5.1 @ 1.275v or 5.2 @ 1.390v i9-10900k
5.1 @ 1.275v or 5.2 @ 1.390v i9-10900k
Hey everyone, I've been experimenting with the new chip and upgraded the whole PC to overclock it. It turns out I didn't quite hit the perfect performance numbers—still running at 5.1 across all cores with a low voltage of about 1.275bios and around 1.266 load on LL3. I'm not satisfied with this level of boost and would prefer to invest in a much better cooler, aiming for something like 1.4 at 5.3 or higher. I'm currently using the NOCuta nh-15.
Given the significant voltage increase needed to reach 5.2, it's worth considering whether it's worth the effort. I've done a lot of research, and 1.390 volts seems safe as long as temperatures stay low during regular use. Alternatively, sticking with 5.1 at 1.275v until I can upgrade my cooling system might be a better path. Any advice would be really appreciated.
CineBench stable 5.1 at 1.275 with peak Core temp of 81°C ... what room temperature is needed?
CineBench stable 5.2 at 1.390 ... what Core temp?
Excessive Vcore rise
(115 millivolts)
from 5.1 to 5.2
...
Only the game stable version
5.3 at 1.410
... what Core temperature?
Vcore is nearly too high.
The significant 115 millivolts needed to jump from 5.1 to 5.2 clearly indicates your i9-10900K sample can't maintain all-core overclocking at 5.2 GHz with acceptable core temperatures. As mentioned before, even the Noctua NH-D15 is
the best air cooler available, but it struggles to keep up with the 10 Core 20 Thread i9-10900K at those voltage levels.
The mentioned overclocking is minimal compared to the effort needed to raise the voltage. Voltage has a strong impact on power loss, changing it significantly. A single 0.1GHz increase raises power use by roughly 2%, while increasing the voltage leads to a nearly 19% rise.
Your voltage and frequency readings seem inconsistent.
A 5.1 at 1.275 V with a 1.390 GHz frequency corresponds to a 110 millivolt rise.
When moving from 5.2 to 5.3 at 1.390 GHz, the increase is only about 25 millivolts.
Here’s an example of how Core Voltage and Frequency typically relate:
If your processor remains stable at 5.1 GHz with a 1.275 Vcore, then at 5.2 GHz it should need no more than roughly 1.34 Vcore, while at 5.3 GHz it might require up to 1.41 Vcore for stability. Your device should follow a similar trend.
Given your 1.39 Vcore for 5.2 GHz, this represents a 110 millivolt increase from the previous setting, but only a 25 millivolt rise between 5.2 and 5.3. This indicates your results might have been slightly off, possibly due to inconsistent testing methods or rushed conclusions.
For optimal overclocking, remember that a 100 MHz boost usually demands about a 50 millivolt increase in Core voltage to keep stability. If further increases beyond 70 millivolts are necessary, it suggests you may be pushing your processor beyond its safe operating limits.
Each microarchitecture has a recommended maximum Vcore. Below are the values from 14 to 65 nanometers since 2006, along with their corresponding degradation curves.
According to these curves, CPUs tend to degrade more with each die-shrink, except for the 14nm technology where FinFET improvements have enhanced voltage tolerance.
The chart also shows that a maximum Vcore of 1.425 is generally safe for 14nm processors, aligning with recommendations from Intel’s Overclocking Test Lab (see paragraphs 8–10).
However, despite being a high-end air cooler like the Noctua NH-D15, it may not provide sufficient cooling for the i9-10900K at the voltages you’re targeting because of its substantial power draw (over 325 watts). For better results, consider using a 360 mm AIO or a custom liquid cooler.
It’s wise to follow the guidance from jay32267 and hotaru.hino, and consult the provided links to understand overclocking risks, especially electromigration and degradation. Pushing your CPU beyond its recommended limits for such gains is unlikely to improve performance meaningfully and could harm long-term reliability.
Thanks for the clear feedback, and you're absolutely correct, I can't really confirm if 5.3 is fully stable since it only performs well at 1.415v. Any game I run might fail under stress.
Cinebench will definitely trigger a BSOD if it doesn't handle thermal throttling properly beforehand.
Yes, it's very reliable at 1.275v around 1.266 load voltage on all cores—looping for hours with max temps near 81°C makes sense that I need 1.390v to run 5.2?
I'm stable at 1.355v but will crash if I try to run below 1.390v in Cinebench R20.
CineBench stable 5.1 at 1.275 with peak Core temperature of 81°C ... what room temperature would be optimal?
CineBench stable 5.2 at 1.390 ... what Core temperature is expected?
Excessive Vcore rise observed (115 mV) when changing from 5.1 to 5.2.
Only the game stable version
5.3 at 1.410 is available — what Core temperature should be considered?
Very high Vcore increase indicates the i9-10900K sample struggles to maintain stable long-term overclocking at 5.2 GHz with acceptable Core temps. Even with the Noctua NH-D15 air cooler, it may not handle the 10 Core 20 Thread configuration at those voltages because of the substantial power draw (over 325 Watts). For reliable results, a standard 360 mm AIO or a custom loop is advised.
CineBench serves as a practical testing tool, but its power usage and Core temperatures closely match Prime95 Small FFTs without AVX, making it a consistent benchmark for thermal evaluation. Here’s a summary of tools grouped by thermal and stability metrics based on TDP percentages across six Intel generations at default settings, rounded to the nearest 5%:
Although tests span from 70% to 130% TDP, Windows Task Manager treats each test as 100% CPU usage, not actual TDP load.
Core temperatures directly reflect power draw (Watts), which depends on workload.
If you aim to keep your 10900K below around 80°C—strongly advised—then pushing it to 5.1 GHz at 1.275 Vcore is the safest overclock choice to prevent issues like electromigration and degradation.
Pushing beyond recommended Core voltage and temperature isn’t worthwhile for a marginal performance gain, as it offers no significant benefit to overall system performance.
Click the links above for detailed guidance.
The room temperature is quite low, around 19°C, and without a side panel on, it could rise slightly above 81°C—though that appears to be the cap. I really appreciated your advice and plan to stick with 5.1 until I try the custom loop (something I wanted). However, I had a bad experience with the new Corsair closed loop; it destroyed two of my Sapphire vapor X 7970s. It took months to figure out because the leak was very small and dried out quickly without leaving any residue. For a while now, I’ve been dealing with water issues, but I still love overclocking. The 11900K didn’t work for me, so I won’t reduce cores in the future. A custom loop I built might help address any leakage worries.
My main concern was about my chip sample—it’s stable at 1.275 LLC3 across all cores with fixed ratio. From what you mentioned, it should only require around 1.35 to reach 5.2? From what I see, a jump from 5.1 to 5.2 seems quite large, yet the sample performed well at 1.415 and 5.2 was stable for many days across different games.
I can’t test Cinebench at 1.355+ due to my cooler.
🙁 What I know is that 5.3 was reliable at 1.415, and 5.2 was stable between 1.355 and 1.365, but it needed about 1.390 to pass Cinebench R20 and cores hit 100%.
UPDATE:
I re-tested 5.1 after reviewing your graphs and links. It’s really strange—I needed such a big jump to reach 5.2. Although the game was stable for many days across various titles, it took around 1.390 to pass Cinebench R20. Cores reached 100 at that point.
Thanks for your assistance!