Why would my 9980XE start reporting incorrect TDP?

[cptninc]

I was trying to resolve a recurring BSOD due to WHEA_UNCORRECTABLE_ERROR, and saw that this apparently is just a known thing on x299 boards. Transitioning the CPU from idle to high load the wrong way brings this about at the motherboard default settings. Increasing Vcore to 1.2 is the much fabled solution.

So, I went poking around the settings on my Rampage VI Extreme to see if I could find that setting. I couldn't, so I did not make that change. I rebooted and figured I'd go on The Google to learn more.

Fired up F@H and checked my cooling monitor. CPU is reporting 100% load, loop temps are what I'd expect from 250-300W.... but HWmon shows the CPU reporting a continuously changing value of only 2-4W.

Any idea what I might have done to break that? CPU TDP is actually something I monitor during some renders because it's how I discover CUDA crashes with Iray, so I'd kind of like to have it working again.

 

[EarthDog]

HWinfo just isn't reading the sensor properly... that's all. You seem to be saying it read it properly previously, as in, before you made a voltage change that you actually ended up not doing? I'm a bit confused.

EDIT: So.........reboot w/o changes and now HWMon isn't reading the sensor right?

Try Hwinfo instead of Hwmon?

Weird issue.

 

[cptninc]

It read properly previously (or at least gave believable values).

I did not make the voltage change but it's quite possible that I changed something else unwittingly.

Mostly unrelated, this processor shows up as a 7980XE in HWMonitorPro for some reason. That's not a new thing though. I don't really know why I'm mentioning it here.

 

[EarthDog]

Have you updated the BIOS on your motherboard to the latest version?

If you are not overclocked, reset the CMOS/optimized defaults and to set everything back to stock and see if it reports.

I have a 7960X which reports fine in Hwmon (Asus X299 board).

EDIT: I lied.... my wattage there is 1W or so, lol... so it isn't reporting right on mine either. That said I am manually overclocked so................. maybe I changed something.

EDIT2: Do you have any other monitoring software running? or other apps like icue or RGB apps, etc?

 

[cptninc]

OMG. iCue. So many expletives come to mind. I uninstalled iCue a few months ago when I finally converted over to Aquaero. That fixed about 20 software issues all by itself.

For other monitoring software, I'm running GPU-Z and Aquasuite. I'm not sure if the latter simply grabs its data from hwinfo or if it runs its own service to poll that data, but that one is also showing the same 2-4W.

I will try resetting the CMOS to defaults later today and see how that goes.

 

[RJARRRPCGP]

If the processor responded with "internal timer error", then you must at least disable turbo core, if not EIST and C1E as well!