F5F Stay Refreshed Hardware Desktop OCCT MHz variations between 3.9GHz and 5GHz

OCCT MHz variations between 3.9GHz and 5GHz

OCCT MHz variations between 3.9GHz and 5GHz

G
GalPlaysMC
Member
55
09-19-2022, 07:45 AM
#1
Hello, your temperatures and power usage are normal and not nearing any limits. I notice the CPU MHz vary a lot during the medium data set test in OCCT, ranging from 3.9GHz to 5GHz in the top left corner. Task Manager and HWInfo appear to be accurate. Could this be an issue with OCCT itself? Right now, turbo is enabled at 5GHz in the BIOS. Your system specs are: 16GB RAM, 3200MHz, Intel i5-13600K, Gigabyte UD AX DDR4 motherboard.
G
GalPlaysMC
09-19-2022, 07:45 AM #1

Hello, your temperatures and power usage are normal and not nearing any limits. I notice the CPU MHz vary a lot during the medium data set test in OCCT, ranging from 3.9GHz to 5GHz in the top left corner. Task Manager and HWInfo appear to be accurate. Could this be an issue with OCCT itself? Right now, turbo is enabled at 5GHz in the BIOS. Your system specs are: 16GB RAM, 3200MHz, Intel i5-13600K, Gigabyte UD AX DDR4 motherboard.

Y
yalo29
Senior Member
641
09-19-2022, 01:33 PM
#2
I'm guessing that it's an issue with the large clockspeed differential for the P and E cores. OCCT probably doesn't know how exactly to display a single number that shows that half of your cores are running at 5-5.2GHz and the other half are at 3.9-4.1GHz. I could imagine that, in the future, OCCT might be coded to display two different frequency numbers, or perhaps only show the frequency of the fastest core, but for now, I'm guessing that it tries to give a sort of "average" and its algorithm for doing that is getting confused.
Y
yalo29
09-19-2022, 01:33 PM #2

I'm guessing that it's an issue with the large clockspeed differential for the P and E cores. OCCT probably doesn't know how exactly to display a single number that shows that half of your cores are running at 5-5.2GHz and the other half are at 3.9-4.1GHz. I could imagine that, in the future, OCCT might be coded to display two different frequency numbers, or perhaps only show the frequency of the fastest core, but for now, I'm guessing that it tries to give a sort of "average" and its algorithm for doing that is getting confused.

K
KablooieKablam
Posting Freak
908
09-20-2022, 04:46 AM
#3
Maybe OCCT combines P and E core MHz to display a single value. If that's true, I shouldn't focus too much on those numbers. During testing, watch the HWiNFO Performance Limit Reasons for any throttling warnings. This will reveal if power restrictions are occurring. Turbo power caps often trigger issues if BIOS settings aren't high enough.
K
KablooieKablam
09-20-2022, 04:46 AM #3

Maybe OCCT combines P and E core MHz to display a single value. If that's true, I shouldn't focus too much on those numbers. During testing, watch the HWiNFO Performance Limit Reasons for any throttling warnings. This will reveal if power restrictions are occurring. Turbo power caps often trigger issues if BIOS settings aren't high enough.

J
JessBrearley
Member
195
09-20-2022, 09:16 AM
#4
I performed the test with XTU and confirmed no power throttling was detected.
J
JessBrearley
09-20-2022, 09:16 AM #4

I performed the test with XTU and confirmed no power throttling was detected.