P775DM3-G : BSOD while gaming

Hey,

I bought my Clevo from mySN. It's the XMG U717 model which is basically the P775DM3-G. I was really happy with this beast... until I ran into a problem recently: After a while the laptop crashes when gaming. The screen goes black, the fans turn up and the system shuts down. I have installed the latest drivers but the problem still occurs. Loading default settings in the BIOS didn't help either. What could be the issue causing all this?

I have the model with the i7 6700K, GTX 1080.

 

loading bios defaults is a good first step in getting to the issue. so we can rule out any unstable undervolts / overclocks or such. another thing to check is your system file integrity by running chkdsk, reseating RAM, CPU and GPU, as well as making sure that no components are overheating (repaste, check proper contact of thermal pads on the gpu). one last culprit could be the PSU. did it shut off as well (did the LED go out)?

 

So I checked following:

• Undervolting / overclocking: Everything is on default
• CPU & GPU temps: cpu is between 70-76 and gpu ~70-ish
• chkdsk on harddrives: All ok
• PSU: LEDs still on. Seems to be working.
• RAM: ?? What do I need to do here? Reinsert them?

 

yup, basically. pull out the RAM, wipe down the contact pins and reinsert the modules. if this keeps happening, try switching out the sticks one by one and alternating the ram sockets. could be faulty ram.

if the problem still persists, check cpu and gpu installation by reinserting those as well, applying new thermal paste and check for proper placement / contact of the gpu thermal pads. might be overheating of components without thermal sensors....

 

I tried to play a game (Miscreated. cry engine game) with an undervolt of -95mV. It still crashes but this time I can alt-tab out to windows. Windows seems to be working fine. In the game I still get a black screen though. Again, temps are normal (max. 77 on GPU and ~70 on CPU).

This looks like a GPU problem. Maybe faulty drivers? Driver version is 376.19

 

Check the source of the BSOD using this instead of guessing.

 

Something other than your GPU core is heating up.
Or if you have an overclock on the GPU, that might need to be dialed down.

 

I'm inclined to agree. If you're alt + tabing to check temps rather than using an OSD, they were likely a fair bit higher than that while playing.

 

Ok. I will post the results from BlueScreenView when I get back from work.

@bloodhawk I have no overclock on the GPU.

 

While I'm at it, I want to replace the thermal pads on the GPU. Do you know how thick they are? Will 1mm suffice?

 

that depends on your gpu and heatsink

Sent from my Huawei Mate 8 NXT-AL10

 

What if the thermal pads are too thick? Will it cause any problems?

 

Yes. less contact with the GPS due to the increased elevation they might create

 

Gotcha. Then there is a good chance something is not making proper contact with the GPU heat sink. (other than the core)

 

Here is the logfile from HWinfo. But I noticed that the system no longer crashes, only the game.

 

what happens if your run AIDA64 Extreme System Stability test for 3 mins? Can you run it and have HWiNFO running in the background to show us the temps then take a screenshot of both and post it here?

 

Here is the screenshot. No crash yet.

 

ok so temps are good. at least that one is out of the way

what WLAN card do you have and what driver?

It would be helpful if you put a sig like everyone here so we don't have to keep asking you about your system specs

 

I just updated my signature. How is the WLAN card relevant though?

 

the reason I asked, is when I had an Alienware 18 and had the Killer Suite installed which installs a network traffic filter driver, that used to cause BSODs so you're fine since you have an Intel WLAN card

Maybe the RAM is causing the issue?

 

Good chance something on the GPU is not making proper contact.
The game crashes / black screens at stock clocks usually point to that.

 

So no BSODs when gaming, right? Just a black screen?

I'd have to agree with your earlier prognosis: faulty GPU (still just a prognosis, though).

 

WLAN cards can cause some crazy issues that you couldn't even believe.


To me, it sounds like it could be a bad GPU.

 

Hey, I experienced something similar recently with my P870DM-G with dual 980m, the screen blacked out, fan maxed out, power LED blinking orange but it didn't shut down, I sent it back to mythlogic and they've been testing it for two weeks, still waiting for them, what did you do to fix it?

 

So, I opened this baby up and added some additional pads / repasted the whole thing. Now it seems working (no crashes yet). But now I have higher temps (GPU before: avg. 75 C -> now: avg. 80 C). The temps are a bit higher because some pads are too thick, right?

BTW, here are the spots where I put the thermal pads:

 

Try using an external image host. The forum one has been broken for ages.

 

Is it just me, or do one of those CPU pins look a little funky...? (Near the top left of the socket.)

 

It's not my Clevo, actually. I just took a picture from google as a reference.

 

A too-thick pad may be one cause, or there could be something not making full contact. Wouldn't be by much or you'd be in the upper 90s in seconds, but I'd take a look at the extra pads first.