Internal Parity Whea Logger 19 - Metro Exodus - 9900K 5ghz

[lpiotrowski211]

HI.

My pc:

9900K 5G OC manually overclocked to 5ghz
LLC level 6,all power limit maxed,sync all cores 50,core voltage was 1.315v. But now actually i bumped to 1.320v to see what happens.

2x16GB DDR4 3000MHZ GSKILL

Asus Prime Z390-A

Corsair 850RmX



Pc is rock stable i think. But i am getting sometimes Whea Logger 19 Internal Parity warning only when i am playing Metro Exodus.
It pop up in event log,like once of 9 hours of play. What can i do now ?

Also get this on new bios and older . Sometimes WHEA LOGGER 19 INternal parity will pop up. Like 1 warning of 9 hours in Metro Exodus. Should i worry?

Its event screen:




I tested 8 hours RealBench stress test 32gb ram and no bsods,no wheas. IHC MEMTEST no errors.

Like i said i bumped core voltage to 1.320v or maybe should i bump too a Dram voltage from 1.350v?

[Jane Doe]

Hello, this is your second post about similar problem. Someone said that 8h of RealBench is enough. They said that new prime95 with small axv can hot your cpu about 115C... bcs of that they advice you to use RealBench. I will note you this and you can decide whay you want to do about your test stability but can tell you that you are not stable and possible that you need 1.37v or more for full stability. This is note from other site:
"Here’s a sample of utilities grouped for thermal testing and stability testing, shown according to % of TDP at Default BIOS settings:

TDP … Thermal Test - Steady Workload

129% … Prime95 v27.7 through v29.4 - Small FFT’s (AVX, No Offset)
101% <-- Prime95 v26.6 - Small FFT’s
89% … HeavyLoad v3.4.0.234 - Stress CPU
87% … FurMark v1.19.1.0 - CPU Burner
78% … CPU-Z v1.82.0 - Bench - Stress CPU
66% … AIDA64 v5.95.4500 - System Stability Test - Stress CPU
55% … Intel Processor Diagnostic Tool v4.0 - CPU Load

TDP … Stability Test - Fluctuating Workload (Peak)

123% … OCCT v4.5.1 - CPU: OCCT (AVX, No Offset)
118% … LinX v0.6.5 - Default
116% … IntelBurn Test v2.54 - High
113% … OCCT v4.5.1 - CPU: Linpack (AVX, No Offset)
110% … AIDA64 v5.95.4500 - System Stability Test - Stress FPU
99% <-- Asus RealBench v2.56 - Stress Test (AVX, No Offset)
94% … Sandra 2017.09.24.41 - Burn in - Processor Tests
92% … CineBench v15.0 - CPU - Render Test
79% … Intel Extreme Tuning Utility v6.4.1.15 - CPU Stress Test"

So as you can see Prime is first place and if you want system stable in any case, you need pass new Prime95. So as I told you in other post, make stable 4.7Ghz or 4.8Ghz and then try to rise clock. In my case I'm keeping my cpu on 4.8Ghz with prime stable 1.262v but for realbench stable I need only 1.207 or more,less depends on LLC lvl. So you need to bump your vcore to 1.380v and try metro again. Better to not run prime with that voltage bcs high temperature. You will fix whea with raise voltage, as I fixed, but'm happy with stable prime 4.8Ghz avx, and that is much better than unstable 5.1Ghz as I got with 1.360v depends on LLC lvl (cinebench, old prime, aida64 fpu...).

[Arne Saknussemm]

Pc is rock stable i think.

Your PC may beg to differ

This is why I recommended you test your PC after the stress testing...after all, all the stress test proves is that you can run the stress test...at the end of the day the easy test of "is it my OC" is to set defaults.

F5 defaults...and play Metro...if it still crashes you can look at drivers etc but if it doesn't...look to alter the OC...

[lpiotrowski211]

Nah it not crashes. I just left running game overnight for 9 hours. When i exit game i just checked for no reason event logs and 1 entry Whea Internal Parity was.

Anyway. I had voltage 1.315v. Now i bumped to 1.320v and see. But anyway its vcore related 100%?


I have LLC level 6 but when i check Level 7 i was getting spikes from 1.315v to 1.360v in load is this normal?

On level 6 , 1.315 to 1.323v

[Arne Saknussemm]

If it isn't crashing I'd forget all about it...

There is a point in stability testing where the instability passes to the tester...if you go on too long

[lpiotrowski211]

Ok last question. I tested LLC and on level 7 voltage is increasing to 1.360v on load from 1.315 idle. Manual entered in uefi : 1.310V.


So i use level 5 LLC. Manual entered in uefi : 1.360V. Idle Windows : 1.350V , load ( AVX ) 1.314v stable.

Should i keep it?

[Jane Doe]

Yes keep it for now and play metro. Do not run new prime95 with that voltage, it will not be good For games will be enough and I think max temperature in game will be spikes about 75C max... LLC lvl 5 is very good. Manual is constant voltage, you can try Adaptive 1.340v + offset (positive) 0.010v or 0.020v or 0.030v untill you get stable. And try to get stable with lvl5 or lvl6.

[mdzcpa]

Yes keep it for now and play metro. Do not run new prime95 with that voltage, it will not be good For games will be enough and I think max temperature in game will be spikes about 75C max... LLC lvl 5 is very good. Manual is constant voltage, you can try Adaptive 1.340v + offset (positive) 0.010v or 0.020v or 0.030v untill you get stable. And try to get stable with lvl5 or lvl6.

Word of advice. Stop looking in the Event Viewer to find problems. I used to do that too. I'd spend countless hours troubleshooting everything from overclock stability issues to driver conflicts. It was never ending.


Now I use the Event Viewer for 2 things only. If my system actually exhibits problems, I use the Event Viewer to trouble shoot. I also look at it after my new builds are complete in the first few days. After that, I don't look. If the system runs well and does everything I want it to, then it is stable and running well. This approach has saved me countless hours of frustration and stress.


Carry on.....

[lpiotrowski211]

Hi. So i used LLC7,but in uefi : manual voltage 1.270V.
Then on idle/gaming have 1.280v-1296v to 1.314v ( depending on load % ). On realbench max steady 1.314v stable.

Can i test metro with that settings? LLC7 is the most stable maybe it will help?


WIth adaptive 1.310v and offset 0.001 i have fluctuations like 0,670v ,0680v on idle or 1.119v in realbench . So i remove adaptive maybe i am stupid for that.

[Luck100]

WHEA errors are hardware errors. You shouldn't get them on a stable system. If you're nearly stable you might get them occasionally and only in some applications. It should go away with more vcore or less aggressive overclock.