I've been using my AMD 9800X3D and G.Skill Royal 48GB RAM for about a year without issues. Recently, I replaced my MSI Godlike motherboard with a Godlike X, and soon after, I started experiencing random freezing and reboots—primarily when playing FFXIV or even just idling in Windows with Brave open. Oddly, there are no BSODs or error codes, and event viewer only logs unscheduled reboots. The problem usually appears after about 48 hours of use, making it seem like I had solved it initially. I've replaced nearly everything: old GPU, PSU, SSD, and even reset to a clean Windows 11 install. I tested the memory with memtest86, checking both sticks individually—both passed.
The only parts I haven't changed are the CPU and the RAM. I've noted some CPU usage and temperature spikes when launching FFXIV, but temps otherwise seem normal. I found some discussions about similar issues with the 9800X3D leading to a possible conclusion that the CPU might be unstable. If swapping out the CPU doesn't fix the problem, I'm at a loss. Any advice would be greatly appreciated, and I hope this helps anyone facing similar issues in the future!
4 Answers
Your RAM could be overheating, especially when your GPU runs hot during gaming. There’s some good info from Wendell at Level1Techs on DDR5 thermal issues that might be worth checking out. Maybe your RAM fan isn't enough?
The lack of error codes in your logs is frustrating! I had no BSOD either, just random reboots. It might be worth checking TPM issues; I found that some updates created conflicts, especially in AMD systems. Here’s a link discussing a similar problem (might be worth a look).
I still need to create a USB with an older Windows version. I keep wondering if the recent updates might be a factor in all this.
I had a similar issue with a less powerful setup, and it turned out that reseating all the cables in my system made a huge difference. It would reset suddenly, just like yours, with no helpful logs. What helped was taking everything apart and putting it back together—sometimes the slightest connection issue can cause chaos.
I thought the PSU might be the culprit in my case because of how quickly it would reboot. When I replaced it, the new unit was defective! It overheated and smelled burnt. Did you notice any changes to your OS installations after the issues went away?
Have you tried lowering the RAM frequency to around 6000? Even if memtest shows both sticks are fine, it might be worthwhile to remove one stick and see if that helps. Also, check your Windows application logs for any errors that might give a clue.
I’ll check the logs tonight after work. Before I wiped everything, I noticed some warnings about an unused Ethernet port, which I disabled in the BIOS, but the issue is still there. Right now, I'm only using one stick of RAM to troubleshoot.
I need to take another look at that. I had a fan running for the RAM, but temperatures during memtest were reasonable. CPU seems suspiciously high sometimes, though.