add arrow-down arrow-left arrow-right arrow-up authorcheckmark clipboard combo comment delete discord dots drag-handle dropdown-arrow errorfacebook history inbox instagram issuelink lock markup-bbcode markup-html markup-pcpp markup-cyclingbuilder markup-plain-text markup-reddit menu pin radio-button save search settings share star-empty star-full star-half switch successtag twitch twitter user warningwattage weight youtube

Possible GPU/CPU Overheating **NEED HELP**

Turf357

17 months ago

Fairly recently, my home-built desktop computer shut down randomly without much warning. A strange flicker or glitch of the screen happened before everything going completely black. My desktop was still powered, but nothing was being displayed to the screens. This has happened once previously, and I didn't think much of it the first time. The same flicker occured, but I was able to shut down the computer and power it back on after a short while. I've noticed that when my PC is running for a considerable amount of time, the room that it is house in gets extremely hot. Other people have said it's hot in this room, but I couldn't feel the heat all that much. When I put my hand near the fans, it did not feel like the air was very hot, but I could've been mistaken. I wouldn't say I had the best airflow with only 2 fans + gpu fan + cpu cooler fan and an semi-enclosed area underneath my loft bed. The door to this room is usually closed while using the PC. I have not been able to fully boot my computer up all day. When I do, my computer powers on and everything seems fine. Sometimes the bios screen appears, but the monitors appear to have many vertical lines on them. These lines take up a wide space, but are transparent with a reddish hue. Then, the windows start up screen appears and loads for a short while. At about the time when I'd expect it to finish booting, the screens turn black again, but without a glitch/flicker. The monitors say there is no input source and then I turn off the PC. My best guess is this is a result in CPU or GPU overheating, but I don't know how to confirm or approach this problem. I was hoping someone on here could tell me more about what is happening. Thank you!

I have tried just using my motherboard plugs instead of the ones from the GPU, and absolutely nothing happens. Before I switched them, I tried to turn it on a couple times. It shows the bios and windows setup screen just as shown above. One of the times, it showed the blue screen and said "windows could not load correctly". The screen was still being strange at this point, but I troubleshoot the error. It said that windows troubleshooter could not fix this error. Another time it got past the windows set up screen, but, instead of showing the sign in screen, it showed a glitched screen with many red vertical/horizontal lines.

Comments

  • 17 months ago
  • 2 points

Is you system in a custom cooled rig? Have you checked temps via HW monitor? Do some testing to figure out if that is the issue and then maybe re apply thermal paste or set higher fan curves, add more fans, etc.

EDIT: I read through more of it and it sounds like a GPU issue. Maybe something is shorting the GPU or it could be temps, try re installing the GPU and see if that helps. From what the rest of the details say, it sounds like you're experiencing artifacts. Have you also OC'ed your GPU?

  • 17 months ago
  • 1 point

It definitely is not a custom cooled rig. As this is only my second build ever, I don't have quite the skills to do anything fancy. Would you explain what HW is? I haven't even tried opening the bios yet, but I plan on doing that soon. The temperatures are not high enough on start up to ruin the GPU now, but it could have damaged it earlier. I have never overclocked my GPU. I will see if bios runs and I will try reinstalling the GPU.

  • 17 months ago
  • 1 point

Try re installing the GPU first. If you can boot to windows download HW monitor from CPUID. That will allow you to check your system temps and will tell you if temps are the issue. You can also re install the drivers. Since you've never OC'ed the GPU, I'd assume you got a bad card. If you still have warranty, you can RMA it and get a new one. This happens sometimes as maybe something is shorting the card out, or maybe an error.

  • 17 months ago
  • 2 points

Okay, reinstalling the GPU now

  • 17 months ago
  • 1 point

I have reinstalled my GPU, and windows 10 booted up first time with no problems. I have downloaded CUPID HWMonitor and I will give you the temperatures of both my CPU and GPU at idol (with hwmonitor and one tab of google chrome open).

AMD Ryzen 5 1600 Temperatures Package (Note 0) 31C (87F) 30C (85F) 41C (106F)

AMD Radeon (TM) R7 360 Temperatures GPU 40C (104F) 37C (98F) 40C (104F)

I am unsure of what a normal temperature and too high of a temperature is. After I send this message, I will be shutting down the PC and waiting for a response.

  • 17 months ago
  • 1 point

What is your room temp at 40c and higher is a bit warmer than it should be, Try putting some load onto them by running a game or benchmark like PUBG or AIDA64. They both will tax both the CPU and GPO enough to show the max temps for both parts.

  • 17 months ago
  • 1 point

My room temperature runs normally around 22.5c (72.5f). I'll be trying Aida64 for the stress test and report the results afterwards.

  • 17 months ago
  • 1 point

This topic has been moved from "Hardware > Other" to "Systems > Troubleshooting".

Sort

add arrow-down arrow-left arrow-right arrow-up authorcheckmark clipboard combo comment delete discord dots drag-handle dropdown-arrow errorfacebook history inbox instagram issuelink lock markup-bbcode markup-html markup-pcpp markup-cyclingbuilder markup-plain-text markup-reddit menu pin radio-button save search settings share star-empty star-full star-half switch successtag twitch twitter user warningwattage weight youtube