Noob looking for opinions on undervolt settings. (9900k)

[jansimek]

Hello! i have an 9900k sitting in a Gigabyte Aorus z390 Master mobo, cooled by a Noctua NH-D15.

Been messing around with adaptive voltage a bit with a goal of reducing temps and also finding the perfect offset to run on as low voltage as possible at a 4.7 all clock speed.

I am looking for opinions on whether or not should i consider my current settings stable and stop tweaking.

Below i added some of my observation from stress testing. I am not interested in overclocking any higher and i am not interested in messing with other settings such as llc as long as im stable. The 4.7 "stock" all core turbo frequency is enough for me, at least for now.

BIOS settings :

CPU clock ratio : 47
MCE : OFF
Turbo Boost : OFF
XMP : 1
Vcore mode : Normal
and a DVID offset.

-0.160 error 5m into p95(version 29.8) small fft (worker error & stop +a whea error)

-0.155 error 12m into p95(version 29.8) small fft (worker error & stop +a whea error)

-0.150 error 30m min into p95(version 29.8) small fft (worker error & stop +a whea error)

-0.140 error 1h 20m into p95(version 29.8) small fft (this time only the worker stop & error happened with no whea log in windows. Sign of getting closer to 100% stable?)

-0.130 passed 3hours 48minutes p95(version 29.8) small fft with no errors. Cpu package max was 86c after almost 4 hours of what is the absolutely worst case scenario, if i understand the prime version differences right? im assuming that running the 26.6 non AVX version after this is pointless as that could probably go on forever.

Vcore under the newest p95 small fft test load im running is between 1.116 - 1.128 with the latest -0.130 DVID offset.

I had no issues, errors, crashes or anything of that sort running games such as battlefield V and other stress tests such as Aida64, Realbench, Cinebench even at an offset of -0.150. It's just the P95 that kept producing errors up until i bumped it up to -0.130.


is this 100% stable? are my bios settings "healthy" for long term?


Thanks. Jan.

 

[EarthDog]

Stable is if your PC doesnt crash for your uses. If it doesnt, I'd call that stable.

 

[jansimek]

Stable is if your PC doesnt crash for your uses. If it doesnt, I'd call that stable.

Hello and thank you for the input. My idea is to achieve 100% stability under all possible circumstances, it just feels right to do so.

Would you think 3 hours 48minutes of small fft latest p95 with no error is enough to confirm that?

 

[wingman99]

Yes I would call running small fft latest p95 with no error 3 hours 48 minutes safe for heavy floating point calculations. Try some heavy integer with floating point calculations, like web browsing or gaming if you game.

 

[EarthDog]

Hello and thank you for the input. My idea is to achieve 100% stability under all possible circumstances, it just feels right to do so.

Would you think 3 hours 48minutes of small fft latest p95 with no error is enough to confirm that?

Maybe... hard to test all possible circumstances...

That said, 4 hours is what many consider to be a MINIMUM for stability testing. If you want "100%" stability I'd at least test for 8+ hours on both small fft and blend for the memory (and even then, things can happen).

 

[jansimek]

Yes I would call running small fft latest p95 with no error 3 hours 48 minutes safe for heavy floating point calculations. Try some heavy integer with floating point calculations, like web browsing or gaming if you game.

Gaming seems to run absolutely fine, had no issues yet.
However im a little confused by the differences in my load voltages. I did a little research on vdroop, but it seems like im 100% stable like this.

4.7g(-0.130 offset) prime95 26.6 NON AVX small fft vcore under load : 1.092 - 1.104 100% stable

4.7g(-0.130 offset) prime95 29.8 AVX small fft vcore under load : 1.116 - 1.128 100% stable

max gaming and general use vcore reading i get is 1.224 when i leave hwinfo running for a couple hours (goes up and down between 1.056 - 1.212 in gaming)

why is the stress test vcore peak lower than gaming? is it because of the steady workload of p95 or am i missing something?

 

[wingman99]

Gaming seems to run absolutely fine, had no issues yet.
However im a little confused by the differences in my load voltages. I did a little research on vdroop, but it seems like im 100% stable like this.

4.7g(-0.130 offset) prime95 26.6 NON AVX small fft vcore under load : 1.092 - 1.104 100% stable

4.7g(-0.130 offset) prime95 29.8 AVX small fft vcore under load : 1.116 - 1.128 100% stable

max gaming and general use vcore reading i get is 1.224 when i leave hwinfo running for a couple hours (goes up and down between 1.056 - 1.212 in gaming)

why is the stress test vcore peak lower than gaming? is it because of the steady workload of p95 or am i missing something?

It is becase transistors need load line resistor with the voltage regulator module buck converter. The voltage goes down when the AMPs go up.

Here is a description of load line. "Load line (electronics)
In graphical analysis of nonlinear electronic circuits, a load line is a line drawn on the characteristic curve, a graph of the current vs the voltage in a nonlinear device like a diode or transistor.
It represents the constraint put on the voltage and current in the nonlinear device by the external circuit." https://en.wikipedia.org/wiki/Load_line_(electronics)