[SOLVED - DEAD CARD]Help - Driver stopped responding...

[railven]

Pretty sure I made a thread, but can't seem to find it now.

System in Sig.

System worked fined Friday/Sunday for my gaming sessions. Today I've had the driver stopped responding a bunch of times, in games, at the desktop, just tried to launch Bioshock got as far as options before driver stopped and game crashed.

I've made no changes to the system except install Steam and download a few games. EDIT: Just checked NV, I was using the WHQL drivers from Oct 21, downloading Oct 28 ones now. Using latest WHQL drivers. Not sure what happened between Sunday (my last gaming session) and today.

Event Log:

Any trouble shooting tips?

 

[railven]

After the first batch of errors, I returned card to stock settings. It ended up working for a bit (20minutes?) before the errors started up again.

PC has hard lock twice since. Gonna do a driver clear and give the newest beta's a whirl.

Maybe it's karma haha.

EDIT: Actually, you do see the 20min gap in which it stopped crashing. Did a little surfing, then went to play Bioshock and it stopped responding and crashed Bio.

 

[railven]

Okay...just to make this weirder...

I did the uninstall, clicked reboot when prompted, Windows reboot, went to remove PhysX to reinstall with new driver and found this:

So it auto installed 311.06 which I have no idea what date they are from, but checking NV driver site, they predate March 2013 (since those were 320).

No prompt, no cancel, just installed that set. This normal for NV or do I got a gremlin some where?

 

[railven]

And it's back. Did the uninstall process again, rebooted, upon Windows boot, 311.06 was installed.

Guess I'll disable my NIC to see if it's a windows auto update garbage nonsense. Well this is frustrating.

 

[bystander36]

Based on the problems, and past experience, the first thing that popped into mind is your PSU is failing.

 

[railven]

Based on the problems, and past experience, the first thing that popped into mind is your PSU is failing.

Possible, and if it is - I'm a sad kitten

While I've been wanting to buy a new PSU, it's usually not after a failure and mostly an upgrade. It is nearing 4 years...

On my driver install issue, disconnected internet, reboot, uninstalled, reboot, no 311.06 reinstalled so not sure if it's NV or MSFT (most likely MSFT) force installing drivers.

System is back up with newest driver (331.65) now to test a game. Wish me luck.

 

[Tweak155]

Windows tries to detect and install drivers for your hardware. My guess is that those are the drivers it finds for your card.

 

[railven]

Holy crap!

See if I can post this before it goes ape shiz again.

Ran Bio: Inf benchmark, ran fine only one spot of weird artifacting, results:

Scene Duration (seconds), Average FPS, Min FPS, Max FPS, Scene Name
32.33, 38.49, 19.40, 47.38, Welcome Center
7.24, 37.57, 13.41, 43.49, Scene Change: Disregard Performance In This Section
21.68, 35.28, 14.31, 48.50, Town Center
8.21, 34.11, 15.41, 45.34, Raffle
9.18, 49.65, 28.63, 55.67, Monument Island
3.06, 54.93, 52.03, 56.46, Benchmark Finished: Disregard Performance In This Section
81.70, 39.01, 13.41, 56.46, Overall

Got to desktop and it started crashing, Event Viewer:

Any way I can test my PSU, or do I have to yank another one out (think I'm out of spares...)

I'm using the IGP on my 2600K to power a second monitor which shows no issues until the driver resets then it just refreshes (since the system locks up during the reset process.)

I wonder if my 660Ti is damaged?

 

[railven]

Tried another Bioshock bench, failed 10 seconds into it, Driver stopped, game crashed...

Time to open up the case and see if something is loose.

 

[bystander36]

Have you checked the GPU and CPU temps during these crashes? You mentioned removing the OC made it last longer originally. Heat could be a factor.

 

[railven]

Have you checked the GPU and CPU temps during these crashes? You mentioned removing the OC made it last longer originally. Heat could be a factor.

Sigh...had to fix GF's PC issue first (unrelated.)

Yeah, temps were fine, I monitor temps like a hawk.

GTX 660 Ti was about 68C (stock fan) CPU was about 35% average load running 42C.

PC is open, all connections double checked and secure. No random dust bunnies, seems normal.

Gonna try Bio bench again, well report findings.

 

[railven]

Well damn, before the driver was not responding, this time I got a windows sad face (BSOD) and reboot.

I tried to write down the temps, but ended up taking a blurry pic on my phone right as it crashed, here is the results, best I can read:

CPU:
AVG Load: 28
temp: 40c?
Core 0 was at >70% load, rest were below 40%

GPU
load 98%
temp: 58C
Clocks: 1034mhz
RAM: 3004mhz
Fan: 30%
power 90%
volts 1.175v

That looks about right.

At no point did temps seem like an issue...I wonder if it's my hDD failing?

Well, the first time this started to happen I was running a game off the SSD (FFXIV), and now I'm running Bio off my standard game drive (7.2K RPM 500KB Seagate model 7200.12)

So possibly not the HDD. Dammit, I finally decide to get through my back log of games and now I can't get it stable haha.

Gonna resit card.

 

[bystander36]

I was having an issue, which I thought was my SSD at first, as it would stop responding. I then started getting crashes and blue screens in SLI, and found that if I turned SLI, it worked. As it turned out, it was a failing PSU, which caused the other parts to fail. I'd personally check there first.

 

[railven]

I was having an issue, which I thought was my SSD at first, as it would stop responding. I then started getting crashes and blue screens in SLI, and found that if I turned SLI, it worked. As it turned out, it was a failing PSU, which caused the other parts to fail. I'd personally check there first.

Welps, tried changing the PCIE connectors...no dice.

I wonder if I can try to run Bio on the IGP and see if it finishes (as terrible as the score may be.)

Are there any PSU tests I can run? I'm fresh out of spares at the moment :/

 

[blackened23]

Are you overclocking your GPU? I get TDRs with nvidia cards when I overclock too high. Or if the temperature is too high. AMD cards tend to blue screen when OC'ed too high, nvidia cards tend to TDR instead. When I push clocks too high on my 780, my fans will revert to auto fan (I usually ran manual fan) and then pause....pause....TDR..... is that what's happening to you? Are you OC'ing the 660? If so , revert the OC. Or raise the fan speed. Some overclocks require higher manual fan settings.

 

[railven]

Are you overclocking your GPU? I get TDRs with nvidia cards when I overclock too high. Or if the temperature is too high. AMD cards tend to blue screen when OC'ed too high, nvidia cards tend to TDR instead. When I push clocks too high on my 780, my fans will revert to auto fan (I usually ran manual fan) and then pause....pause....TDR..... is that what's happening to you? Are you OC'ing the 660? If so , revert the OC. Or raise the fan speed. Some overclocks require higher manual fan settings.

Only CPU is OC'ed, which I'm about to reset to stock just to test. Soo will report back with CPU stock too.

 

[railven]

No dice, everything stock, crashed seconds into Bio bench.

Got GF to approve me testing her GTX 680, gonna stick my 660 into her rig.

Let's find out what's dying...might be both haha. Will report back.

 

[Majcric]

do you have another GPU you can try in your system? I've seen GPu's be finicky like that before just enough to be annoying. Half-dead, Half Alive. A pain to troubleshoot.

 

[railven]

About to test GTX 680 in my rig. Going to assume GTX 660 Ti is dead or failing.

Put 660 in her system after taking out her 680, turned on one of her games about 30 seconds in, it started artifacting badly, driver crashed, then fine, then more artifacting and hard system lock up.

About to run Bio benchmark on my system with her 680, I expect no issues. Will report when done.

 

[railven]

Bio bench finished, no issues:

Per Scene Stats:
Scene Duration (seconds), Average FPS, Min FPS, Max FPS, Scene Name
32.69, 47.58, 23.66, 83.61, Welcome Center
6.64, 48.19, 19.89, 55.72, Scene Change: Disregard Performance In This Section
22.02, 45.37, 11.41, 54.46, Town Center
8.15, 43.68, 32.86, 55.08, Raffle
9.20, 61.05, 9.66, 73.32, Monument Island
3.06, 68.98, 41.78, 72.40, Benchmark Finished: Disregard Performance In This Section
81.75, 48.96, 9.66, 83.61, Overall

Note, CPU is stock now so may have lower min that first benchmark.

Gonna run another bench to see, wish I had downloaded Metro 2033 this morning.

 

[railven]

Another run passed of Bio with 680 in my system:

Per Scene Stats:
Scene Duration (seconds), Average FPS, Min FPS, Max FPS, Scene Name
32.75, 47.33, 17.96, 63.91, Welcome Center
6.68, 45.85, 10.25, 59.24, Scene Change: Disregard Performance In This Section
22.29, 43.44, 3.36, 56.23, Town Center
8.19, 41.79, 8.09, 55.42, Raffle
9.17, 61.31, 9.32, 72.94, Monument Island
3.04, 68.81, 63.87, 70.68, Benchmark Finished: Disregard Performance In This Section
82.12, 47.99, 3.36, 72.94, Overall

I couldn't even get her system to boot with the GTX 660 after the last hard lock. Would see the Win logo, artifacting, and just reboot.

Welps so much of my pleasant experience with NV. In under 3 days it went from ecstasy to disdain.

I had bought the 660 Ti super cheap from an online friend, it's an EVGA, wonder if I can RMA it...should see if he registered it or has the receipt.

Worst case, I run of my HD3000 for a few days while I get a replacement card.

 

[railven]

Windows tries to detect and install drivers for your hardware. My guess is that those are the drivers it finds for your card.

Sorry missed this, I know Windows does try to detect but it's always prompted me when changing drivers on my AMD cards. It's never fully installed CCC and optional software on it's own which I found really weird. It would at most just install the display driver.

In the image above, you can see it installed everything, 3d Vision driver, even HDMI audio drivers. The full shabang without ever prompting me to confirm.