The crashes happen at random times and at random places, has happened at the desktop even. But most often they occur when gaming, especially graphically intense games. But what's weird about them is that, before crashing, the GPU behaves totally normal, normal temps, fan is doing it's job as always, no stuttering, no fps drop, no artifacts nothing out of the usual. And as mentioned before, they are random. Can occur 5 min after turning on the PC, or it may not even crash at all, even after hours of gaming. Once, using Heaven Benchmark, I benchmarked/stress tested the GPU for more than 4 hours straight and absolutely nothing happened, and then the next day it would crash after 10 minutes of CS:GO at low settings.
What I've tried so far: obviously always using the latest drivers, doing that recovery time thingy on the register, CLEAN drivers reinstall, reinstalling windows after formatting everything, switched PCI slots, underclocking gpu, cleaning the graphics card, heck I even updated the BIOS.
Now I did try it on another PC, which also used a GTX 690 (how convenient, not having to install driver after switching heh), no crashes there, but I wouldn't count that as a test, because it was only for 30 minutes due to limited time. Might try for a longer period sometime.
DETAILS OF THE CRASHES: the crashes can behave in 3 different ways, but lately, the third one seems to stand out:
Type 1. Screen freezes (for about 2-8 secs), then screens start flickering (for about 1 second) (and while it flickers, the gpu fan revs up to max speed [oddly, sometimes it doesn't]), and then the driver recovers (nvidia display driver has stopped responding and has recovered) and fan slows down to normal speed.
or type 2. Screen freezes, works again, freezes again, works again and so on, until I force restart. (On very rare occasions, instead of freezing again, it would just resume working as usual, but that only happened once or twice). I posted a video about this one (don't judge my music taste pls). Video: https://youtu.be/n3gMXtk5XJk At 1:08 you can hear the sound glitching too.
or type 3. Total blackout, screens just go black with "No signal detected" on both monitors, PC is still ON, and the graphics card's fan rev up to maximum speed. Sounds like a GPU fail-safe perhaps? But why, if the temperatures are normal? Also, PC seems to not respond either, tapping the power button to shut down safely from Windows doesn't work, also the drive writing light on the front of the PC stops flickering, no activity. Forgot to mention: this happens INSTANTANEOUSLY, in a blink of an eye it goes from totally fine to this.
Sound either stops or glitches on all of them.
I can't remember when this started, because I didn't give it much importance because I only played CS:GO, where it hardly ever crashes (BUT STILL DOES), but I would say it's close to a year now.
These details just YELL that it's the graphics card's fault, but it couldn't be the GPU chip itself because the symptoms are so weird, right? As the temps only show the main chip, maybe its the memory? or perhaps the power delivery inside the graphics card? Or maybe it's the PSU overvolting the graphics card at random intervals? Pshh, I'm an amateur anyway, I shouldn't even be making assumptions. Is this fixable? My beloved GTX 690 still stands up to today's games very well, so don't want to get rid of that.
Anyway, thanks for reading, hopefully someone can help.
PC SPECS: (was bought and assembled (by professionals) on summer of 2012
OS: Windows 7 Home Premium
CPU: Intel Core i7-3960X (not overclocked)
GPU: 1x Reference Nvidia GeFroce GTX 690 (not overclocked)
RAM: 4x 4GB Kingston DDR3 PC3-10700 (667 MHz)
Motherboard: Intel DX79SI
Two screens*: 1x 1080p @144Hz (BenQ XL2411Z), and 1x 1920x1200 @60Hz (Hanns G HZ281)
Drives: 1x 120GB SSD (Windows [and drivers I suppose?]) and 1x 1TB HDD (everything else)
Power Supply: Chieftec 850W 80+ (only 80+, neither silver nor gold nor anything like that)
* My main monitor for gaming is the BenQ, I don't have the screens set up for surround or anything like that, and I have set the nvidia control panel for single screen performance.
NOTE!!!: Decided to jump to the event viewer after a type 2 crash and saw something interesting. There's this one warning the kept repeating about 20-50 times EVERY SECOND (pictures below).
Event viewer: https://gyazo.com/05451fae7285d886a81bdadf6bb7a478
Details of the warning: https://gyazo.com/8386de163b17220d7798770ebc1d9f34
https://gyazo.com/4501e7882e44ce561515c1bcb0526f7f
I believe this error appears on all crashes: https://gyazo.com/ff48107c7d4980a4bb141652a00b5bef
https://gyazo.com/bcf744edd01481c17b8397db648153ec
What I've tried so far: obviously always using the latest drivers, doing that recovery time thingy on the register, CLEAN drivers reinstall, reinstalling windows after formatting everything, switched PCI slots, underclocking gpu, cleaning the graphics card, heck I even updated the BIOS.
Now I did try it on another PC, which also used a GTX 690 (how convenient, not having to install driver after switching heh), no crashes there, but I wouldn't count that as a test, because it was only for 30 minutes due to limited time. Might try for a longer period sometime.
DETAILS OF THE CRASHES: the crashes can behave in 3 different ways, but lately, the third one seems to stand out:
Type 1. Screen freezes (for about 2-8 secs), then screens start flickering (for about 1 second) (and while it flickers, the gpu fan revs up to max speed [oddly, sometimes it doesn't]), and then the driver recovers (nvidia display driver has stopped responding and has recovered) and fan slows down to normal speed.
or type 2. Screen freezes, works again, freezes again, works again and so on, until I force restart. (On very rare occasions, instead of freezing again, it would just resume working as usual, but that only happened once or twice). I posted a video about this one (don't judge my music taste pls). Video: https://youtu.be/n3gMXtk5XJk At 1:08 you can hear the sound glitching too.
or type 3. Total blackout, screens just go black with "No signal detected" on both monitors, PC is still ON, and the graphics card's fan rev up to maximum speed. Sounds like a GPU fail-safe perhaps? But why, if the temperatures are normal? Also, PC seems to not respond either, tapping the power button to shut down safely from Windows doesn't work, also the drive writing light on the front of the PC stops flickering, no activity. Forgot to mention: this happens INSTANTANEOUSLY, in a blink of an eye it goes from totally fine to this.
Sound either stops or glitches on all of them.
I can't remember when this started, because I didn't give it much importance because I only played CS:GO, where it hardly ever crashes (BUT STILL DOES), but I would say it's close to a year now.
These details just YELL that it's the graphics card's fault, but it couldn't be the GPU chip itself because the symptoms are so weird, right? As the temps only show the main chip, maybe its the memory? or perhaps the power delivery inside the graphics card? Or maybe it's the PSU overvolting the graphics card at random intervals? Pshh, I'm an amateur anyway, I shouldn't even be making assumptions. Is this fixable? My beloved GTX 690 still stands up to today's games very well, so don't want to get rid of that.
Anyway, thanks for reading, hopefully someone can help.
PC SPECS: (was bought and assembled (by professionals) on summer of 2012
OS: Windows 7 Home Premium
CPU: Intel Core i7-3960X (not overclocked)
GPU: 1x Reference Nvidia GeFroce GTX 690 (not overclocked)
RAM: 4x 4GB Kingston DDR3 PC3-10700 (667 MHz)
Motherboard: Intel DX79SI
Two screens*: 1x 1080p @144Hz (BenQ XL2411Z), and 1x 1920x1200 @60Hz (Hanns G HZ281)
Drives: 1x 120GB SSD (Windows [and drivers I suppose?]) and 1x 1TB HDD (everything else)
Power Supply: Chieftec 850W 80+ (only 80+, neither silver nor gold nor anything like that)
* My main monitor for gaming is the BenQ, I don't have the screens set up for surround or anything like that, and I have set the nvidia control panel for single screen performance.
NOTE!!!: Decided to jump to the event viewer after a type 2 crash and saw something interesting. There's this one warning the kept repeating about 20-50 times EVERY SECOND (pictures below).
Event viewer: https://gyazo.com/05451fae7285d886a81bdadf6bb7a478
Details of the warning: https://gyazo.com/8386de163b17220d7798770ebc1d9f34
https://gyazo.com/4501e7882e44ce561515c1bcb0526f7f
I believe this error appears on all crashes: https://gyazo.com/ff48107c7d4980a4bb141652a00b5bef
https://gyazo.com/bcf744edd01481c17b8397db648153ec
Facebook
Twitter
Instagram