Intermittent BSODs

[EmptNor]

Ryzen 9 7950x3d
16gbx2 DDR5 6000mhz gskill Z5 neo
XFX 7900XT
Asrock x670e steel legend
Samsung 980 2tb nvme x2

seagate barracuda 4tb hdd
Corsair RM1000x PSU

 

Hello, I’ve been having intermittent BSODs on my windows 11 build since May. I tried running SFC and DISM scans, reinstalling chipset and gpu drivers, running memtest86 and HCI memtest (both found no errors), and updating my bios, but still got BSODs. Next I tried reformatting and reinstalling w11, but I’ve still suffered BSODs.

 

Also, the original build is from December but that had looping BSODs on boot, and fixing it required refunding/replacing both the ram and the mobo. The full rebuild from February ran perfectly fine until my problems started again last month. Oh, and the ram has been running with expo off the whole time.

 

Here are the minidumps from after the reformat along with ones I backed up from before:

https://www.mediafire.com/file/9zbddmjm8frco81/minidumps.zip/file

 

Any help would be immensely appreciated. Thanks.

[Bjoolz]

1 hour ago, EmptNor said:

Oh, and the ram has been running with expo off the whole time.

The dump files show the RAM at 6000MT/s which is the EXPO speed. I checked one from May 3rd and one from June 6th, both showed the same thing. 

 

In general it looks like memory. Memory doesn't have to mean RAM, but it's usually the main suspect. Windows puts low priority data from RAM into the page file and loads it back in when needed so storage can look like memory (And memory can look like storage). The memory controller is in the CPU and if this fails it will just look like memory.

 

It's weird that it stopped crashing for 5 months if it's the same issue. From these dump files, I would say it's RAM or the memory controller (CPU). Try it at 5200MT/s or lower as 5200MT/s is the highest speed AMD will officially support. If it still crashes, try it with one stick at a time. I don't trust memory testers with DDR4 and newer, they miss bad RAM too often. 

 

[EmptNor]

Huh? But I've specifically made a point to never touch the expo settings in bios ever since I got the new mobo. The reason being that I replaced an MSI board for an asrock one that didn't include my ram on its qvl list.

 

If memory testers are no good, is there anything that is reliable to force an error from faulty memory/cpu? If I do get crashes from both sticks individually, should I immediately conclude a faulty cpu, or is there anything else to consider?

 

Thanks a lot.

[Bjoolz]

14 minutes ago, EmptNor said:

If memory testers are no good, is there anything that is reliable to force an error from faulty memory/cpu? If I do get crashes from both sticks individually, should I immediately conclude a faulty cpu, or is there anything else to consider?

Prime95 is usually one of the better ways because it chews through such large amounts of memory, but if it's an electrical problem with the RAM it could be basically random when it has an error. When I say electrical, I mean something else than the memory part of the RAM. The CPU would be my main suspect if it crashes with either stick, but you can't really rule out the motherboard or PSU. It never really looks like the motherboard, it just looks like the component connected to the bad part of the motherboard. Less common than the CPU, but both are pretty rare in general. The PSU is also a candidate, but it would have to hit that sweet spot of unstable enough where it causes components to give errors, but not cause shutdowns. Also rare.