I've been experiencing some serious issues with my Radeon RX 5700 XT Mech OC From MSI. Currently, it is in a state when under full gaming loads, will do a plethora of different things to crash in ways I never thought possible. Here are some of them.
- Monitor goes black, turns back on with all applications and background being completely black with white boarders surrounding them, seizing the system up shortly after. Full shutoff and reboot required to get out of.
- Applications going black and then returning to be in abnormal states, such as all applications being blacked out and unable to return without a PC restart.
- Applications go black for a second and returning to normal, then Whole PC crash and lockup shortly after.
- Monitor goes black, whatever sound was playing repeats at a fast frequency for half a second, with the PC restarting shortly after. No error shows up when windows recovers.
- I once was able to recover the system from one of the black application hangs, with finding that AMD stated there was a driver crash. I hope this can mean something for this whole problem.
For troubleshooting, I know for a fact hardware wise that is is indeed the graphics card being the culprit, as I have tried with the system 2 different cpu's (r5 2600, r7 3700x), motherboards (b450-f gaming ASUS, x570 Aorus pro Wi-Fi), power supplies (evga 650w b3 bronze, seasonic focus PX 850w platinum, and ram dims (XMP on and off for both kits, single and dual channel, single and dual stick for each 2x8gb kit).
For the drivers end, I've gone through many different versions, from I believe version 19.7 and above with the issue still persisting. No mater how many time I uninstall or reinstall, the issue persists.
Also, for temperatures, GPU core peaks at 67C, Memory peaks at around 90c, and Hot spot reads at about 113C, unsure if this is an issue, as I've sent the card back with this remark and got nothing fixed for it.
If there is any more Information you need to help troubleshoot the problem, or any recommendations to relieve these problems, I would be totally open to take them.
Sorry, I know this thread isn't brand new but I'm commenting to put my input in.
What you're having sounds identical to my problem (I too have a 5700 xt with almost identical symptoms to you,) and I'm commenting to keep this thread in mind if you do get a logical response.
At one point I had crashes recover and got the "Display driver has stopped responding error."
After contacting AMD support, they provided me with this link (https://support.microsoft.com/en-us/help/2665946/display-driver-stopped-responding-and-has-recovered...) which helped to correct my issue for about a week, but the issue has started to happen again. (For me, I used Method 3: Modify registry entry to increase GPU processing time.) Although this issue still occurs, I have found that instead of crashing EVERY TIME, my system can recover more frequently although it takes time.
I encourage you to try this also to see if it helps you at all, at least for a little while. If this solution works for you (I don't have high hopes,) or if you get another solution, please message or reply to this as I still have found no permanent fix.
@TheRandomMudkip the temps of your card might be a problem.
Hotspot over 90°C already causes thermal throttling and therefor performance loss.
I had blackscreen and crash issues with my RX 5700 XT as well, when it was brand new.
This solution worked for me and fixed all my crashes: [crash fix] RX 5700 XT driver 20.8.3
I am using driver 20.9.1 myself and did not have any crashes, after using that solution.
My temps are around 60-65°C under load for GPU, Memory and VRM.
The junction temp (hot spot) is around 80-85°C and sometimes spikes to 87°C for a short time.
I am using a manual fan curve and allow my fans to reach 100% if needed.
I have seen posts from other people, that also reported problems with blackscreens and too high temps.
Found the problem! BIOS was WAYYYYYYY out of date! Updating it fixed all my problems. It's a really good troubleshooting step that I overlooked. I'd recommend this to others who are in need of help.
UPDATE: Bios was not the problem. Issue has returned. This issue seems to go away with a major software/hardware change, and slowly become more and more rampant and extreme, leading to the point in which the system is unusable. Any advice is much appreciated on how this could even occur. I have tried old drivers, different CPU's, motherboards, ram kits, storage drives, power supplies, even different cases, and the problem persists after even a windows reinstall. I'm completely lost.
You have already tried a lot. It is difficult to point you in the right direction, if all your steps did not work.
Did you check Windows Event Logs and Reliability Monitor to see what kind of errors are causing the crashes?
Does your GPU still have warranty? Maybe think about a replacement from your vendor, if nothing else works.
I have the same issue on Linux, after major driver changes it works for a few days, and then it just comes back. Got all the same symptoms, I noticed that especially once it reaches 90°C it tends to crash. At this point my only option is selling it and getting a new one, or just suffering and rebooting my pc sometimes up to 5 times per hour