Are you checking the voltages? I'm reading this correctly.
Are you checking the voltages? I'm reading this correctly.
you're checking if the adjustments you made affect the readings on your Aorus B450m with the 2600x. under the volt setting, the third pstate option was set to 18, showing 140000 in that window. after trying it, the main screen in Pop OS shows the CPU-x app now reads 1.4 volts. with these changes, the dynamic vcore is at +0, dynamic VCORE SOC is normal, and DRAM voltage is 1.330v as before. when I left the pstates alone and only changed the dynamic vcore to .215, the SVID would read 1.135, which didn't seem right. i'm wondering if the current reading from CPU-x is accurate. you can check the image here: https://imgur.com/a/hCnYmfM. also, the 1.44 volt mentioned might refer to one of the 1.35 values for RAM.
It seems you're p-state pushing the limits... good luck with that. I attempted it with my 1700 but gave up because I just couldn't manage the state configurations properly.
Also, the BIOS P-state settings I've encountered are almost always in HEX format, not decimal. This might be a key reason why I struggled to get the right settings. P-State was a short-lived trend in early BIOS versions with Zen 1 on 300 series boards, but it's mostly faded away since later BIOS releases offer better control. Notably, starting with Zen+ (2000 series CPUs), PBO became quite effective for them.
Lastly, I'm not certain what CPU-X refers to... it could be something like CPU-Z, possibly a copy? If it is CPU-Z, the voltage readings are extremely unreliable for Ryzen because its strobe delays take minutes to measure. This makes every reading significantly delayed compared to the actual processor state, especially since Ryzen dynamically adjusts voltage—even in 2nd Gen models. For accurate information, I recommend using HWInfo64 instead.
Ryzen doesn't stay completely still, it's always active. An operating system such as Windows, with many background tasks running, increases its activity. It often puts individual cores into deep sleep when voltage drops significantly—possibly even to zero when the core is off—but no utility can detect this low voltage. The issue arises because if a utility requests the voltage, it would wake the processor expecting a higher value. Therefore, a reliable utility will report the last measured voltage before entering sleep.
You can observe much of this behavior using HWINfo64. Check the C6 sleep status per core, for example. The time spent in C6 means the voltage is nearly zero. You can also track SVI2 VCore and examine the MINIMUM voltage. Alternatively, averaging the Vcore over a period when the computer is idle provides a clearer picture of what "idle" voltage looks like.
Creating a graph on your desktop to visualize VCore changes over time is also helpful. Graphs are especially useful for a processor like Ryzen that constantly changes state.