PC Crashes while playing games

[Knightmare10]

Hello,
So a couple months ago I had a buddy install a PNY GTX770 4gb into my PC. My current gfx card was a GTX260. With the 770 installed, whenever I play games my PC sometimes shuts off. Sometimes it won't happen for a week or so, but eventually its going to shut down (no blue screen).

I play a game called Path of Exile a lot, and whenever I play this really graphic intense build, I can't get through a single zone without my PC shutting down. Yesterday I uninstalled all my nvidia drivers and reinstalled the latest one. I was able to play PoE without my shutting down for many hours, but towards the end of the night, it crashed.

If I go back to using my GTX260, I can play games without my PC crashing. I'm wondering if my power supply is not giving enough power or it is overheating after long playing hours, even though the recommended for GTX770 is 600W. Any idea of what can be causing this?

Im not very computer savvy, but here are some of my specs:

Windows 7 Ultimate 64-bit (6.1, Build 7601) Service Pack 1
AMD Phenom(tm) II X4 965 Processor (4 CPUs), ~3.4GHz
GTX260 and GTX 770 (770 is the one crashing)
8.00GB memory
DirectX 11
Thermaltake Toughpower 700W

 

[ruecklm14]

Unless your psu is dying it should be enough. Make sure your GPU isnt overheating. Run a GPU benchmark for a few hours and check you GPU's temp. Also, when doing that, touch your psu. If its to hot to touch than thats bad.

 

[Knightmare10]

Unless your psu is dying it should be enough. Make sure your GPU isnt overheating. Run a GPU benchmark for a few hours and check you GPU's temp. Also, when doing that, touch your psu. If its to hot to touch than thats bad.

Any recommendations for an application to use for the GPU benchmark?

Also, what do you mean by "dying"? Like getting too old? I bought the PC used with the GTX260, and I'd say it's about 6 years old?

 

[ruecklm14]

Heaven Benchmark on ultra is pretty GPU intensive. Also use HWmonitor or GPU-Z to see the temps. (HWmonitor has temps for your cpu too)

 

[Knightmare10]

Heaven Benchmark on ultra is pretty GPU intensive. Also use HWmonitor or GPU-Z to see the temps. (HWmonitor has temps for your cpu too)

Ok, thanks. I actually have argus monitor to monitor my temps, but if you think that's better than I'll try that out..

 

[ruecklm14]

Unless your psu is dying it should be enough. Make sure your GPU isnt overheating. Run a GPU benchmark for a few hours and check you GPU's temp. Also, when doing that, touch your psu. If its to hot to touch than thats bad.

Any recommendations for an application to use for the GPU benchmark?

Also, what do you mean by "dying"? Like getting too old? I bought the PC used with the GTX260, and I'd say it's about 6 years old?

Is your psu 6 years old?

 

[Knightmare10]

Unless your psu is dying it should be enough. Make sure your GPU isnt overheating. Run a GPU benchmark for a few hours and check you GPU's temp. Also, when doing that, touch your psu. If its to hot to touch than thats bad.

Any recommendations for an application to use for the GPU benchmark?

Also, what do you mean by "dying"? Like getting too old? I bought the PC used with the GTX260, and I'd say it's about 6 years old?

Is your psu 6 years old?

Yeah about that. I bought the PC used and put the GTX770 in myself. Ever since I installed the 770 the PC has been crashing during games. It can go for a week or so without crashing, but eventually it does.

When I monitor the temps with my 260, the GPU temp hovers at around 60°C during gameplay and 45°C idle. Whenever I try the 770, the temp is constantly fluctuating anywhere between 50-70°C during gameplay, and 30-35°C idle. Is it normal for the GPU to fluctuate so much?

I really feel like the 770 is drawing so much more power and that something is overheating somewhere.. But like I said, Im not very PC savvy and still learning.

 

[ruecklm14]

Unless your psu is dying it should be enough. Make sure your GPU isnt overheating. Run a GPU benchmark for a few hours and check you GPU's temp. Also, when doing that, touch your psu. If its to hot to touch than thats bad.

Any recommendations for an application to use for the GPU benchmark?

Also, what do you mean by "dying"? Like getting too old? I bought the PC used with the GTX260, and I'd say it's about 6 years old?

Is your psu 6 years old?

Yeah about that. I bought the PC used and put the GTX770 in myself. Ever since I installed the 770 the PC has been crashing during games. It can go for a week or so without crashing, but eventually it does.

When I monitor the temps with my 260, the GPU temp hovers at around 60°C during gameplay and 45°C idle. Whenever I try the 770, the temp is constantly fluctuating anywhere between 50-70°C during gameplay, and 30-35°C idle. Is it normal for the GPU to fluctuate so much?

I really feel like the 770 is drawing so much more power and that something is overheating somewhere.. But like I said, Im not very PC savvy and still learning.

The temps are good and its normal for fluctuation. PSU degrade over time so if your PSU is really old then it could be at the end of its life. Also, your PSU could be overheating. clean out all the dust from your case and make sure that the case fans are not being blocked by anything. Also, it would be good to get your mobo and cpu temps while gaming. HWmonitor can do that. I think its most likely some part is overheating because it's not random and is after long hours of gaming. To get cpu temps use heavyload and put it to test your cpu. run it for a few hours then see the avg temps and max temps. if your system crashes try to be constantly looking at temps so you can see what it was at when it crashed. When I say dying I mean getting old, not working properly. Also, does your GPU fan seem to be running fine (i.e any weird noises from your GPU or any part. When my PSU started to die it was buzzing really loudly .)

 

[aaronfield]

Another thing to try. Start a game or intensive task and check your voltages in HWMonitor if you still have that installed. I believe it shows voltages at the top, and check to make sure your +12v is near 12v. Small fluctuations such as .5 MAX is safe. If you cant conclude enmough from that id buy a new psu for cheap. Ultra is a very cheap manufacturer that makes okay psus. I bought one for my girlfriends family gaming rig and its been chugging for nearly a year.

 

[Knightmare10]

Unless your psu is dying it should be enough. Make sure your GPU isnt overheating. Run a GPU benchmark for a few hours and check you GPU's temp. Also, when doing that, touch your psu. If its to hot to touch than thats bad.

Any recommendations for an application to use for the GPU benchmark?

Also, what do you mean by "dying"? Like getting too old? I bought the PC used with the GTX260, and I'd say it's about 6 years old?

Is your psu 6 years old?

Yeah about that. I bought the PC used and put the GTX770 in myself. Ever since I installed the 770 the PC has been crashing during games. It can go for a week or so without crashing, but eventually it does.

When I monitor the temps with my 260, the GPU temp hovers at around 60°C during gameplay and 45°C idle. Whenever I try the 770, the temp is constantly fluctuating anywhere between 50-70°C during gameplay, and 30-35°C idle. Is it normal for the GPU to fluctuate so much?

I really feel like the 770 is drawing so much more power and that something is overheating somewhere.. But like I said, Im not very PC savvy and still learning.

The temps are good and its normal for fluctuation. PSU degrade over time so if your PSU is really old then it could be at the end of its life. Also, your PSU could be overheating. clean out all the dust from your case and make sure that the case fans are not being blocked by anything. Also, it would be good to get your mobo and cpu temps while gaming. HWmonitor can do that. I think its most likely some part is overheating because it's not random and is after long hours of gaming. To get cpu temps use heavyload and put it to test your cpu. run it for a few hours then see the avg temps and max temps. if your system crashes try to be constantly looking at temps so you can see what it was at when it crashed. When I say dying I mean getting old, not working properly. Also, does your GPU fan seem to be running fine (i.e any weird noises from your GPU or any part. When my PSU started to die it was buzzing really loudly .)

So I actually have a liquid coolant for the CPU installed because the guy I bought the PC off of thought that might be the cause. Me not knowing much just went along with it because I wanted my PC to run correctly.. So the CPU temp never goes above 40°C while gaming. I'm constantly monitoring the temps, it's become painful because Im always worrying that it's too high and its going to crash..

Ever since we put in the liquid coolant, my PC is very loud. I kind of feel like the liquid coolant might be a cause of the PC crashing since that took the place of the rear exhaust fan and now air is pushing in to cool the radiator.. The only exhaust fan I have is one on top of the case and it's not very big.

What I wanted to do was remove the liquid coolant and put a regular CPU heatsink+ fan in its place, and then install a new exhaust fan, as well as, another exhaust fan on top of the case. So I would have 1 fan in the front pushing air in, 1 exhaust in the back, and 2 exhaust on the top. Maybe this would cool my overall system more since I never really needed the CPU liquid coolant in the first place, I'm not overclocking or anything..

 

[Knightmare10]

Another thing to try. Start a game or intensive task and check your voltages in HWMonitor if you still have that installed. I believe it shows voltages at the top, and check to make sure your +12v is near 12v. Small fluctuations such as .5 MAX is safe. If you cant conclude enmough from that id buy a new psu for cheap. Ultra is a very cheap manufacturer that makes okay psus. I bought one for my girlfriends family gaming rig and its been chugging for nearly a year.

Hey man so the 12v values are:

Min: 11.880 V Max: 11.968 V

 

[Knightmare10]

Another thing to try. Start a game or intensive task and check your voltages in HWMonitor if you still have that installed. I believe it shows voltages at the top, and check to make sure your +12v is near 12v. Small fluctuations such as .5 MAX is safe. If you cant conclude enmough from that id buy a new psu for cheap. Ultra is a very cheap manufacturer that makes okay psus. I bought one for my girlfriends family gaming rig and its been chugging for nearly a year.

What should the temps of the PSU be around? If Im reading this correctly, the TMPIN1 temp is MIN: 85°C and MAX: 94°C.

Compared to the TMPIN0 and TMPIN2, the TIMPIN1 is much, much higher... The 0 and 2 sit at around 40°C, while the 1 is at 86°C.

 

[aaronfield]

The +12v is fine then.

Those are just the temp sensors on the mobo. You will have to physically feel the psu while playing a game or bench-marking. 85 degrees is quite a bit even for a mobo. Could be a faulty sensor too though. My last board said that it was 236 degrees C constant and was faulty to all hell. Do you have any friends with computer parts? If you do just borrow a friends psu and test it out. If that is the case buy an Ultra psu for cheap and live happily ever after. also, while i dont think this could crash a computer completely, did you do a "clean" uninstall of the nvidia drivers before reinstalling? it could be a setting from your 260 that messes with the 770 due to age difference/differing lineups. If you dont know what I mean a quick Google search will remedy that.

While doing anything i guess reach into your computer(with proper grounding) and feel the heatsinks attached to the mobo. If any of them feel extremely hot then try to aim a fan at it. I suppose you wouldn't get a BSOD if your northbridge or southbridge overheated and shutdown.

Plus you can still use the liquid cooler, just have the fans blowing out instead of in. But definitely add a second fan on the top.

Another note: What you want to do with your fans is going to shoot dust into your computer like crazy due to negative airflow. Better stock up on canned air if you do it.