Path of Exile overheating GPU

"
f3rret wrote:
"
Astasia wrote:
You simply shouldn't be having problems with GPUs overheating if you aren't overclocking. They are designed to be able to run at 100% on factory settings and keep themselves at a safe temperature with just basic case fans helping with airflow. If you clean your system out and it's still overheating (and isn't overclocked) then there is something wrong with it.


Well to tell the truth, I'm not actually sure it's the GPU overheating.

Here is what happens (it can happen during the Vaal fight too):

I'm playing merrily.
Sudden screen goes dark, computer freezes.
Computer reboots.

SpeedFan checks of GPU temperature shows it hovering in the 64 - 70 degrees Celsius soon as computer comes online again.
I could be something different, I've not noticed the problem with anything other than PoE.
It seems like it might be overheating, as I've had it happen before with older graphics cards.



Could be the memory. I once built my own computer, and had similar problems before finally finding out that one of my memory sticks was bad.
Should be able to find a program somewhere to test it.
That's not the GPU overheating, but the temperature could be related. It sounds more like an inadequate power supply, your card might be kicking the fan into high gear creating a power shortage and causing the system to reboot.
"
Astasia wrote:
That's not the GPU overheating, but the temperature could be related. It sounds more like an inadequate power supply, your card might be kicking the fan into high gear creating a power shortage and causing the system to reboot.


Unlikely.
The computer does not lose power at any point. When the computer crashes I can hear whatever sample was playing at the time playing on endless buzzing repeat until the machine finally reboots.

When I get back into windows, it sometimes pops up with a message saying 'windows had an unexpected crash' (paraphrasing) which lets me access memory dumps similar to those generated by any other BSoD.

I can play solo all day long, only time I've EVER seen this crash is on the Vaal Boss fight and earlier today with a 6-man group.

However, I am not a hardware guy.

So if you could take a look at the computer stats listed in first post and make a guess as to whether or not that could run on a 500W PSU.
Closed beta member since: March 19, 2012
Last edited by f3rret#1161 on Feb 15, 2013, 5:53:57 PM
It wouldn't appear to lose power, the fans might not even slow down, it's just the CPU/mobo not getting enough power for a split second that will force a reboot. It most likely wouldn't play sounds for a period of time like that though, so you are now describing something that sounds more like a severe game crash or as somebody else mentioned a memory related system crash.

Only thing I can say for certain without looking at your machine in person is it's not the GPU overheating. 70C isn't even hot, GPUs these days can get up to 95C while working hard and still be safe, and if they do get too hot what you would see are graphical anomalies on your screen or just a GPU crash, it likely wouldn't crash your whole system.

I can tell you I had a crash during Vaal myself a while back, which was the only time the game actually crashed on me.
"
Astasia wrote:
It wouldn't appear to lose power, the fans might not even slow down, it's just the CPU/mobo not getting enough power for a split second that will force a reboot. It most likely wouldn't play sounds for a period of time like that though, so you are now describing something that sounds more like a severe game crash or as somebody else mentioned a memory related system crash.

Only thing I can say for certain without looking at your machine in person is it's not the GPU overheating. 70C isn't even hot, GPUs these days can get up to 95C while working hard and still be safe, and if they do get too hot what you would see are graphical anomalies on your screen or just a GPU crash, it likely wouldn't crash your whole system.

I can tell you I had a crash during Vaal myself a while back, which was the only time the game actually crashed on me.


Hm, alright. I'll try running memtest and see what it gives me.

That said, the temperature of 70C I mention is not the temperature at crash-time, it is what it shows after the computer reboots, for obvious reasons I cannot see the temperature at crash-time.
Closed beta member since: March 19, 2012
Try checking your Windows System Logs. It should give you a little more information about why your computer is rebooting suddenly.

You can access them by Right-Clicking on "My Computer" and selecting "Manage"

From there you can access Performance Logs and Alerts on the left hand pane. I can't recall offhand which category the logs you're looking for will be in (and I'm unable to check at our work computers) but it should be fairly obvious. Big red exclamation points will be the major issues.
"
f3rret wrote:
That said, the temperature of 70C I mention is not the temperature at crash-time, it is what it shows after the computer reboots, for obvious reasons I cannot see the temperature at crash-time.


Ya but it would take more time to drop from 110C to 70C.
I'll try looking through logs next time I have a crash.

As it happens right now there's so much other stuff in there and I cannot remember what time the crash occurred at.

I'll run memtest+ tomorrow and see if I get any errors.

Will update with results after.

"
Astasia wrote:


Ya but it would take more time to drop from 110C to 70C.


You'd be surprised actually, a while back I was messing about with mining bitcoins and doing GPU mining would easily push the GPU to 80 or so degrees, stop the mining and the temperature drops back to ~40 in like 10 seconds.
But I'm getting off topic here.
Closed beta member since: March 19, 2012
Last edited by f3rret#1161 on Feb 15, 2013, 6:28:56 PM
Hey there!
What's your result with memtest?
i had theese problems too, changed my memory and voala works perfect.
i have 2x4Gb 1600Mhz Kingston HyperX Blu
be sure that you put the memory modules into same colored memory sockets like so:
Here's your motherboard: /S=socket M=memory module/
Sockets could be S S S S or S S
If you have four sockets modules should be like:
M S M S
oh and you should try disable X.M.P/Extreme Memory Profile/ in your bios.
But if you think that your GPU is the problem set up a custom fan profile with msi afterburner.
My ophinion is it could be the memory or your overclocked cpu is not stable, you should try put the cpu clock and voltages to stock.
If you don't know your defaults, search in google for your processor and on ark.intel.com you find every value of your processor.

Hope that helped, Cheers.
DjSall's LL Tank Crit EK: 1568944
Youtube Channel: http://youtube.com/proksalevente

Report Forum Post

Report Account:

Report Type

Additional Info