Question Blue Screens and crashes with 4 DIMMs in, but not 3 or less

Dec 20, 2020
5
0
10
Hello. I built my computer myself:

CPU: Ryzen 7 2700X
GPU: Gigabyte RTX 3080 Vision OC
MB: Gigabyte GA-AX370 Gaming
RAM: 32 GB (4x8GB) Corsair Vengeance Pro RGB 3200MHz
Storage: Samsung EVO 860 1TB SSD, Seagate Firecuda 1TB SSHD Hybrid
PSU: Gigabyte P GM 850W

I am having crashing issues. This includes bluescreens and game crashes (several different games). It seems to happen at random; I have been monitoring task manager and there is no spike in utilization of anything before a crash. The PC sometimes will even bluescreen when it is simply idling on the desktop. Bluescreens seemed to happen every 1-3 hours or so. So far, I have seen 6 different blue screen codes. They are:

SYSTEM_SERVICE_EXCEPTION
DCP_WATCHDOG_VIOLATION
KERNAL_SECURITY_CHECK_FAILURE
0xc000021a
DRIVER VERIFER DETECTED VIOLATION
ATTEMPTED_WRITE_TO_READONLY_MEMORY

The crashes first started about a week after I began overclocking. My overclock was nothing serious: my CPU from 3.7 GHz to 4.0, and the RAM from 2133 MHz to 3200 MHz (essentially XMP). The voltages were set to auto so there shouldn't have been any problems there. Upon searching for a solution online, I was led to checking all my drivers. I ran the windows driver check tool that checks the status of all system drivers upon restart. Well, when I ran the tool and restarted, my computer refused to boot ever again no matter what I tried. So I began to think maybe I had either somehow corrupted my windows files or my SSD was going bad. I checked the drive using command prompt and turned up no errors. I then proceeded to wipe the drive and install a fresh copy of windows 10. My PC booted like a brand new computer and I redownloaded all of my stuff. I thought I was in the clear, but then I crashed again about an hour later. At this point I was fuming and decided it must be my RAM. I ran the windows memory diagnostic tool and it turned up a hardware error. Just to be safe, I ran the memory diagnostic tool 4 more times, each time with 3 sticks out of the 4 to test which stick(s) was the bad one. Just to my luck, each of those tests with 3 sticks in came back with no errors. Also, let it be known that I never crash at all or have any problems when the 4 sticks were running at 2133 MHz, nor did I have any problems or crashes when I had 3 or less sticks in while running at 3200 MHz. The only time I crash is when I have all 4 sticks running at 3200 MHz. I need my memory to be running at 3200 MHz because if it isn't, the RTX 3080 % utilization refuses to go above 30% and I get extremely poor gaming performance. I would like to have all 4 sticks in mainly for the aesthetics, but also why not have 32 GB instead of 24. So at this point I determined that each of the RAM sticks are working just fine on their own, paired with another, or two others, but when all 4 are in and running at 3200MHz, the PC crashes games and bluescreens consistently. At this point, and given this information and everything I have tried, I have to believe that the problem lies either in the hardware of slot #4 on my motherboard, or there is some motherboard or windows software or setting that I am missing that is causing problems with 32GB RAM at 3200MHz. I have not been able to find anyone else on any forums on the internet who have had the same problem as me. Can anyone confirm my suspicious about my motherboard? I'd rather not go and buy a new MB and do all that work to replace it just to find that the problem persists. Thanks.
 
Dec 20, 2020
5
0
10
Looks alike all could be an unstable memory OC.
What voltage are you running on those?
You might have to increase the volts or drop the speeds until you find stability.
The voltage is currently set to auto. I would hope that means that the MB automatically adjusts the voltage accordingly based on how many DIMMs are in and what speed I told them to run at. I have tested different voltages before, and none of them solved the problem. I am not 100% sure, but I think I remember even trying setting the voltage as high as 1.4V, which (correct me if I am wrong) is the max safe voltage for most DDR4 RAM. I will try this again when I am home though. I get back tomorrow night.