling my card is dead, though. It seems to be getting worse.

 

[iluvdeal]

Can you try the card out in another computer to verify the problem is reproducible on a different system? This rules out it having to do with your mobo, ram, software config, psu, etc. You don't want them sending you the card back because they could find nothing wrong with it.

 

[(sic)Klown12]

I had this issue with my old 6970 whenever it dropped to 2D speeds of 300/150. I made a profile in CCC and set the clocks to 400/300 and the issue went away. Since you can mod the BIOS, try raising both the idle clocks and voltage to something slightly higher, or you could just disable power saving all together.

 

[Nintendesert]

lol, I probably would have thrown the pc out the window by now. I am having my own audio issues right now that are pissing me off. plug and play consoles are starting to look pretty good right about now...

Yeah, after messing around with XBOXes and the red ring of death too often and even faulty PS3s I'll stick with PC gaming where at least I can try to fix stuff and not just use a hair dryer.

 

[SickBeast]

I had this issue with my old 6970 whenever it dropped to 2D speeds of 300/150. I made a profile in CCC and set the clocks to 400/300 and the issue went away. Since you can mod the BIOS, try raising both the idle clocks and voltage to something slightly higher, or you could just disable power saving all together.

If I turn off the adaptive power saving the problem still happens, and that's at the full max clockspeeds.

I think my card is dying. I had some strange purple screen corruption happening as well.

 

[Throckmorton]

I have this issue on both my work laptop, a Dell XPS with dual 8700GTs, and my own laptop which is a Sony CW with 230M. I haven't been able to find any fix.

 

[railven]

Not sure if too late, but I went through 2 GTX 460s (EVGA) with the same issue. I never did find out what was the cause.

The cards played games fine (would however TDR during gaming sometimes) but after reformatting, try the handful of drivers available and 2 RMAs EVGA gave us a GTX 460 SC for the difference (<$20) and it stopped.

Until I updated her drivers again (forgot which version) and the TDR issue came back. When I swapped her to the temp GTX 560 Ti the problem persisted, it would always involve her ALT-Tabbing to Firefox, that much was a given.

Someone somewhere said it was a power issue from the cards switching clocks due to various different states. I think it would be related. For example, now with my HD 7970, if I load up Youtube my card locks to 501mhz which creates gaming lag, if I close the Youtube page it clocks back up to 1150mhz.

I think these cards are getting too sophisticated and unfortunately, we'll run into errors along the way. Thankfully, the GTX 680 has yet to TDR on us (knock on wood.)

 

[WT]

Got my first TDR yesterday after having never seen the crash on ANY of the 5 PCs I'm running. Of course this had to happen to my GTX580. I was using the 295 driver, but once it started I installed the 30 driver .. no change. One time I did note artifacts on the screen (banner ad here an AT that had weird purple blotches on it) but I'm really at a loss on what to try next other than pull the card and swap it with something else.

System specs:
2500K @ 4.1 using an H80
Zotac GTX580 @ stock
Antec 750 High Current PSU
Patriot 12800 DDR3 - 8GB
Gigabyte Z68XP-UD3-iSSD

Nothing changed from working fine yesterday to crashes every minute today. So frustrating to go from a perfectly functional PC to one that crashes every minute.

 

[ShintaiDK]

Btw, the TDR was a new features in Windows Vista. I helped both AMD and nVidia users with this as changing drivers etc didnt help.

Sometimes people need a clean Windows install for it to be fixed. This is basicly a Microsoft problem. Microsoft thought they was clever....they wasnt.

http://msdn.microsoft.com/en-us/library/windows/hardware/ff569918(v=vs.85).aspx

A good advice would be to increase the TDR timeout. Specially when you using any sensoring software you increase the chance of getting timeouts. An increase can fix your problems (TdrDelay). For example set it to 8. Setting TdrLevel to 0 disables it.

 

[WT]

Thanks, Shintai, I will try the MS fix. I was on that page yesterday but after 4-5 TDRs, my PC would BSOD with an Nvlddmkm error.

I can also try to wipe and restore the PC from earlier this week using my WHS. Now that I think about it, I did have the TDR or a similar weird issue over 3.5 years ago, and that turned out to be failing RAM on my rig at that time. I'll try and run a memtest later tonight.

 

[Stuka87]

Setting my CPU clock down to 3800 instead of 4000 helped me to get rid of this little a**hole of a problem. Stock is 3500 phenom x4 deneb and I am currently using a Sapphire 6850. I'm assuming my PSU couldn't handle it probably (Thermaltake 600w non certified version). I also set ram to auto instead of 1333 but anytime I try to go 4k CPU with an overclocked GPU it seems to cause this problem.

OP what are the specs on your card? I have bad momentary memory but did you mention whether it was OCed or not?

600W is more than enough for your setup, assuming it does have multiple 12V rails which may be resulting in overloading one of them.

As for your CPU, what all are you setting to boost the speed? If its just the multiplier, there is your problem. You need to look into voltage and CPU-NB in order to OC an AMD.

 

[The_Golden_Man]

My previous Gigabyte GTX 570 (Dead now) also started out with "Display driver has stopped responding and has recovered error". This happened in idle, youtube videos and regular use. Eventually my card died and couldn't even start my system

 

[NFarnzy]

I have had ongoing TDR's since I have bought my gtx 560 TI.. I have tried almost everything I can to remedy it.. Nothing has worked.. sent the original 560 Ti back got a different one.. and still happend.. I have tried several versions of the nvidia drivers.. still happens.I have also tried a couple different MB and still happens. For me it is driver related.. it at least is linked to it. To Clear the TDR... I just shut my PC down completely and it clears it. then it is good again for a while!

 

[birthdaymonkey]

I started getting this error (almost daily) on my Galaxy GTX 560Ti a while ago after a driver update. I rolled back to 275.33 and never experienced the problem again. I've been hesitant to update that machine since, so it's still going strong on 275.33.

 

[samboy]

I had to run my GTX 460 1GB on V275.33 for a long time to avoid TDR's (and tried many of the new drivers).

The V296.10 actually worked for me without any TDR issues. However, given the time I wasted before I'm not going to try newer drivers without any reason......... (once upon a time I used to update the drivers as they came out)

 

[sandorski]

I wonder if some of these problems could be HDD/File Corruption issues. It's been a few years since I have had similar problems with a 4850 and X2 3800 all at Stock Win XP 64. The problem happened for a few months, then eventually subsided, don't recall what I did, but likely updated Drivers during that time. Anyway, what also occurred later on with that system was that one of my HDDs began to develop Bad Sectors and later when I retired the system I discovered half a dozen bulging Caps on the Motherboard.

 

[WT]

Two weeks since my last mysterious 'display driver error' message. I made the change to the registry that Shintai recommended, upping the TDR timeout from 2 to 8.

I'm not sure if a somewhat flaky dat update came out from Kaspersky that may have caused my issues, but I never had it again after that one day. I'm sure I'm probably cursing myself by posting this, but so far it looks like its resolved for me.

 

[Magic Carpet]

Had the same issue with my ex (Geforce 250). Think I rolled back to some drivers and it got fixed.

 

[ShintaiDK]

Two weeks since my last mysterious 'display driver error' message. I made the change to the registry that Shintai recommended, upping the TDR timeout from 2 to 8.

I'm not sure if a somewhat flaky dat update came out from Kaspersky that may have caused my issues, but I never had it again after that one day. I'm sure I'm probably cursing myself by posting this, but so far it looks like its resolved for me.

Happy to hear

For others, remember sensoring tools etc can cause TDRs too.

 

[Ventanni]

This may have been said during the post already, so I apologize for repeating anything, but from my understanding the crash stems from Windows 7 thinking the gpu drivers crashed during up/down clocking. I used to get this a lot with my Geforce 285. I thought it was a heat related issue 'cause it didn't crash as much when I took the side off the case, but when I popped in my 5850 and saw it crash (although not nearly as often), I began to wonder. So yeah, it's not an Nvidia specific issue.

So I'm wondering, is there a way to keep the GPU from downclocking itself? Just run at the same mhz/voltage all the time?

 

[Vesku]

Do you run with more than one monitor?

 

[ShintaiDK]

Its not while up and downclocking as far as I am aware. Its simply the driver that "timeouts" and windows restarts it. The driver can be occupied by another program and cause the error that way for example.

MS made a horrible concept with TDR in assuming the driver would always be ready to "check". Hence you get the TDR error in normally valid situations.

A paste from MSDN:

The following is a brief overview of the TDR process:

1. Timeout detection: The Video Scheduler component of the Windows Vista graphics stack detects that the GPU is taking more than the permitted quantum time to execute the particular task and tries to preempt this particular task. The preempt operation has a "wait" timeout&#8212;the actual "TDR timeout." This step is thus the "timeout detection" phase of the process. The default timeout period in Windows Vista is 2 seconds. If the GPU cannot complete or preempt the current task within the TDR timeout, then the GPU is diagnosed as hung.
2. Preparation for recovery: The operating system informs the WDDM driver that a timeout has been detected and it must reset the GPU. The driver is told to stop accessing memory and should not access hardware after this time. The operating system and the WDDM driver collect hardware and other state information that could be useful for post-mortem diagnosis.
3. Desktop recovery: The operating system resets the appropriate state of the graphics stack. The Video Memory Manager component of the graphics stack purges all allocations from video memory. The WDDM driver resets the GPU hardware state. The graphics stack takes the final actions and restores the desktop to the responsive state. As mentioned earlier, some older DirectX applications may now render just black, and the user may be required to restart these applications. Well-written DirectX 9Ex and DirectX 10 applications that handle "Device Remove" continue to work correctly. The application must release and then recreate its Microsoft Direct3D device and all of its objects. DirectX application programmers can find more information in the Windows SDK.