![]() ![]() Yes I know those high end hardware components are able to handle more load and temperatures but I did not expect it to oscillate so closely to their limit. 75 mV for cache is stable for most 10th Gen mobile CPUs. Everyone wants a thin and light laptop so they run at temperatures that used to be considered sky high. Nvidia GPUs running at 80☌ and Intel CPUs running at over 90☌ is normal for any recent laptop while playing a game. If you do not want this to happen, clear the Thermal Velocity Boost box in the FIVR window. It slows your CPU down 100 MHz when it gets over 70☌. Your Limit Reasons window shows TVB as the reason for throttling. That way you can display the main ThrottleStop window, the FIVR window and the Limit Reasons window all on the same screen at the same time. Open Limit Reasons first and then open the FIVR window. You can include Core Temp too but there is no reason to. Why not open the main ThrottleStop window on top of your Cinebench window? ThrottleStop shows temperatures and CPU MHz. I already know what Cinebench looks like. Most of your screenshot shows the Cinebench window. This is log from Event Viewer of that BSOD (source WHEA-Logger): Maybe BSOD in my NFS test was not affected by undervolting and simply the OS bailed because of thermals with Turbo enabled (I was playing only with Turbo disabled before). Maybe a dumb question but all needed to do is to set offset and hit apply so it changes offsets in the top right table of FIVR window, right? Or there is need to restart computer? Sorry I'm new in this and never done it before. But unfortunatelly I did not notice any difference and the temps during rendering were hitting 90s for all tests. Just downloaded the Cinebench and ran some tests with -0, -75, -100 for core and -75 for cache. I was forced to play with disabled Turbo. Well I was buying the laptop knowing thermals are poor but I did not knew that its so bad that it will prevent me from playing because of temps. One error in the TS Bench test is too much.ĭownload Cinebench you very much for response. If the TS Bench reports any errors at the top, your undervolt is not stable so you need to reduce it. Full load tests are important but partial load tests like 8 or 10 threads are just as important. Reduce the amount of undervolt until you are 100% stable.Īfter you make any voltage adjustments, run a few different TS Bench tests. If you are, this confirms that your voltages are not stable and you have undervolted too much. You should never be seeing any WHEA errors. Many 10th Gen mobile CPUs like the 10750H are only stable with the cache at -70 mV or -75 mV. The biggest mistake people make is they set the cache offset voltage too high. The core and cache offsets do not have to be set equally. These other three voltages are not that important. You do not want undervolting these to interfere with stability when trying to get your core and cache setup correctly. When gaming, you are using the Nvidia GPU. Leave these three at an offset of +0.0000. Consider replacing the thermal paste with something like Noctua NT-H2 or a similar paste that works well at high temperatures.ĭo not undervolt the Intel GPU, iGPU Unslice or the System Agent until you have finalized your CPU core and cache settings. Click to expand.That sounds like a poorly designed laptop.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |