GTX 570 The problem of sudenly crashing down with Black Screen, BSOD and "Driver Stop

[ninoaoe]

I'm sure that many of you have seen threads like this one many times but really i haven't seen any of them completely solved.

My GTX570 suddenly after installing Thief,started doing all the these things..Specifically:

#Crashing giving black screen while sound still plays but no button except hard restart works
#Blue Screen Of Death saying "VIDEO_SCHEDULER_INTERNAL_ERROR"
#"Driver stopped working" message
#"Driver stopped working" message multiple continuous times one after another giving some artifacts-like
WHAT I HAVE TRIED
+Clean install drivers (20 times maybe)/clean old drivers
+Bought a new PSU (CX750) and i rarely see the hard crash where i have to restart the PC
+Overclock/underclock(not too much)/ increase voltage from 1000mV --> 1020/1050/1090
+Various changes at Nvidia control panel/manage 3D settings
+Change PCI slot
+Change DVI slot
WHAT I HAVE NOTICED
- I have an extra monitor(TV-HDMI) connected and i think that when i remove it crashes happen less often
-When i'm on Chrome,crashes happen like every 5 minutes
-The card never did these things before me installing some games and overclocking(probably overclocking damaged the card?
-Underclocking from 732 to 600 and removing the 2nd display seems to work just fine for a long time but again crash.Underclocking to 550 and still no crash..

Help will be appreciated! I can't sleep if i can't fix it!

 

[tweakboy]

Try another game. It can be the game. Also set your nvidia settings

See what happens. gl

 

[tweakboy]

Also could be a nvidia panel problem.. make sure nvidia panel settings are set in default. Stop your OC if you are OCing,, and lets take it from there.

 

[Sunburn74]

Maybe its a CPU problem or some other component. What kind of stress testing have you done?

 

[aShufflinZombie]

nVidia killed the 500-series with bad drivers starting in about September '13. My 560ti has the same problems even when just using chrome.

 

[tweakboy]

LOL, my firefox and thunderbird dont even launch,,,,it just goes into memory. I can use sunbird,, but they need to fix this bs.. gl

 

[ninoaoe]

I don't think that it's another component cause i've seen and other people having the same problem..But i never had a problem with the drivers before..Why now?

 

[tweakboy]

Ya same here, smooth for me!

 

[stahlhart]

Were you running 334.89 prior to having installed Thief, or did you install this version because you installed Thief?

 

[amenx]

nVidia killed the 500-series with bad drivers starting in about September '13. My 560ti has the same problems even when just using chrome.

Still have a 570 in another machine. Running fine with latest drivers. No problems with Chrome either.

 

[Mk pt]

With my old GTX570 i was getting a lot of drivers issues on Battlefield 3/4 and Max Payne 3.
But didn't have problems with browser's or other games [but i didn't try with some other AA games, like Metro: LL or any CoD's].

Now i have a r9 290 [Tri-X] and no issues with any game.
Even in Bf4 there's no issues. But i didn't try Mantle yet.



Nvidia didn't kill GTX 5xx in latest drivers - i tried drivers from back the days the GTX570 was a high end gpu and was getting same drivers issues in those games......

But 'everybody' - who did a fast search about it - know's that Nvidia reduced GTX5xx performance via drivers after GTX6xx launch, especially in CUDA performance.
Still, a GTX580 smoke's a GTX770 in Cuda performance because Nvidia want us to pay hundred's $$ for a weak gpu with acceptable CUDA performance [low/mid-range quadro's].



I was very happy with my old Nvidia.
But i couldn't support nvidia pratices, at least for now. That's why i went to AMD R9 290, wich is also the gpu with the best price/performance in the mid-high segment.

 

[ninoaoe]

Were you running 334.89 prior to having installed Thief, or did you install this version because you installed Thief?

No i had the latest drivers before i think..I mean,i always install new drivers when they come out.

Right now i formated my SSD adnd installed windows again..So far no problem at all! Even with stock speeds.

Could it be a software problem that was making all these crashes?I can't understand..

 

[stahlhart]

No i had the latest drivers before i think..I mean,i always install new drivers when they come out.

Right now i formated my SSD adnd installed windows again..So far no problem at all! Even with stock speeds.

Could it be a software problem that was making all these crashes?I can't understand..

And probably something you'll now never be able to definitively trace back to the root cause, unfortunately. Maybe some sort of DirectX issue? A corrupt file? Who knows.

But the good news is that it appears to be working correctly now...

 

[ninoaoe]

And probably something you'll now never be able to definitively trace back to the root cause, unfortunately. Maybe some sort of DirectX issue? A corrupt file? Who knows.

But the good news is that it appears to be working correctly now...

DirectX problems? Really never heard of it...
Whatever,thanks a lot guys for the help! I'll inform if i find out what was it of if it starts again

 

[BonzaiDuck]

DirectX problems? Really never heard of it...
Whatever,thanks a lot guys for the help! I'll inform if i find out what was it of if it starts again

I've been troubleshooting a very (very!) intermittent problem for too darn long. I've replaced all the drivers. I've rechecked and reset my BIOS settings, to eliminate any overclocks as culprit. I replace RAM needlessly, although I probably wanted to go from four modules to two for 16GB anyway. I've tried different monitor configurations with the HD 3000 and my GTX 570.

I decided to replace the old GTX 570 with a new GTX 7XX card -- but not for the sole purpose of troubleshooting. But replacing the 570 seems to coincide with a disappearance of any troubles. I discovered that the fan on the GTX 570 was clotted with kruft, and I had a better idea why the idle temperatures on the card were pushing 50C. The new card hasn't even broken 40C under medium load conditions.

Some of my symptoms seem similar to your symptoms. But I think my symptoms have disappeared.

 

[RampantAndroid]

Have you tested your system RAM? Run Memtest86+ for at least 1 pass for the sake of ruling that out. I've seen bad RAM make the "Video driver stopped responding" appear repeatedly.

 

[tweakboy]

This story needs gang green.

 

[el etro]

*In Control Panel\All Control Panel Items\Power Options\Edit Plan Settings, chose edit advanced settings, and in PCi Express, set Link State Power Management to off(solved it for me, GTX 460);
*Deactivate all hardware acceleration in browsers;
*Use 314.22 Driver to test, Go 327.23 driver if everything gone right, then try R331 driver.

 

[ninoaoe]

Have you tested your system RAM? Run Memtest86+ for at least 1 pass for the sake of ruling that out. I've seen bad RAM make the "Video driver stopped responding" appear repeatedly.

*In Control Panel\All Control Panel Items\Power Options\Edit Plan Settings, chose edit advanced settings, and in PCi Express, set Link State Power Management to off(solved it for me, GTX 460);
*Deactivate all hardware acceleration in browsers;
*Use 314.22 Driver to test, Go 327.23 driver if everything gone right, then try R331 driver.

I tried memtest for 3 passes and they were ok!
I did turn off the link State power managment as well but no luck..
Now everything works just fine with the new fresh install of windows 8.1 64bit and 331.65 drivers !At least for now..

 

[ninoaoe]

Ok i just overclocked my CPU and enabled the XMP (extreme memory profile) too see if would happen anything and AGAIN DRIVER STOPPED RESPONDING and crashes of the screen...
So how is this connected? Is it the overclocking? The XMP?The PSU i'm sure it's enough (it's a CX750 corsair)

 

[stahlhart]

Your overclock isn't stable. Gradually roll your settings back and re-test until the crashing stops.

 

[ninoaoe]

Your overclock isn't stable. Gradually roll your settings back and re-test until the crashing stops.

I don't get Blue screens and the temps are just fine..I don't even understand how is my cpu OC connected to my gpu crashes...The only thing i can think of,is becaus of the PSU..Which is brand new and quite good.

 

[BonzaiDuck]

Ok i just overclocked my CPU and enabled the XMP (extreme memory profile) too see if would happen anything and AGAIN DRIVER STOPPED RESPONDING and crashes of the screen...
So how is this connected? Is it the overclocking? The XMP?The PSU i'm sure it's enough (it's a CX750 corsair)

Which CPU?

I say this because I'm beginning to suspect a cause of my own problem. All this time I had OC'd my SB-K chip, I had left the HD 3000 enabled (if only with "Render Standby->Enabled.") I had otherwise set "Enable Lucid Virtu / multi-monitor" enabled -- so that I had that option in the OS if I chose to use it.

The iGPU graphics scales with the CPU overclock. There are very few tweaks available for the iGPU in my Z68 BIOS, except for "iGPU LLC" and "iGPU Offset" sign, "Maximum iGPU frequency" and iGPU offset voltage. I don't think regular stress-tests like LinX or Prime95 would reveal any susceptibility of the iGPU to cause a crash, but it might during any sort of graphics-intensive activity.

If one wanted to insure iGPU stability with CPU OC'ing, it might be necessary to use the "maximum frequency" setting for the iGPU to limit how it scales with the CPU overclock frequency.

I'm just speculating about this, but I finally did everything possible in BIOS to disable the iGPU on my processor. I can't validate any of some few causes for my trouble, because I took all these precautions at once and replaced my GTX 570 with the GTX 780. So far, the once-in-ten-day instability seems to have disappeared.

 

[ninoaoe]

Which CPU?

I say this because I'm beginning to suspect a cause of my own problem. All this time I had OC'd my SB-K chip, I had left the HD 3000 enabled (if only with "Render Standby->Enabled.") I had otherwise set "Enable Lucid Virtu / multi-monitor" enabled -- so that I had that option in the OS if I chose to use it.

The iGPU graphics scales with the CPU overclock. There are very few tweaks available for the iGPU in my Z68 BIOS, except for "iGPU LLC" and "iGPU Offset" sign, "Maximum iGPU frequency" and iGPU offset voltage. I don't think regular stress-tests like LinX or Prime95 would reveal any susceptibility of the iGPU to cause a crash, but it might during any sort of graphics-intensive activity.

If one wanted to insure iGPU stability with CPU OC'ing, it might be necessary to use the "maximum frequency" setting for the iGPU to limit how it scales with the CPU overclock frequency.

I'm just speculating about this, but I finally did everything possible in BIOS to disable the iGPU on my processor. I can't validate any of some few causes for my trouble, because I took all these precautions at once and replaced my GTX 570 with the GTX 780. So far, the once-in-ten-day instability seems to have disappeared.

I've got a pretty old cpu
i7 860 @ 2.8Ghz. and Corsair Vengeance 8GB Ram

 

[BonzaiDuck]

I've got a pretty old cpu
i7 860 @ 2.8Ghz. and Corsair Vengeance 8GB Ram

Then no integrated graphics. So . . . scratch my speculation about the iGPU complications.

The problem I had with troubleshooting an instability that occurred every week and half: It would take a month or longer to check off possible causes to identify which was the culprit. You begin to worry about the effect of even occasional BSODs on your disks, OS install and data. So it's easy to tweak just about everything you can think of until the problem goes away -- no more the wiser for which of those things caused it.