No bluescreen, no freeze, just instant restart.
Would get one every couple weeks since I built the PC a few months ago. Got several last week on the same day, replaced PSU and rerouted cables, and it ran perfectly since... until this morning, when they're coming randomly.
Would suspect the PSU, but it ran through a windows install and setting up my old PC (handing that down to my mom) and the last few days in this one just fine.
Not repeatable by program or gaming, that I can find. No telling if they're going to be five minutes after startup or three hours. So far the only load has been Firefox. Idles at 35-48 degrees.
No minidumps from them.
AMD Ryzen 3600
ASUS TUF Gaming x570 Plus (wi-fi)
MSI GeForce GTX 1650 Super
64gb (4x16gb) Team T-force Vulcan Z DDR4 3000 memory
Intel 665 m.2 1tb SSD, 1TB WD Blue SATA SSD, 4TB WD Blue SATA HDD
(now) Corsair RM850x PSU
Samsung DVD/CD read-write drive, if that matters.
SO FRUSTRATED. Any help, please? Dump from WHEA: https://drive.google.com/file/d/11lmC6fsNTQM7YrS9gynA6MDMptm__N7R/view?usp=sharing
No known changes aside from the PSU and rearranging (which didn't cause immediate problems) until they started.
Troubleshooting so far: Deleted unnecessary programs. Updated drivers. Disabled what I could. Routed SATA cables to relieve stresses. Reseated RAM, reseated CPU and reapplied thermal paste. Updated BIOS and reset to factory. Reset CMOS.
Those are the same errors I get and I was told that it's the GPU! I have a Ryzen 5 3600X, AsRock X570 Phantom Gaming 4S, Gigabyte Windforce RX 5600 XT Gaming OC, 16GB TeamForce Xtreeme Gaming DDR4 3733 RAM, topped off with a Corsair H110i GTX V2. Now I have to wonder is this a CPU or chipset issue... Those events are "processor failure" and "processor hierarchy error". All the searching comes up as a GPU driver issue. You have an Nvidia. So what's going on? I even saw a few posts elsewhere with Intel/Nvidia users getting this error, possibly a Windows 10 thing?? I have Windows 10 Home x64 version 2004, but these complaints go back to build 1903. I've done all the same troubleshooting, reseating stuff you did to no avail. Card heat could be an issue still yet to be addressed further. Imagine if all this crap is caused by a Windows bug! It'll take years for them to figure it out!
The 18s seem to have calmed quite a bit with some OS repair.
Think it's in large part drivers not playing well together or some corruption. Possibly exacerbated by hardware things.
Next step: a clean install of Windows.
Unfortunately that means that, yes, if it's a compatibility issue there may be little I can do about it, and it's just waiting and hoping for relevant updates.
did you install a Zen3 compatible BIOS version on your motherboard. especially with AGESA 22.214.171.124? Those ID 19 / hard reboot loops are a common "feature" of this AGESA version. A had the same problem, I downgraded to a non Zen3 BIOS, now my system is stable again for over 3 weeks.
Even the AGESA 126.96.36.199 Patch C BIOS versions seems to have some issues. Maybe AMD should have finalized their AGESA ahead of releasing Zen3.
Not that I'm aware of.
When I built the PC, I updated the BIOS to 1407 (for this particular board) and did the normal run of driver updates. It had crashing issues until an update or two later, but I think mostly because the stock cooler and thermal paste wouldn't drop the idle temps down to under 55 at best, and the Ryzen's normal temperature spikes would trigger the failsafes.
After an aftermarket cooler (and, by timing, a couple Windows updates) it was pretty near rock solid until these issues. At least on the 18s/reboots. The 19s (corrected hardware issues) show up plenty in the event log, but don't seem to cause much issue.
Not that they aren't, just that they aren't triggering reboots themselves.
The only BIOS update since then was during these issues. Not sure if it helped or not, but I can only assume that 2802 had a round of minor fixes for various stuff anyway. At least if gives me the option for the new Ryzens; 1407 was pre-release.
The only thing that seems to have seriously helped the reboots is the OS repair, but the batch of new drivers and updates may have aided it.
Also switched back to the old PSU just to take out one variable. Will switch back to the new one on another day off to test.
Update: fresh Windows 10 install.
No more Event 19s (corrected hardware errors), and WHEA 18 isn't being written, just Event 41, Kernel-Power after the (maybe slightly rarer) restarts.
Across two PSUs. Memtest reported 0 errors, haven't found anything with OCCT...
I have no idea. At this point I'm almost certain it's hardware. The CPU or motherboard.
Just wait, those errors will come back, maybe not the crashes. Not to be the "harbinger" of bad news, but I've done the OS reinstall deal and it stays good for a little, then it starts tossing the same codes. Code 41 always is there. It acts like it woke from sleep or s3, when I have mine set to "high performance", no sleep, never turn off display, etc. One thing to make sure of is to set the SSD to never turn off. Spin drives can power down, SSD's never, ever. That will lead to corruption. Windows should do that automatically but the AI isn't there yet. Glad it's working out so far. I highly doubt it's hardware because there's so many of us complaining about virtually the same thing. That would mean we all need new CPU's and mobo's all in less than a year.
Now I used DDU to uninstall my graphics driver and put an older one in and mine has behaved for 2 days so far. So if you get the crash issue, try using DDU and reinstall the graphics driver. A lot less work than doing the whole OS, drivers, games, etc.
I expect the issues will come back. Unexpected cycling does that.
I haven't re-upped the drivers yet. It's a fresh install, and so far is acting the same. I just got rid of some corruption issues, and now it's down to restarting every five hours or so instead of every half hour. And unless it updated things without asking me, they're just the base drivers for now.
I guess I'll do clean updates as a further test and hope.
Just frustrated because it would do it every few hours when I first built it, then it tamed down to only a couple times until this last week. And that didn't coincide with me updating anything (at least not manually) except the PSU, which I switched back for testing. Windows will do what it does, though.
I even checked for anything I could be shorting but didn't see any scratches, don't think passing wires would cause interference enough, so unless the 4+4 extensions lost shielding where I can't see (my case has almost zero room on that edge), there's nothing to short. And I disconnected the extensions just in case, and routed cables in a direct but ugly path.
I'd be less frustrated if Windows would even dump the crash info now, so I could check that.
First off this error is very generic and lots of things can trigger it. From bad cpu, memory issues, power issues, gpu issues, malware, registry corruption. You get the picture.
However as of late there are just a ton of these issues being reported some have replaced their AMD Navi GPU other their Ryzen processors and they say the problem is then fine. It seems to me disproportionate but more often than other cpus it seems I see this complaint with 3600 and next the 3700x both 65 watt parts. I have 2 3600 parts and a 3700 myself, all on MSI Tomahawk boards and no issues. I have not updated to the latest bios and will not at this point unless I upgrade to Zen 3 at a later date. As I have seen to many that complain of this issue being on the latest bios. I should add there are already several new reports of this trickling in on 5xxx chips too. Which also has me thinking, it is bios related.
It sure seems in the cases where it is a Ryzen cpu and green team gpu many report that replacing the CPU has helped. So I am not sure if their is something that with the latest bios has pushed CPUs that may have teetered on that edge of just passing validation over the edge?
I for one am not sold on the idea that all these CPUs are suddenly bad. To many having this issues are people that ran these chips for months before having a problem while others are new builds.
I think it is possible something introduced into the bios in a recent update could be a contributing factor.
Any of you having this issue report it to AMD: https://www.amd.com/en/support/contact-email-form
I would then highly suggest talking to the support department for your motherboard and let them help troubleshoot the issue.
Most likely it is them that can isolate the cause. Maybe even offer an older or unreleased newer bios to help.
Worth also doing is to run OCCT from OCBASE dot COM. It is free diagnostic software you can test your system with to see if it is finding problems with any of the major components.