Are the BSoDs causing instability?
Are the BSoDs causing instability?
Hey everyone.
My i7 9700K has been running smoothly at 5.0GHz for the past eight months, but a week ago I suddenly encountered BSoD errors.
Over the past week, I've had one BSoD each day, with messages like "IRQL_NOT_LESS_OR_EQUAL".
Today I noticed another type of BSoD, this time showing "KERNEL_SECURITY_CHECK_FAILURE".
I considered lowering my CPU speed to 4.9GHz to see if that would resolve the problem, but I wanted your thoughts—do you think this could be related to my overclocking and might have caused instability?
Looking forward to hearing from you all.
Best regards
Other factors can cause instability in an overclock, such as Windows updates, microcode updates, BIOS updates, and an initially unstable system. High vcore or IMC voltage may lead to CPU wear. In your situation, memory issues seem likely—inspect the IMC and/or RAM voltages. It's interesting how boosting the vcore can resolve the problem. Comprehensive checks are recommended: verify RAM, VRM, and CPU cooling. Dust buildup around the VRMs and CPU heatsink can block airflow and trigger overheating. Updating Windows 10 to the latest version and removing faulty drivers might also help. Consider using a tool like https://www.resplendence.com/whocrashed to identify crashes. A troubleshooting guide is available for resolving kernel security check failures in Windows 10.
Thanks for your prompt reply. I saved the file and performed an analysis, which shows my two previous crashes:
https://gyazo.com/72142b8190d211dc978f58acc6ab8a9d
Although it doesn’t provide detailed information about the exact cause.
I understand from WhoCrashed that it’s a driver-related issue, but it doesn’t specify which one.
😀
However, during an OC of my CPU, whenever stability issues occurred, the BSoD message appeared as irql_not_less_or_equal – maybe I misremembered.
Regarding details, here are my current OC settings:
5.0GHz (with adaptive voltages)
Multiplier: 50
XMP: Enabled
AVX: 0
MCE: Auto
SVID Support: Auto
SVID Behavior: Best Case Scenario
AC/DC Load Line: 0.01 / 0.01
Load Line Calibration: Level 5
Adaptive Voltage: 1.350v (1.332v @ load)
Voltage Offset: -0.018v
VCCIO: 1.150v
VCCSA: 1.150v
ADD: I lowered the frequency to 4.9GHz just to check if it resolves the problem; everything else remains unchanged.
EDIT: My Windows auto-updates were active, but I verified myself to ensure all updates were current. A new update was found – ".NET Framework 3.5 and 4.8 to Windows 10 Version 1903 for x64". Additionally, I haven’t updated my GPU driver in a while. The latest driver is from 1st October, while the newest one is from 3rd February.
I’m not sure if this is related to my issue, but I wanted to share it.
If down-clocking resolves the problem, it suggests the core issue lies in overclocking. Which program are you employing to verify stability? A 5GHz AVX overclock works well for the 9700K, ranking among the top 9% of CPUs.
Coffee Lake Refresh
SSE Frequency
AVX2 Frequency
Vcore
% Capable
9700K
4.90GHz
4.70GHz
1.337V
100%
9700K
5.00GHz
4.80GHz
1.350V
Top 90%
9700K
5.10GHz
4.90GHz
1.362V
Top 38%
9700K
5.20GHz
5.00GHz
1.375V
Top 9%
Top 1%
source
Each driver version contains its own set of bugs; just because it's newer doesn't guarantee stability if it causes crashes.
Yeah I think you're right. But fixing it is really tough because I'm only getting one BSoD each day, and even today the message was different from the ones before.
I tried Realbench, OCCT, Prime95 and IBT.
My OC has stayed stable for 8 months until these problems started a week ago.
I've seen no WHEA errors, no BSoDs, no throttling issues.
I guess it might just have become unstable after eight months, I don't know if that's true.
Also checked the chart, 1.375v seems too high for a 5.0GHz processor.
My prediction seems to be about values where the most CPUs can hit that frequency. It’s possible you’re just one small voltage change away from stable at 5GHz, I’m not sure. Don’t overlook BSoD. If you verify your overclock stability, you’ll see if it’s due to the overclock itself or something else. It took me 2-3 years for my 2500k to degrade, but I got it from a binned chip. Most overclocking efforts needed adjustments over time.
The biggest issue with overclocking is that BSoD consistently pushes me to run stability tests. My old audio card triggered BSoD, and it took some time to figure out the cause. Leaving an unstable system only leads to data corruption.
I understand.
Of course I don't want to overlook the BSoD, which is why I reduced my OC settings.
Now I just have to wait and observe if it happens again. But definitely, if I experience another BSoD today, it must be something other than the CPU OC causing the problem—especially since I'm now running at 4.9 GHz with a higher vcore than necessary. Voltage shouldn't be the issue.
It could also be related to RAM.