PC intermittently freezes and resets. No BSOD

Toggle sidebar Toggle sidebar
C

CapitalFlight

Honorable
Feb 23, 2014
19
0
10,510
I purchased a new 1080 for a new rig. I put it in my current rig just to make sure it wasn't DOA. It all ran fine and was very pretty. I switched it back to my GTX 770. When I was doing so, I knocked one of my ram ports. I've previously had big issues with my RAM.

Anyway, long story short, after doing the above my PC is intermittedly freezing and auto-resets at seemingly random moments. I'm not getting a BSOD


  • WhoCrashedIt doesn't have any details.

    I re-seated my RAM. Memtest made 3 passes with 0 errors.

    I've double checked power cables going into GPU. They're snug.

    I did a clean install of GPU Drivers by running DDU and then clicked "install clean" on the Nvidea installer.

    CPU and GPU temperatures look perfectly normal. Right now CPU cores are at 30 deg Celsius, GPU is idling at 32 deg (using EVGA Precision). I just managed to play Dirt Rally for an hour, GPU temp was ~55.

My running theory is that it's either the GPU or the PSU

How do I check that voltages are normal? What else can I check? (One thing I haven't tried is to re-seat the GPU)

CPU: i5 2500
PSU: Coolermaster TX 750w
GPU: Gigabyte Windforce GTX 770

Appreciate any and all help
 
Sort by date Sort by votes
Roland Of Gilead

Roland Of Gilead

Titan
Aug 8, 2008
10,341
2,266
77,690
Very clear description :)

It's possible it's the PSU. Great in it's day, but is it nearly 5-6 years old now? It's possible it's failing and not providing enough power, seeing as the crashes are random.

It could also be the ram slot being slightly damaged. Have you tried it with just one stick in the slot that you feel wasn't damaged?

You can run something like HWMon/Info to monitor power draw on certain components. CPU/GPU/MOBO etc. So you can get an idea of the draw at idle/load. The TX750 if it were running well, would be plenty yo power your system. Your probably not even hitting 400w usage.
 
Upvote 0 Downvote
C

CapitalFlight

Honorable
Feb 23, 2014
19
0
10,510
Cheers.


Well, I'll add some details on the RAM. This is the "long story" part I skipped in OP:


- A couple years ago I upgraded from some Corsair XMS3 DDR3 1333 8GB (2x4) RAM to Corsair Vengeance 1600 16GB (4x4). It worked fine.

- One day (a year or so later) my PC wouldn't boot, with no display. Motherboard beeps led me to believe it was RAM. I opened her up and found the Hyper 212 CPU Cooler was pushing against the tall Vengeance card in Slot 1. I downgraded to 8GB (2x4) using slots 2 and 4 and (with a lot of trial and error) eventually got it running nice and stable again.

- I had some similar issues arise this time around (i.e. PC not booting/no display, giving off the same number of beeps from motherboard beeper). I got it working again by putting the old Corsair XMS3 into slots 1 and 3. Notably, it took a fair bit of trial and error switching sticks around, getting one in Slot 1 working solo, then placing in the second to slot 3.


So OVERALL: I am getting the exact same issue when I have either the Vengeance in slots 2 and 4, or the XMS in slots 1 and 3. This commonality, plus the successful memtest (I honestly expected plenty of errors) has lead me to believe its PSU or GPU related - especially as everything was perfectly stable before I trailed the new 1080.


The mobo, ram, and PSU were all purchased in 2011. I'm currently building a new rig, but would preferably be able to sell this one for a bit of cash, so I need to get it working.

I just don't understand how it could be the PSU if I had it stable before trialling the 1080. I will try your suggestion to try HWMon/Info and report back.
 
Upvote 0 Downvote
Roland Of Gilead

Roland Of Gilead

Titan
Aug 8, 2008
10,341
2,266
77,690
It could be a number of different things as you've pointed out. Including PSU, Ram, Mobo (damaged sockets) and age of components.

The additional power draw of the 1080 could have pushed the PSU a bit (if it were starting to loose efficiency or power output). It could be the ram also, although with ram it's more likely to have a BSOD and as you've said it's come back clear on memtest.

I'm leaning toward PSU/Mobo.
 
Upvote 0 Downvote
C

CapitalFlight

Honorable
Feb 23, 2014
19
0
10,510
I've installed HWMonitor and it simply says the GPU has 0.950 V. In precision X it says the same, except when in game where it goes up to 1200 V.

Today the PC hasn't crashed, and I've played Dirt Rally just fine. However PUBG ran terribly, despite Steam telling me I was getting 60FPS, it felt more like 20 FPS. CPU and Memory were ~60% usage, GPU at 100%.

I'm not really sure what to do. Everything appears to be running normally, but something is clearly causing an issue with select games.

 
Upvote 0 Downvote
Roland Of Gilead

Roland Of Gilead

Titan
Aug 8, 2008
10,341
2,266
77,690
Run a game, and HWMon at same time. Alt-tab out of the game and take a screenshot of HWmon. We can look at the metrics and give a better assessment. Power usage, voltages, CPU/GPU usage, mem. Seeing the full picture with all details may help us.

are you using Geforce experience for games? Have you changed any nvidia control panel settings?
 
Upvote 0 Downvote
C

CapitalFlight

Honorable
Feb 23, 2014
19
0
10,510
Hi Keith, sorry for the slow reply. It's been a strange week.

My PC is plugged into a surge protector power board. At night, I switch off this power board because even when shut down the PC is giving power to my USB peripherals which keeps their bright LEDs on; I've not been able to find out how to turn this off in my bios.

Everytime I switch it off this surge protector power board, the "fault" light quickly flashes off with it. I've never thought anything of this. A few nights ago when I switched it off, it actually tripped the safety switch for the entire house. Could it have been the root cause to my issue, or sheer coincidence? I'm not sure.

I've since stopped switching off this board. My PC has stopped intermittently freezing, however it auto-restart after I put it to "sleep", and games are running very poorly - especially PUBG. It feels like I have a major download going on in the background - everything is a bit laggy. The Steam FPS counter says I'm on high FPS - Rocket league as high as 140, however it feels no better than 30 FPS. PUBG feels like 15 FPS

Here's a shot of hardware monitor as I was in a game of PUBG: https://i.imgur.com/j9TScQV.png

Are you able to identify anything that looks off?

 
Upvote 0 Downvote
C

CapitalFlight

Honorable
Feb 23, 2014
19
0
10,510
Hmmm, I've just found in my windows 10 notifications "Display driver failed to start; using Microsoft Basic Display Driver instead"


It would appear my graphics driver isn't kicking in? This is strange, I would have thought I uninstalled and reinstalled it perfectly fine. I'm at a bit of a loss of what to do, this would most likely be the reason for the functional computer but shocking game FPS


Edit: I've just reinstalled the gpu driver again, but used an older one from Dec 2017. I suspect this could be the fix...
 
Upvote 0 Downvote
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom