Core0 clock drops to zero

nebulus

New Member
Hello!

I'm not sure if this is a bug in HWINFO or a problem on my side, so I decided to post here first. I have an AMD Phenom II X2 555, clocked at 3200Mhz (default). I noticed that while idle, both summary and sensors page shows me very short drops to zero in Core0 clock speed, while Core1 clock speed is relatively stable. In order to even see those zero values i had to set refresh interval to 200ms. Even more, the minimum value displayed by the sensor page is not zero, even though current value briefly drops to zero. Cool'n'Quiet is disabled from BIOS, but even so, this happens when the CPU is idle. Any ideas ?

Thanks!
Adrian
 

Attachments

  • ZeroClock.PNG
    ZeroClock.PNG
    34.6 KB · Views: 8
Hi Adrian,
thanks for the report.
1. Does this problem occur also when you close the System Summary window and leave only Sensors window open?
2. Does this problem occur also when you disable the "Use HPET" option?
Please enable sensor logging to store all values into a CSV file and try to capture the situation. Then please attach or send me the the log to analyze it.

Martin
 
Hello!

1. The problem seems to occur only when the System Summary window is opened. When I close that page, the reading appears to be stable (as this is not a fully reproducible issue, I can't be 100%).
2. The problem also occurs when I disable the "Use HPET" option.

I will attach a part of a log which shows the zero clock issue.
Adrian

Edit: I continued to log the values when showing only the sensors page. A drop to zero doesn't seem to manifest itself in these conditions, but I still noticed lower values for the clock for some measurements (see Log2.csv)
 

Attachments

  • LogZeroClock.CSV
    41.4 KB · Views: 2
  • Log2.CSV
    5.8 KB · Views: 1
Fluctuating clock is absolutely OK since all modern CPUs change their clock several times per second (to preserve power when there's no load).
However the 0 MHz is definitively a bug and I can clearly see it in the log. I'll implement a patch for this issue in the next build.
Thanks for the feedback !
 
Thank you Martin for posting your log files! I noticed the same issue this morning after 8 hours of FPU testing my i9 9900k. So I was a touch spooked by the error. Thank you Nebula for confirming there is a bug! Amazing support. I will be donating to HWiNFO as soon as I am able. It seems good support is hard to come by these days.
 
On my z390 master I get something similar 1 core will drop to 800mhz yet should stay at 4.9ghz. Not sure if it's a bug or a windows power profile bug

Screenshot 2021-01-17 175706.png
 
Last edited:
Back
Top