Last few weeks, every so often i get this crash screen while playing (it always looks like that). Sometimes it's a hard crash and other times it just crashes the driver and after i can continue playing for 6 hours with no issues. It usually only happens when i hit Espace and go the main menu or when i bring up an inventory screen, map,... and never in the middle of gameplay. As a matter of fact it most often happens while i'm closing the game. I update my drivers every time the new ones come out and i uninstall the old ones using DDU in Safe Mode. Has anyone else experienced this before and was it a hardware failure or a software issue? I did the memory tests and it doesn't seem to be a RAM problem. Here's a small giveaway as a thank you for listening :)

The GPU in question is MSi RX480 8GB
current driver Radeon Software Version
18.10.2

6 years ago*

Comment has been collapsed.

Does this happen with every game you play?
A specific game or only high-end games?

6 years ago
Permalink

Comment has been collapsed.

just the demanding games it seems, most recently today while i was playing Hellblade. Yesterday it happened in DmC Devil May Cry and Steam error said it could be hardware failure or overheating but msi Afterburner if it's to be trusted says the GPU temperature never went above 75°. The GPU in question is MSi RX480 8GB btw. since i forgot to mention it in the OP

6 years ago
Permalink

Comment has been collapsed.

Simple, feel it, can you keep your hand on it or not? If it's the last then it could be a problem.

6 years ago
Permalink

Comment has been collapsed.

Artificating like that is due to video memory corruption. In most cases, you can just clock the memory down and it'll be fine with minimal impact on performance. Try clocking the memory down 100. If it doesn't go away, try to RMA it if it's still under warranty (MSI is three years, so it should still be).

On a side note, 480s like to stay under 70, so I'd also recommend increasing airflow as well.

6 years ago
Permalink

Comment has been collapsed.

Include the name of the GPU for more research, bump.

6 years ago
Permalink

Comment has been collapsed.

If you got access to a replacement card then switch and try that.

I wouldn't even rule out a bad motherboard.

Macbook but sounds like something similar pointing to either graphics or motherboard.
https://discussions.apple.com/thread/2121464

6 years ago*
Permalink

Comment has been collapsed.

My experience with GPU problems is this: You only get 100% certainty if you test the GPU in another rig. If there is no problem in the other rig, your GPU is not broken and you have to look elsewhere to find the problem. If the problem shows in the other rig, your GPU is going to Valhalla.

Most of the time strange artifacts on screen are an indicator of broken VRAM on the GPU, but this behaviour should be more prominent the more VRAM is used, meaning in 3D applications such as games. I once had a broken GPU i still could use for approx. 8 weeks in 2D, only browsing, mailing, such things.

6 years ago
Permalink

Comment has been collapsed.

We need more info. Add it to the OP and then let people know when you've updated it.

What GPU do you have?
What drivers do you have?
Have you tried re-installing drivers already?
What games does it apply for?
Have you tried starting your PC on safe mode?
What temperatures is your GPU running at?
Have you cleaned your PC?
Have you tried switching your GPU for the same model or for any other model?

6 years ago
Permalink

Comment has been collapsed.

GPU is RX 480 8GB
drivers are the current ones Radeon Software Version 18.10.2 and i've gone through all the latests ones since it started happening one by one.
It mostly happens in demanding games, most recently Hellblade and The Witcher 3 and when it happens it happens very early in the play session, like the first time i pull up a map or go the main menu and after that if it's not a hard crash but only the driver crashes i can relaunch the game and the driver and usually play for hours and hours with no issues. MSi Afterburner and FurMark don't show it going above 75°
PC is regularly cleaned and i don't have a spare GPU to test

6 years ago
Permalink

Comment has been collapsed.

You could try stress testing with FurMark and if your GPU doesn't crap itself on it, it's probably not a hardware issue.

6 years ago
Permalink

Comment has been collapsed.

FurMark has been runing for around 20 minutes now and it seems to be holding it's own

6 years ago
Permalink

Comment has been collapsed.

I was just searching and found this person that has the exact same artifacting. I didn't read the entire thread, but he has an RX480 and at least one other person in the comments also had the same issue with an RX480. You may want to read and see if they mention what happened or if they found out what the issue was.

https://www.reddit.com/r/pcmasterrace/comments/7tjvbn/give_it_to_me_straight_has_my_gpu_died_at_the/

More links:
https://answers.microsoft.com/en-us/windows/forum/windows_7-gaming/problem-with-my-rx480/0dbde62d-3eb6-430c-9029-763cac18f5ed
https://forum-en.msi.com/index.php?topic=285787.0
https://www.reddit.com/r/pathofexile/comments/4y4k8d/did_path_of_exile_kill_my_graphics_card/
https://www.reddit.com/r/techsupport/comments/629ft0/gpu_checkerboard_artifacts_rx480/
https://www.youtube.com/watch?v=q0XzUwVIbnE

Edit: I have a GTX 580 (8 years old) and my drivers would sometimes crash in certain games. Lowering the clocks a tiny bit or increasing the voltage would fix it in those games and everything was fine. Then like 6 months later, it started crashing the driver and lowering the clocks didn't help, even well below factory settings. Then is started crashing the entire computer and was artifacting even in the bios and desktop. That all happened in about 20 minutes and I considered the card dead and took it out. Then a few months later I heated the GPU die and brought it back to life because my issue was failed solder under the die. It is still working and what I am gaming on now. Link to that thread is here.

6 years ago*
Permalink

Comment has been collapsed.

I got computer knowledge but with the dozens of options, i wouldn't even have a clue with all the overclocking, and i think many would turn it up too high and thus are more quickly to break their stuff. I am fine with just stock.

Good find though, looks very similar.

6 years ago
Permalink

Comment has been collapsed.

I'll try lowering the clocks a little bit, mine is the Gaming X version so i think it came a bit OC'd out of the box, maybe that'll help and then go down the list if it doesn't. thank you for the links

6 years ago
Permalink

Comment has been collapsed.

I added some more links in my comment above. I didn't read through them, but I did see one of the first comments where someone said that it is a VRAM issue on the RX480, so core clocks and voltage probably won't help. You can try lowering the memory clock, but I'm not sure if that will fix it either. Maybe read through those threads or do a search for RX480 VRAM artifacting and see if you can find more info. Or search for RX480 checkerboard artifact.

6 years ago*
Permalink

Comment has been collapsed.

It might be worth trying out older drivers just in case it's not a hardware issue and that it's a buggy driver. Try to rule out any software related issues and possible conflicts. Also check your Event Viewer. It can provide you additional info.

6 years ago
Permalink

Comment has been collapsed.

You could try to rollback to a old driver version, maybe it's driver issue.

6 years ago
Permalink

Comment has been collapsed.

My diyng R9 270X looked like this... nearly 1:1
It was the vram...

6 years ago
Permalink

Comment has been collapsed.

First of all, updating your drivers itself might be the cause. For a real example from myself, my GTX 1050 starts crashing (very similar to your screen, just not patterned like yours) if I use a newer driver, have to send in a support ticket but lazy. Use DDU and install an older driver model, start from the oldest driver, if you're unsure (Of course do check if the said older driver still supports your GPU).

First and half, MSI Afterburner (driver) is known to cause blue screens, I wouldn't recommend installing/using it too often, do be wary.

Now for the worse case, your card is dying and can't work at it's optimal frequency anymore, so underclocking until you can replace it, I guess.

6 years ago
Permalink

Comment has been collapsed.

Obvious solution but just in case; make sure your screen wires are plugged in properly, and that your GPU is seated properly (remove it and put it back in). Probably a deeper problem than that but if you haven't tried this basic troubleshooting yet, now would be a good time.

6 years ago
Permalink

Comment has been collapsed.

Try underclocking your GPU a bit and see if it still occurs, I had something similar with my old one.

6 years ago
Permalink

Comment has been collapsed.

My step sons R9 380 had a simillar issue, I took it apart to see the factory applied thermal past was mostly on the pcb, not the chip. I applied some new paste and there hasnt been a problem since. That was 2 years ago

6 years ago
Permalink

Comment has been collapsed.

Every suggestion I would give you has already been mentioned so just have a bump if someone else has any better ideas.

6 years ago
Permalink

Comment has been collapsed.

maybe it does need more power ( weak psu or low power delivery )?!!! :/ or needs a cleaning from dust or maybe a driver problem , do a clean install and try or maybe its from the monitor , did you try a deferent monitor ?!!

6 years ago*
Permalink

Comment has been collapsed.

After air-dusting your card and card fans, grab some fresh thermal paste and replace the old paste on your GPU die after wiping it clean with isopropyl alcohol. Also check to ensure that the thermal pads (such as on your VRAM) are placed properly on their respective zones. If they've slid or ripped about, you'll want to readjust or replace them. Doing all of these steps will allow for better cooling (and thus less chance of thermal throttling and artifacting). I would expect that most errors after that are due to game bugs and/or a particularly bad driver, if not PSU problems (which would be more evident in regular use). You don't need to be underclocking, rolling back to very old drivers, or doing anything else like that. Just maintain your parts.

6 years ago
Permalink

Comment has been collapsed.

It's an odd one for sure. Some good suggestions, but it is strange that it seems tied in with pressing esc. Crazy as it sounds, try a different keyboard as a test. Had an issue once where pressing a certain key (also think it was esc) would very occasionally flip the screen 180deg. Also a Windows reset / reinstall might fix it up.

6 years ago
Permalink

Comment has been collapsed.

Deleted

This comment was deleted 2 years ago.

6 years ago
Permalink

Comment has been collapsed.

could be your ram. i had a issue where my pc would crash when exiting killing floor 2 (it loads alot of stuff into your ram) and my one ram chip was cutting out on me. my ram would work fine but after a crash it would tell me i only had half my ram but on reboots it would show up as normal.

6 years ago*
Permalink

Comment has been collapsed.

If it helps: When my last graphics card started dying, I just got black screen for a second or two, about twice a month, then it would go back to normal. In later stages, I started getting small, strange artifacting across my desktop, which progressed to some multicolored vertical lines. Finally, graphics became really bad, and strange ASCII characters started appearing in place of normal alphabetical characters, vertical lines took up a lot of my screen, I got lots of crashes, and then that's when I retired that card. I switched to onboard graphics while waiting for replacement card, and never had an issue with crashes.

As an aside, you might want to try running Anti-virus / Anti-malware to see if you find anything else that could be causing the problems (in case it might be a virus or crypto-miner using your GPU). Also, if you buy a new Graphics card from a store with a good return policy, you can check to see whether the new card fixes your problems and return it if the problems persist.

6 years ago*
Permalink

Comment has been collapsed.

No idea but have a bump.

6 years ago
Permalink

Comment has been collapsed.

Ded already

6 years ago
Permalink

Comment has been collapsed.

Sounds like it's dying, yes. Especially if you can rule out that it's the RAM, drivers or mainboard.
My old GPU started like this. Crashes of the driver and game. Then it had some rare display bugs on hardware-hungry games.
I kept using it until one day green blocks and lines filled the screen. Even while booting in DOS.
Bought a new card and everything works like a charm.
Using your card until it's on the brink of death is scary!

6 years ago
Permalink

Comment has been collapsed.

Will it be such a situation over time?
If that is the case, we also doubt the thermal runaway and poor power supply.
However, as long as you see the screenshot of the screen, doubt Memory corruption more.
It is difficult as it states that it has been tested.

that? by the way··.
Did you update the latest drivers?
However, if the period from the previous driver to the last updated driver is longer than one month, an error will occur if you do not perform a clean installation.
If time and environment are affordable, check whether the same phenomenon occurs even when connecting to another PC.
I am worried not to damage other equipment when the graphic board is damaged.

6 years ago
Permalink

Comment has been collapsed.

Sign in through Steam to add a comment.