-
Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
CORE TURBO RATIO LIMITS
- Thread starter Nuhaiz
- Start date
unclewebb
ThrottleStop & RealTemp Author
- Joined
- Jun 1, 2008
- Messages
- 8,116 (1.33/day)
You need to disable VBS including core isolation memory integrity so ThrottleStop can work correctly. Reboot and delete the ThrottleStop.INI configuration file after you do that.
Read the disable VBS link in my signature.
Read the disable VBS link in my signature.
Thank you very much for the guidance. And for the TS bench, do i need to test all the box? or just the default one? cause when i am testing with the default, it doesnt have any error but when i change the value, it present an error. So, do i need to test all the boxes and make sure there are no single errors?
unclewebb
ThrottleStop & RealTemp Author
- Joined
- Jun 1, 2008
- Messages
- 8,116 (1.33/day)
No need to test all of the TS Bench settings. I like the 960M test. If any errors are reported in any test, increase the CPU voltage.
What CPU do you have and what voltage settings are you trying to use?
Post an updated FIVR screenshot.
What CPU do you have and what voltage settings are you trying to use?
Post an updated FIVR screenshot.
Last edited:
unclewebb
ThrottleStop & RealTemp Author
- Joined
- Jun 1, 2008
- Messages
- 8,116 (1.33/day)
I just wanted to double check that VBS was disabled. The ThrottleStop FIVR window looks correct. The default turbo ratios and memory speed are being reported correctly. The FIVR monitoring table confirms that your undervolt is working correctly. It also shows that Power Limit 4 (PL4) is set to only 40. Open the TPL window and set Power Limit 4 to a value of 0.
A maximum offset of about -60 mV is typical for similar 10th Gen mobile CPUs. You can start to lose stability if you try to go much beyond that.
Have you tried setting the core offset to a bigger number compared to the cache offset? Do some Cinebench testing. Some Intel CPUs do not need to have the core and cache offsets set equal to each other. With older 8th and 9th Gen mobile CPUs, many users saw better performance or better temperatures by setting the core offset higher than the cache offset. This trick seemed to reduce the voltage when AVX instructions were being executed by the processor. Cinebench uses lots of AVX instructions so it is a good way to confirm any improvements. If Cinebench temperatures or performance improves, you should see some improvement in any recent game that also use a lot of AVX instructions.
Maybe you can do some testing at -60 mV cache offset, -100 mV to -150 mV for the core offset. Do some TS Bench testing to make sure there are no errors and then run Cinebench to see if there is any difference. There might not be. If you cannot prove that there is a difference then set the two offset voltages equal to each other.
Check the Log File box when you play a game so you have a record of your CPU performance and temperatures. Attach a log file to your next post so I can see if there are any throttling issues.
The FIVR monitoring table is reporting that Speed Shift EPP is set to 178. This can reduce maximum performance. Are you using the Windows Balanced power plan? That typically sets EPP to 84 on most computers. Did you check the Speed Shift EPP box on the main ThrottleStop screen? I prefer not to check the ThrottleStop EPP box. This allows Windows to manage the EPP value with no interference from ThrottleStop.
A maximum offset of about -60 mV is typical for similar 10th Gen mobile CPUs. You can start to lose stability if you try to go much beyond that.
Have you tried setting the core offset to a bigger number compared to the cache offset? Do some Cinebench testing. Some Intel CPUs do not need to have the core and cache offsets set equal to each other. With older 8th and 9th Gen mobile CPUs, many users saw better performance or better temperatures by setting the core offset higher than the cache offset. This trick seemed to reduce the voltage when AVX instructions were being executed by the processor. Cinebench uses lots of AVX instructions so it is a good way to confirm any improvements. If Cinebench temperatures or performance improves, you should see some improvement in any recent game that also use a lot of AVX instructions.
Maybe you can do some testing at -60 mV cache offset, -100 mV to -150 mV for the core offset. Do some TS Bench testing to make sure there are no errors and then run Cinebench to see if there is any difference. There might not be. If you cannot prove that there is a difference then set the two offset voltages equal to each other.
Check the Log File box when you play a game so you have a record of your CPU performance and temperatures. Attach a log file to your next post so I can see if there are any throttling issues.
The FIVR monitoring table is reporting that Speed Shift EPP is set to 178. This can reduce maximum performance. Are you using the Windows Balanced power plan? That typically sets EPP to 84 on most computers. Did you check the Speed Shift EPP box on the main ThrottleStop screen? I prefer not to check the ThrottleStop EPP box. This allows Windows to manage the EPP value with no interference from ThrottleStop.
Sorry unclewebb. My laptop currently having the same problem like last time. Whenever my laptop got hot, the screen flickering and when it cools down, the screen goes back to normal. I am afraid it had to do with my graphic card. So, I cannot do intensive test like cinnebech or any other intensive task. I will repair my laptop first and went back here. Thank you so much for your advice and guidance.