Computer Help Please! My computer keeps crashing after installing new GPU.

[JaydeDancer]

(1/2)

Hi all,

I upgraded my graphics card in January, and I am unable to play FFXIV because I will randomly lose my display and nothing I have tried to do has resolved the issue permanently. I will have to split this post to accommodate the forum restrictions. The rest of this post basically goes over what I HAVE done to try to resolve the issue.

Original computer specs were that of an HP Pavilion 500-054, purchased in 2013. Link to full specs here.

Current computer specs:
OS: Windows 8.1, 64 bit
Motherboard: MS-7778 JasmineR
PSU: Rosewell Photon 850 80 Plus Gold
CPU: AMD A8-5500
Ram: 8 GB DDR3-1600
GPU: XFX Radeon 560 RX (4GB DDR5 virtual memory) (driver 17.12)
HDD: 1.5 TB SATA
Monitor: Samsung S22C300H 21.5” LED TN
We purchased the GPU in 2017. We tried to install it when it was purchased, but my computer's original PSU wasn't strong enough to run it, and the case was too small to fit a larger PSU.

Fast forward to 2019. I bought a CoolerMaster Masterbox K500 (Link here) so that I could fit a larger power supply in with all the other computer innards. I bought an EVGA 600 w 80 Plus PSU to help run the new GPU, took my old computer apart, put everything in the new case, hooked everything back up, and turned it on.

At first, I couldn't get my display to run. My husband had the idea of turning off the RGB lights on my case and then restarting the computer. Lo and behold, I had a display again. However, I really didn't want to have to live without the lights. So I bought a stronger PSU. Went to a Rosewell Photon 850 80 Plus Gold PSU. I was finally able to get a display AND the lights on my computer. After that, I tried to switch from the integrated graphics card to the one I just installed, and I encountered issues with that too. One of the things I tried was changing the secure boot configuration in my BiOS to “Legacy”. That seemed to work because I was able to see the GPU and install the driver software.

But now I am encountering a different problem, one which is still occurring and is driving me up a wall. This is the issue I am trying to solve.

I can play FFXIV and have it on the max settings without any FPS lag.

BUT.

I will randomly loose my display. Losing the display is preceded by the screen turning one single, solid color (it varies every time), the sound glitches out, and the display goes dark and starts flipping between HDMI and analog, like it's looking for a signal. Then it goes dormant after a few seconds of trying, and failing, to find a signal.

[JaydeDancer]

(2/2)

Here is a list of things I have tried in an attempt to fix the problem:

• switching out the HDMI cable for a different one.
• Unplugging the HDMI and plugging it back in.
• Doing a hard boot on the computer, waiting a few seconds, and then restarting it.
• adding additional RAM to my computer so that I had a total of 16GB of DDR3 RAM (the other two sticks were Team Vulcan 2x4GB DDR3-1600.)
• When I kept having the same issue I ran a memory check and my computer found nothing wrong with my RAM.
• I removed the additional RAM b/c FFXIV is fine with 8GB.
• I checked the windows event log to see if there was something that was happening consistently right before the crash. The only thing I found was that the last thing that windows logged before the crashes was Bonjour Service having a critical error. But that might have been because of the thing that was making me lose my display, not the cause of it.
• contacting AMD support about which driver I should use since they no longer make drivers for Windows 8.1, and their response was “upgrade to windows 10”. -_-
• Trying different AMD driver versions (17.1, 17.7, 17.11, and 17.12)
• Installing MSI Afterburner, and changing the settings for the display to “force constant voltage”.
• Using MSI Afterburner to tell my computer not to allow the CPU to get over 120 degrees F. (This one actually seemed to have the most impact. I was able to log into the game and play it for a few hours without losing my display. However, not allowing the CPU to get very hot made it so the graphics card couldn’t function as well as it normally would and I got fairly significant lag in some spots.)
• Adjusting the temp settings to allow for a PSU temp of 153 degrees F so that I could play without lag (I tried this on Friday night, and the last reading I saw on the MSI Afterburner HUD was 136 degrees F. Shortly after that, I tried to run the benchmark and I lost my display again and couldn’t get it back, even after letting the computer sit for a few minutes.)

What I am wondering, is, if temp seems to be the primary cause of losing my display, and if I have to let the computer cool ALL THE WAY back to room temp before I can restart it and have my display come back on, does that mean that all of these problems are most likely related to the fact that my CPU is from 2013 and it is old? I don’t want to spend money on a new CPU if it isn’t going to fix the problem, but I don’t know how to diagnose the issue to determine if it really is the CPU.

Plus, I don’t know if it is possible for me to get a new CPU and have it work with my old motherboard. I also don’t know if it is possible to find a CPU/motherboard combo for under $300.

I’m hoping someone on here might be able to help me fix this so that I can play the damn game without worrying that my computer is going to crash on me. Ask questions if you need more information; I’ll try to answer them as best as I can.

[Valkyrie_Lenneth]

Sounds like your GPU is dying.

[Vaer]

Check the CPU temp first by downloading a utility like coretemp or hwinfo, turn the game on and monitor the temperature. For hwinfo use the sensors table. If it rises really rapidly the heatsink might be loose.

If the entire computer is shutting down the cpu is probably getting too hot. According to the spec sheet, the cpu will automatically turn itself off at 71C/160F. The cpu is kinda weak so it might be fully loaded when running FFXIV.

It's also a possibility that you have a dud power supply but less likely since using the integrated graphics does the same thing. (That processor and a 560 let alone just the processor by itself is barely using an actual working 600-850W 80 Plus)

Check cpu temps first. If they are ok since you have an extra PSU I would try to switch to the other one. If all that doesn't work it might be the GPU. You can also remove the card and run it off the integrated for a bit and let it sit there to see if it is stable (it will lag but just check for stability/no crashing for a couple minutes) to check if it's the gpu.

[LashL]

I had a similar problem years ago when i blew my 780ti screen would go black but i had sound in 14 still (could move around etc just not see) signaling a gpu failure, but if the whole thing just shuts off its most likely way to taxing on your cpu.

[Rannie]

Ok complete nm

[Amnmaat]

Did you do anything in the BIOS? Like did you overclock the CPU using BIOS settings? Also did you build the computer? If you did did you connect computer fan to the "CPU FAN" pins? Check that the CPU has a CPU fan connected properly.

[Endeleon]

Solution... AMD is rife with issues and it's the bargain for a reason, switch to Nvidia.

[Callinon]

That sounds like you're losing a graphics card. The description here sounds like a hardware fault of some kind, not a driver problem. Also you were right to scoff at the "switch to Windows 10" suggestion from AMD. That isn't going to help you. I mean you should do it if you can because 8.1 is a pile and 10 is not, but that's beside the point.

You can check temps on your CPU and GPU with a utility like SpeedFan. If your CPU were overheating the entire computer would shut down. If your GPU is overheating it could manifest itself like this. It's worth checking to see what kind of temps you're getting when the game is running. But honestly, it sounds like you've just got a dying card.

NB: Mmm also, please report computer temperatures in Celsius. It's a far more standard scale even in the US when measuring something like this. Easier to tell at a glance if something's up without having to convert first. As a rule a computer will shut itself down over 100C and anything over 90C should be concerning.

[Zeonx]

Cousin says might be heat issue on your graphics card.

He said if you have adrenalin 18.4.1 get rid of it OC is bad.
He said usually says sometimes the thermal paste is crap on cards that changing it helps also, if nothing else helps time to change it out for nvidia as AMD graphic cards are junk.

Make sure you have proper cooling as HP computers tend to overheat as well windows 8 also outdated platform.

10 gets more support which means hardware support and devices, but if not its probably a issue with overheating.

Noctua NT-H1 Thermal Compound