Hwinfo high usage while minimized

R

Ryxes

Member
Hello,

I'm using the latest beta build and I noticed really high CPU usage while the program is minimized to tray. At times I've seen it climb as high as 18% and I have no idea if it's related, but at the same time as it goes into the double digits, my CPU temperature spikes to 79-85C. When the CPU usage drops my temperature goes back to normal as well. I'm using the program at default settings, with the only 2 options enabled being report to gadget, because of Fan Control and an alert set to inform me if my GPU pin voltage ever drops below 11.8V (RTX4000s series GPU). I've enabled profiling time, to see if anything is out of the ordinary there, but the highest delay I've seen is 90ms on a sensor I disabled. I really have no idea what's going on and it seems to happen at random times, regardless if under load or idle.

My system is:
Motherboard: Asrock B650E Steel Legend WiFi
CPU: Ryzen 7900X3D / Curve Optimizer -25 / Limited to 85w and 85C
RAM: G.Skill Trident Z Neo 2x32GB 6000Mhz CL30 Expo on
GPU: MSI RTX4080 Suprim X on latest driver
BIOS: Latest version
OS: Windows 11 Pro Build 22621.3007
 

Attachments

  • hwinfocpu.png
    hwinfocpu.png
    4.1 KB · Views: 6
Does this high load happen only when HWiNFO is minimized? So when you open the sensors window, does the load drop back?
 
So the problem is not the minimized state. Are you running some other monitoring/tweaking tools along with HWiNFO? They might be blocking it.
 
Just it, Fan Control, but that one gets it's sensor readings from HWiNFO, and MSI Afterburner to undervolt/memory OC the GPU. I don't run any other monitoring software.

Fan Control I use, because Asrock's BIOS does a very poor job of controlling the fans during Ryzen's infamous core spikes.

But I've had those 3 running alongside each other for a long time and never noticed any issues before.
 
Then please attach the HWiNFO Debug File of a short run when the usage is higher.
 
I'm hoping this is the right thing. It took me a few tries, because the file got too big to attach here if I left debug mode on longer than a minute.
 

Attachments

  • HWiNFO64.DBG
    1.7 MB · Views: 1
I can see there problems accessing the Nuvoton sensor and also the VRMs. This can be due to a collision with other monitoring tools.
Try to disable monitoring of the Nuvoton and Renesas (VRM) sensors.
 
Disabling them just made the cpu usage spike not pass 9% but it's still there. Could it be a collision with either Afterburner or RivaTuner? I don't have any other monitoring tools running, apart from Fan Control for my fans.
 
Alright disabling PresentMon and Renesas only makes it spike way less, only up to 6%. If I disable Nuvoton too then it's between 0.4-2%. In both cases this is CPU usage while HWiNFO is open. Disabling Nuvoton makes me lose readings for quite a lot of sensors sadly. Looking at it RivaTuner and Afterburner don't read any of them though.
 

Attachments

  • HWiNFO Nuvoton sensor.png
    HWiNFO Nuvoton sensor.png
    30.7 KB · Views: 1
Maybe FanControl does?
I will try to reduce the higher load caused by this problem in the next HWiNFO build but that won't fully resolve the conflict.
 
I think it could be the fans. Fan Control only gets readings from HWiNFO for the GPU temp, CPU temp and fan speed sensors, I haven't enabled any other ones in the report to gadget. And Nuvoton has the fan speed sensors.
 
I just had a look at FanControl. It uses LibreHardwareMonitor and that doesn't properly handle access to the Nuvoton.
 
You must log in or register to reply here.
Back
Top