I'm glad I found this post because I've been getting this issue right now on my PC and only happens on certain games. I use the DDU method in safe mode tried multiple amd radeon software versions same issue. I thought it was my power supply as I had a 850w platinum psu that's now in my son's build working flawlessly with his 6800xt and I bought a new 1000w Titanium psu and still having random reboots when gaming and is super annoying. I'm thinking of buying a 1200w PSU, do you think that will fix the issue?
@N8iveWarMachine check out the post from Matt. You've been on 2 PSU's I think it's safe to say you've ruled out a PSU issue, an 850W would be fine. Time to rule out other components, could also be a software issue.
However many have been replacing their CPU and fixed it. I ruled out everything except CPU, MOBO, RAM and SSD. I replaced my CPU several days ago and no issues since and something about the way it runs and loads the cores differently gives me a gut feeling it's fixed, however far too early to know for sure.
Rule out what you can, make sure you update everything BIOS, chipset drivers and all that stuff, maybe try a fresh install. Once you've narrowed it down the best you can, you're left with some parts that potentially need replacing.
While my small local PC store was testing my CPU another person brought in another CPU with the same issue. This doesn't mean in your case it's definitely the CPU but it's something to look closely at and rule out.
Make sure the power supply is safe. I had the problem on the 1000 Watt EVGA G3 Power and when I returned to the EVGA G2 Power. 1000 watts The problem is gone, as well as make sure all the power connections are in their proper place. If you have a friend, try his psu.
Okay, sorry for the late reply. After having my gpu in California for the last month getting replaced I finally have a working PC. I ran my PC with furmark for 10 minutes, which it crashed within 2 minutes. I can't say it's an easy fix as I got a nitro+ 6900xt vs my reference 6900xt, bought a new motherboard (because my 16x slot was broken), and updated my bios and installed a fresh version of windows. After all that my computer now works. I don't know if it was the bios update (a new one came out within the month I didn't have a working computer), the drive wipe, or the new motherboard that fixed the issue, but the issue is now fixed on my system.
Small update on my issue (because getting detail on it has not been easy) and that is that there is "kind of" no update.
Basically, I sent the PC (with my case AND 1000W PSU - which weren't part of the Scan build) but not my drives to the company that made the build, and they have not been able to replicate the issue that occurred literally every time I ran the Time Spy benchmark.
In my gut I believe that the incident that (for those genuinely getting it) is happening to everyone isn't actually individually related to the Graphics Card ... and maybe not the CPU, either ... but more to how the systems themselves are handling sudden power requests through the MB / BIOS.
The BIOS updates that we've been seeing keep speaking about USB problems, and there's ZERO mention from AMD in any communications anywhere acknowledging that the issue that is being seen here (and I believe across multiple, differently titled, threads here) is related to either;
However ... I wonder if any of us have attempted to check USB in any of this? I think (for my part) I know that I'm pretty sure I've had any and all USB 3 hubs plugged in each time ... which ... *maybe* it's a thing? I dunno.
Either way ... I'm asking if I can have a video call with the engineers at my retailer.
@eliotcole the closest I could find that they're aware of the issue is https://www.guru3d.com/news-story/amd-agesa-combo-pi-v2-1-2-is-coming-for-msi-500-series-and-400-ser... they talk about power management changes and trying to fix WHEA errors with BIOS updates.
I suspect it's to do with load line calibration or balances and with infinity fabric and ram over 3200Mhz it just becomes too unstable. Also adding more PCIe gen4 items I suspect makes the stability worse. This coupled with a CPU that is just barely stable and you'll have issues.
This kind of issue has been around since Ryzen 3000, however not everyone will have the same cause, lots of things can cause random restarts. I'd research Ryzen random restarts and if your issue seems similar I'd rule out your CPU.
I've been running my new 5800X for 10 days straight now, Zero issues. I could always reproduce the issue in less that 6 or 7 days before, the only thing I've changed is the CPU, nothing else.
Cheers, man ... I mean, what *is* a positive response from the 'no real update' was that they indicated no issues with the PSU.
We all suspected there wouldn't be, but I dearly didn't need to have that be an issue on top of this issue. Although I hold my hands up to not having tested using purely modular cables on the GPU (not possible for the MB+CPU), I will admit to my brain being a bit on the wrong side of things there.
Still, I wasn't a fan that they literally didn't test the thing that has (without fail in my testing) caused the incident. I respect that they have their process, but they said that they couldn't do a video call with the testing (like, why?) and they hadn't tested the reproducable events that I could do in a second to prove the incident. I still stand by Scan as an excellent company, but this particular experience has been sub-par in general (with some great service from certain staff, though).
I've asked them to check how many ranks the ram has, whilst it's in there (as I can't get the exact SKU and details for some reason), because thanks to Linus it's another thing I don't wish to be losing performance from.
I also respect that from AMD's perspective, a whole new way of managing how memory/disks/GPU memory is handled in game is going to have some way to go before it beds in nicely.
Solved for me