F5F Stay Refreshed Hardware Desktop Changing CPU speeds and timing in Cinebech

Changing CPU speeds and timing in Cinebech

Changing CPU speeds and timing in Cinebech

X
xPerez
Junior Member
27
07-13-2023, 10:47 PM
#1
Hey! Your setup looks solid with the new case and cooler. The temperatures stayed well below 60°C, so you didn’t hit thermal limits. After trying a 4.4GHz boost, you saw some frame rate spikes but also crashes in Hellblade. The Cinebench data shows CPU core clocks jumping between 4.4Ghz and 800Mhz, with temps dropping from 67°C to 27°C—definitely not due to overheating. The VRM temps hit 91°C, which is high but still within normal ranges for a high-end i7. It seems the issue might be related to power delivery or thermal management under load, possibly causing instability rather than throttling. You might want to check your power supply and ensure it can handle the increased load at higher frequencies.
X
xPerez
07-13-2023, 10:47 PM #1

Hey! Your setup looks solid with the new case and cooler. The temperatures stayed well below 60°C, so you didn’t hit thermal limits. After trying a 4.4GHz boost, you saw some frame rate spikes but also crashes in Hellblade. The Cinebench data shows CPU core clocks jumping between 4.4Ghz and 800Mhz, with temps dropping from 67°C to 27°C—definitely not due to overheating. The VRM temps hit 91°C, which is high but still within normal ranges for a high-end i7. It seems the issue might be related to power delivery or thermal management under load, possibly causing instability rather than throttling. You might want to check your power supply and ensure it can handle the increased load at higher frequencies.

T
TameNaz436541
Junior Member
9
07-13-2023, 11:39 PM
#2
I'll also mention that the CPU package power is decreasing between 98W and 54W in line with this trend. When I lowered the CPU to the default 4.2GHz using Gigabyte easy tune, the package operated at 4Ghz (not 4.2Ghz) during the Cinebench test, maintaining the same throttling pattern but slightly less often and occasionally dropping to around 1.2Ghz.
T
TameNaz436541
07-13-2023, 11:39 PM #2

I'll also mention that the CPU package power is decreasing between 98W and 54W in line with this trend. When I lowered the CPU to the default 4.2GHz using Gigabyte easy tune, the package operated at 4Ghz (not 4.2Ghz) during the Cinebench test, maintaining the same throttling pattern but slightly less often and occasionally dropping to around 1.2Ghz.

0
07GmanBEAST07
Member
213
07-18-2023, 08:43 PM
#3
This might be the problem. 91C isn't extremely risky for a VRM, but the sensor placement could mean the hotspot reaches 120°C, which is unsafe. Consider adding a fan and directing it at the VRM to test. This seems like power throttling. You can also look in the BIOS for any limits on power or temperature for the VRM and try adjusting them, though I don't have a Gigabyte Z170 board to point you toward specific sources.
0
07GmanBEAST07
07-18-2023, 08:43 PM #3

This might be the problem. 91C isn't extremely risky for a VRM, but the sensor placement could mean the hotspot reaches 120°C, which is unsafe. Consider adding a fan and directing it at the VRM to test. This seems like power throttling. You can also look in the BIOS for any limits on power or temperature for the VRM and try adjusting them, though I don't have a Gigabyte Z170 board to point you toward specific sources.

Y
YoloGeek22
Member
152
07-29-2023, 06:34 AM
#4
Leave HWiNFO, open ThrottleStop, and follow the instructions. https://www.techpowerup.com/download/tec...ottlestop/ Share a screenshot of ThrottleStop with the Limit Reasons window active while Cinebench runs. Red boxes in Limit Reasons will indicate the cause of throttling—possibly VR temperature or an insufficient power limit.
Y
YoloGeek22
07-29-2023, 06:34 AM #4

Leave HWiNFO, open ThrottleStop, and follow the instructions. https://www.techpowerup.com/download/tec...ottlestop/ Share a screenshot of ThrottleStop with the Limit Reasons window active while Cinebench runs. Red boxes in Limit Reasons will indicate the cause of throttling—possibly VR temperature or an insufficient power limit.

K
Komodo88
Senior Member
749
08-04-2023, 08:49 PM
#5
You shared your findings and adjustments during testing. The observations suggest that thermal management might be the key factor, especially with VRM limitations. It seems your setup is handling higher clock speeds without immediate throttling, but you're still experiencing elevated VR temperatures. The results at different frequencies indicate VR thermal throttling is occurring despite stable clock speeds, which could point to insufficient cooling or power delivery under stress. You might want to evaluate the board's thermal design and consider upgrading components if needed.
K
Komodo88
08-04-2023, 08:49 PM #5

You shared your findings and adjustments during testing. The observations suggest that thermal management might be the key factor, especially with VRM limitations. It seems your setup is handling higher clock speeds without immediate throttling, but you're still experiencing elevated VR temperatures. The results at different frequencies indicate VR thermal throttling is occurring despite stable clock speeds, which could point to insufficient cooling or power delivery under stress. You might want to evaluate the board's thermal design and consider upgrading components if needed.