Voltage might be freezing my GPU?

Kurby_

Commendable
Oct 22, 2016
5
0
1,510
Hello, I've been experiencing these crashes and freezes that happens totally randomly (happens more while playing a game but never happens under FurMark load) for about a month now and I've ran out of ideas on how to find out, finally I came here to ask if someonelse could help.
Crashes happen randomly, sometimes even on desktop, it freezes for about 3 seconds to 40 seconds (varies). Usually the game freezes for that time and then comes back like nothing happened, sometimes sound starts looping and the game crashes and sometimes game continues running (I can hear what I'm doing).
The computer does not freezes totally, I can move the mouse and there's a loading icon on the mouse during the whole freeze.
After I analyses a lot of logs from GPU-Z, GPU seems to totally stops.

Screenshot: http://imgur.com/dcW5hSN


Specs
OS: Windows 10 x64
CPU: Intel Core i5 760 2.8GHz (With a Watercooler)
GPU: NVIDIA GeForce GTX 760 (a little OC but problem happens even when underclocked)
RAM: 2x Kingston DDR3 4GB 1300MHz Dual channel
Motherboard: Gigabyte H55M-S2HP
HDD: - 1 TB Seagate (don't know exactly which one)
PSU: Thermaltake 500W (don't know which one it is, I bought it on 2012)


What I have already done:
Cleaned RAM and GPU connectors with isopropilic alchool and ereaser.
Changed GPU thermal paste (seems to have lowered the temperature a lot after that).
Uninstalled all old GPU drivers and installed clean latest one (and tested with two older versions)
Updated BIOS
Changed RAM slots
Memtest98 showed 108 errors after 12 runs.
Cleaned the whole machine
Defragmented HDD and made a CHKDSK (no errors)

What more can I do? I hope this is enough info to help me out, although i'm already desperate because this ghost has been haunting me for about a month or more. I'll try to provide any extra info about it if you're willing to help! :D

Thanks in advance
 
Solution
I'd be inclined to look towards the PSU given the troubleshooting you've already performed.

I can;t think of a 'good' 500W Thermaltake, especially when you add 4+ years of use.
How do the +5, +3.3 & +12V look? HWMonitor will show you, but I'm sure other programs can too.

Barty1884

Retired Moderator
I'd be inclined to look towards the PSU given the troubleshooting you've already performed.

I can;t think of a 'good' 500W Thermaltake, especially when you add 4+ years of use.
How do the +5, +3.3 & +12V look? HWMonitor will show you, but I'm sure other programs can too.
 
Solution

Kurby_

Commendable
Oct 22, 2016
5
0
1,510


Hey Bart!

Yea, my PSU is kinda old maybe it is the problem.

Anwsering your questions:
+3.3V varies from 3.128V to 3.200V
+5V varies from 4.896V to 4.946V
+12V varies from 0.704V to 9.152V (varies a lot?)
all of these values while in desktop.

And here's a screenshot in the exact moment of the game freeze (after about 20 minutes playing witcher 3 it froze, as always, got frozen for about 20 seconds and then came back normally)
http://imgur.com/9L9SIEg
 

Kurby_

Commendable
Oct 22, 2016
5
0
1,510
Today as I woke up, turned on the computer, this alert message from BIOS showed up: https://i.imgur.com/sWs4XsD.jpg
I have never overclocked the CPU nor RAM. Only GPU.
As time goes by more things makes me think the problem is on the PSU and not GPU.

Should this thread be moved to PSU issues?
 

Barty1884

Retired Moderator
Something is really wrong there.

Those +12V numbers are really off.... Your CPU also seems to be hitting max speed at 8.1 to 8.5GHz and 100% CPU utilization at times.

I'm on my phone at the moment, so can't do too much research, bug I'll try call in some other opinions.


 

Kurby_

Commendable
Oct 22, 2016
5
0
1,510


That CPU Usage looks like something punctual and a visual bug, Core #1 is showing his Max at 8373740 MHz



 

Kurby_

Commendable
Oct 22, 2016
5
0
1,510
Just to state the solution if someonelse has a similar problem, the problem was the Power Supply Unit (PSU). Bought a new one and after some hours of testing, the problems seem to be fixed.