PC shutting off during heavy gaming

[MacDurious]

First, here are the specs for my rig (everything is 8 months old except PSU)

GPU=GeForce GTX 980ti 6gb
CPU=Intel(R) i7-6700k cpu @4.00GHz (not overclocked)
RAM=16 gb DDR4
PSU= Thermaltake smart m 850w (3-4 years old though)
MB= gb tech co. Z170-HD3-CF

During games like DA inquisition, witcher 3, and ME andromeda, my pc will shut off for a few seconds, turn back on with fans roaring but display never comes back up (I can get 5 minutes-2 hours game time before this happens). Games like BF1, SWBF1, TW Attila, SWTOR, World of Warships I have no problems running however. I thought overheating of the GPU was problem however my rig turned off while it was barely warm to the touch just this morning while playing ME andromeda for no more than 3 minutes. This has been happening total for 2 months now. I have no idea how to measure PSU or what volts/amps/watts readings are good or bad. Googling will only get me so far and ive run out of things to look for. Does anybody have an tips or tricks to do or figure out what my problem is?

p.s. pc never turns off while doing normal everyday things on it, so far just certain games forces a shutoff

 

[bboiprfsr]

try fully uninstalling your graphics drivers (link: http://www.guru3d.com/files-details/display-driver-uninstaller-download.html)
then install the latest drivers again. check cable connections to and from power supply. sometimes loose cabling to your hard drive may be an issue. could also check temperature readings during a stress test (i.e. prime 95) and see for anything abnormal. Since you mentioned normal work tasks are A-ok, the fact that certain games force a shutdown should be graphics card related. I suspect video drivers and/or faulty graphics card. Run some graphics card benchmark tests and measure temps. Computer has a tendency to shut down if things go unstable for safety/preventative measures.

We need to also analyze your operating system's crash dump files to further diagnose what could possibly be crashing your computer system.

 

[chemmajorp53]

Sounds like a power supply issue to me, but it could conceivably be something else I guess. Getting readings off your PSU won't help, it's probably not capable of putting out nearly 850W now that it's so old considering that it was a pretty low quality PSU even when it was new.

 

[MacDurious]

try fully uninstalling your graphics drivers (link: http://www.guru3d.com/files-details/display-driver-uninstaller-download.html)
then install the latest drivers again. check cable connections to and from power supply. sometimes loose cabling to your hard drive may be an issue. could also check temperature readings during a stress test (i.e. prime 95) and see for anything abnormal. Since you mentioned normal work tasks are A-ok, the fact that certain games force a shutdown should be graphics card related. I suspect video drivers and/or faulty graphics card. Run some graphics card benchmark tests and measure temps. Computer has a tendency to shut down if things go unstable for safety/preventative measures.

We need to also analyze your operating system's crash dump files to further diagnose what could possibly be crashing your computer system.

I would like to analyze the operating sytems crash dump files but havnt found any google ways to find them. Any suggestions?

 

[MacDurious]

Sounds like a power supply issue to me, but it could conceivably be something else I guess. Getting readings off your PSU won't help, it's probably not capable of putting out nearly 850W now that it's so old considering that it was a pretty low quality PSU even when it was new.

I didnt even know it was a low quality one to begin with. I am starting to think PSU is the culprit because of the age, now taking into account its a lower quality one, its possible but I dont know how to diagnose a PSU, any suggestions?

 

[JaredEN]

The first thing that comes to mind is cpu overheat protection. Do you have a good cpu cooler? Nobody has had real issues with old PSUs, as far as I know. You won't be able to tell the temperature by just touching the externals, the core temps can be vastly hotter, especially without a good cooler.
To monitor core temperatures, here is some accurate freeware.

Use afterburner to check GPU temps:
https://www.msi.com/page/afterburner

Core Temp for CPU temps:
http://www.alcpu.com/CoreTemp/

Real Temp is another good CPU temp monitor, and it's designed for Intel specificly:
https://www.techpowerup.com/download/techpowerup-real-temp/

 

[bboiprfsr]

again, i thought i had power supply issues too (given a 3-year old OCZ power supply), so i spent over $100 on a new 750w corsair psu). That wasn't the case! I just uninstalled my old graphics drivers (b/c each time you rewrite over it through upgrades, it causes instability issues). Just my two cents.

 

[chemmajorp53]

You can't really diagnose the PSU with regular tools. I'd try all the things that the other guy said (Use DDU to uninstall GPU drivers, reinstall, check all your power connection points) but I don't expect the GPU to be faulty in this case. If it was, you would be having problems all the time, not just when it's under the most load. If a computer shuts down due to temperatures, it will usually blue screen, and it doesn't sound like that's what's happening. If you or a friend have another power supply then I'd try it and see if it fixes anything, if not then you might be stuck having to buy one. For what it's worth, most people here would suggest that you replace that PSU even if it isn't faulty.

Something like this

https://www.newegg.com/Product/Product.aspx?Item=N82E16817151132&cm_re=gold_psu-_-17-151-132-_-Product

Or this

https://www.newegg.com/Product/Product.aspx?Item=9SIA2F84BD0013&cm_re=gold_psu-_-17-438-060-_-Product

would be high quality replacements that have long warranties and should last you a lot longer than 3-4 years.

Refer to this list for a vague idea of how quality a PSU is.

http://www.tomshardware.com/forum/id-2547993/psu-tier-list.html




TL;DR Try out some things that don't involve spending money and if they don't work try getting a new PSU

 

[MacDurious]

The first thing that comes to mind is cpu overheat protection. Do you have a good cpu cooler? Nobody has had real issues with old PSUs, as far as I know. You won't be able to tell the temperature by just touching the externals, the core temps can be vastly hotter, especially without a good cooler.
To monitor core temperatures, here is some accurate freeware.

Use afterburner to check GPU temps:
https://www.msi.com/page/afterburner

Core Temp for CPU temps:
http://www.alcpu.com/CoreTemp/

Real Temp is another good CPU temp monitor, and it's designed for Intel specificly:
https://www.techpowerup.com/download/techpowerup-real-temp/

I am not familiar with acceptable CPU temps. I think, with my specific GPU, 80 C is ok, 90 c is where i should start to worry. However CPU I havnt the foggiest idea

 

[JaredEN]

between 0-40C is amazing, 40-60C is good, 60-70C is not good, 70C+ you should be very concerned.

 

[MacDurious]

The first thing that comes to mind is cpu overheat protection. Do you have a good cpu cooler? Nobody has had real issues with old PSUs, as far as I know. You won't be able to tell the temperature by just touching the externals, the core temps can be vastly hotter, especially without a good cooler.
To monitor core temperatures, here is some accurate freeware.

Use afterburner to check GPU temps:
https://www.msi.com/page/afterburner

Core Temp for CPU temps:
http://www.alcpu.com/CoreTemp/

Real Temp is another good CPU temp monitor, and it's designed for Intel specificly:
https://www.techpowerup.com/download/techpowerup-real-temp/

So I uninstalled/reinstalled gpu drivers, tuned down the graphics to high on ME andromeda (by default the game put damn near everything to ultra or its equivalent), cpu temps were at 40, gpu temp at 60 and it still shut down. Im really starting to think something is simply wrong with the psu now.

 

[chemmajorp53]

On a 6700K? I think you're thinking of AMD temps, 70C is perfectly fine for an intel processor. Anything under 80C is fine and it won't shut down until 100C or 105C depending on the processor

 

[JaredEN]

On a 6700K? I think you're thinking of AMD temps, 70C is perfectly fine for an intel processor. Anything under 80C is fine and it won't shut down until 100C or 105C depending on the processor

Mine would die at 98C theoretically. It maybe fine at 70-80C for little bits of time, but it's probably shortening the cpu's life span.

 

[MacDurious]

On a 6700K? I think you're thinking of AMD temps, 70C is perfectly fine for an intel processor. Anything under 80C is fine and it won't shut down until 100C or 105C depending on the processor

I promise. Game was running for only 5ish minutes and I was looking at both the gpu and cpu temps when it shut off. Currently, as im looking at my cpu temps, its at 15-18C. Either way, my temps were well in the norm right before it shut off. What else could you think is happening?

 

[MacDurious]

How can i find the crash code?

 

[chemmajorp53]

On a 6700K? I think you're thinking of AMD temps, 70C is perfectly fine for an intel processor. Anything under 80C is fine and it won't shut down until 100C or 105C depending on the processor

Mine would die at 98C theoretically. It maybe fine at 70-80C for little bits of time, but it's probably shortening the cpu's life span.

Maybe shortening the life span from 15 to 14 years but who cares at that point, 70C is lower than normal. Intel wouldn't ship a stock cooler that the CPU hit 80C on if they weren't confident that it was ok.

@Macdurious, 15-18C isn't a valid temperature for a CPU unless 0C is room temperature. We can't know for sure what's happening, this is trial and error. Until you try the fixes we told you earlier then I personally can't help you anymore.

 

[MacDurious]

On a 6700K? I think you're thinking of AMD temps, 70C is perfectly fine for an intel processor. Anything under 80C is fine and it won't shut down until 100C or 105C depending on the processor

Mine would die at 98C theoretically. It maybe fine at 70-80C for little bits of time, but it's probably shortening the cpu's life span.

Maybe shortening the life span from 15 to 14 years but who cares at that point, 70C is lower than normal. Intel wouldn't ship a stock cooler that the CPU hit 80C on if they weren't confident that it was ok.

@Macdurious, 15-18C isn't a valid temperature for a CPU unless 0C is room temperature. We can't know for sure what's happening, this is trial and error. Until you try the fixes we told you earlier then I personally can't help you anymore.

Ok i guess im just reading it wrong then, any way i can post a picture on here? its just giving me links for shit

 

[MacDurious]

SO went a little risky and just bought a new power supply. Played 6 hours straight on ME Andromeda without a single crash (before it would crash after 5 minutes of playing). Seems this was the culprit

 

[chemmajorp53]

Yeah I figured that was it, hopefully you don't get any more crashes. What did you get?

 

[MacDurious]

Yeah I figured that was it, hopefully you don't get any more crashes. What did you get?

Evga 750 gold. Purrs like a kitten now!