GeForce GTX 660 Ti causes PC reboot... but why?

biluinaim

Commendable
Feb 21, 2016
10
0
1,510
I have had this graphics card for a while now and I can't figure out what's wrong with it. I've had it replaced by the manufacturer once (Zotac) but the problems persisted.
I am wondering now whether it could maybe be a fault with the port on my motherboard but as I have no other PCs to test the components on I am wondering whether there is a way to diagnose exactly what the issue is before I start buying new parts.

It seems that the card doesn't like stress, except its tolerance is very low. My PC (specs below) should be able to run serious games with no issue, but as soon as anything a bit intensive comes up (certain cutscenes, or areas load - in one game, switching the lights on in a room would do the trick) the screen fills with one colour, the sound starts looping and in a couple of seconds the PC restarts itself.

When it boots back up, there's no error messages or anything. I only know it's the graphics card because I tested it on another kit months ago when I had a chance and reproduced a crash that gave me a blue screen with a code that indicated a generic graphics card issue (I'm afraid I don't have the error code anymore but I did google it thoroughly at the time and came up with no details).


Full specs (I don't see how to attach a dxdiag and I'm not sure you'd want it all pasted in here):
Motherboard: Asus M5A78L-M/USB3
Processor: AMD FX-6100 Six-Core Processor ~3.3GHz
RAM: 16GB
Graphics Card: NVIDIA GeForce GTX 660 Ti
Running Windows 10 Pro 64 bit

Thanks for any help!
 

biluinaim

Commendable
Feb 21, 2016
10
0
1,510


Sorry, forgot to add that! The PSU is a Corsair CX750 and it's fairly new as I originally upgraded from 550W when I changed from a GTX 550 to a 660.
 

biluinaim

Commendable
Feb 21, 2016
10
0
1,510
I managed to find more information about the crashes/reboots in the Reliability History.

For each crash there are 3 events with the same timestamp:
Source Summary
Windows Windows was not properly shut down
Windows Windows stopped working
Windows Shut down unexpectedly

The details to "Windows stopped working" show this:

The computer has rebooted from a bugcheck. The bugcheck was: 0x00000124 (0x0000000000000000, 0xffffe001a89fe038, 0x0000000000000000, 0x0000000000000000). A dump was saved in: G:\Windows\Minidump\022216-15406-01.dmp. Report Id: 5a3c5f8c-329d-48a5-bb64-18e4714f3fad.

I've found out from the internet that this is a "hardware complaint" (http://www.sevenforums.com/crash-lockup-debug-how/35349-stop-0x124-what-means-what-try.html) but I've no idea how to figure out which part it is.

Thanks in advance for any more help.
 

biluinaim

Commendable
Feb 21, 2016
10
0
1,510


Thanks very much, I wouldn't have suspected the PSU.

I found some software to check the voltage going through my PSU but I'm way out of my depth. Any chance you can make sense of these? These were taken while just running Firefox.

Hardware monitor ITE IT8728
Voltage 0 0.91 Volts [0x4C] (CPU VCORE)
Voltage 1 1.50 Volts [0x7D] (VIN1)
Voltage 2 1.87 Volts [0x9C] (+3.3V)
Voltage 3 3.41 Volts [0xA9] (+5V)
Voltage 4 8.06 Volts [0xA8] (+12V)
Voltage 5 -9.70 Volts [0xCA] (-12V)
Voltage 6 -7.97 Volts [0xA6] (-5V)
Voltage 7 3.63 Volts [0xB4] (+5V VCCH)
Voltage 8 1.62 Volts [0x87] (VBAT)


Hardware monitor NVIDIA NVAPI
Voltage 0 0.99 Volts [0x3DB] (VIN0)
Power 0 15.15 pc (GPU)
Temperature 0 42°C (107°F) [0x2A] (TMPIN0)
Fan 0 1230 RPM [0x4CE] (FANIN0)
Fan PWM 0 30 pc [0x1E] (FANPWMIN0)
Clock Speed 0 324.00 MHz [0x144] (Graphics)
Clock Speed 1 324.00 MHz [0x144] (Memory)
 

biluinaim

Commendable
Feb 21, 2016
10
0
1,510
Hello again,
It seems like it's not the PSU. I found my old 500W one and it gave me the same problem, if worse (the PC just froze, completely non-responding, did not reboot). Since this morning, my SSD also refuses to boot saying there's some problem with Windows (I currently work around it by booting my HDD, which then prompts the Windows 10 disk selection and my SSD works just fine from there).

I've also downloaded FurMark and ran the 1080p test with both PSUs and had no crashes (results in the 1700~).

Really no idea what's going on but it feels like something's messing everything up now, little by little :/