[SOLVED] How To Fix Low Utilization ?

Feb 23, 2022
34
0
30
I have been having an issue recently where my FPS drops very substantially in all the games I play. I made sure all my drivers are updated, and windows updates are done but it still occurs. I used DDU to remove all graphics drivers and reinstall them but still nothing. I downloaded HWMonitor and I am currently running Dying Light 2 and its showing my Utilization Limit to be 1, which from what I've read means it isn't being used fully. I've included the full HWMonitor screen in the screenshot below and my current PC specs.

Specs:
Processor: Intel(R) Core(TM) i7-10700KF CPU @ 3.80GHz, 3792 Mhz, 8 Core(s), 16 Logical Processor(s)
Motherboard: MPG B560I GAMING EDGE WIFI (MS-7D19)
GPU: NVIDIA 3060TI Founders Edition
RAM: 32GB of 3200GHz
PSU: 650W SFF Gold, cant find make and model without opening it up.
Disk Drive: WDC WDS100T2B0C-00PXH0 (thats what it says in task manager, its an M2 NVME SSD).

View: https://imgur.com/a/1CKzagn

View: https://imgur.com/a/q0pURFX
 
Solution
that voltage limit means that GPU is limited by voltage limiter....not much to render, high fps, maxed out gpu clocks
you could rise voltages, gpu would clock higher


hmm your bus interface has pretty high load, thats unusual since direct storage is not yet available
possible reason:
you dont have enough VRAM, so GPU runs from system RAM (thus high pcie load)
your PCIE runs at low version, check that with GPU-Z while gaming
imTTKGW.png
PerfCap reason is why the GPU won't boost further, and utilization is a default reason. I wouldn't get too hung up on it.

Considering that GPU % is at 100 and the GPU's clock speeds are 300MHz above the default spec, I'd say there's nothing wrong here. The only time you should really worry is if PerfCap reason is Thermals, because the video card is overheating at that point.
 
PerfCap reason is why the GPU won't boost further, and utilization is a default reason. I wouldn't get too hung up on it.

Considering that GPU % is at 100 and the GPU's clock speeds are 300MHz above the default spec, I'd say there's nothing wrong here. The only time you should really worry is if PerfCap reason is Thermals, because the video card is overheating at that point.
My FPS are way worse than they normally are so something is wrong, any ideas?
 
PerfCap reason is why the GPU won't boost further, and utilization is a default reason. I wouldn't get too hung up on it.

Considering that GPU % is at 100 and the GPU's clock speeds are 300MHz above the default spec, I'd say there's nothing wrong here. The only time you should really worry is if PerfCap reason is Thermals, because the video card is overheating at that point.

your gpu is 5C before thermal shutdown
maybe bad airflow in case or GPU need repaste
3060ti fe has 70C under load, yours sits at 88C

The HWMonitor screenshots above were taken with Dying Light 2 running, these next screenshots are with CS:GO running. The problem with Dying Light 2 was my GPU had utilization limit on which I'm assuming it meant its not being fully utilized hence my horrible FPS drops. CSGO on the otherhand, I got a different problem, it seems Reliable Voltage Limit and Operational Voltage Limit are being reached now. (On the screenshots below it shows that Operational Voltage Limit is now back at 0, but it was at 1 while I was running CSGO.

View: https://imgur.com/a/2XKJFRG

View: https://imgur.com/a/q6phLGu
 
that voltage limit means that GPU is limited by voltage limiter....not much to render, high fps, maxed out gpu clocks
you could rise voltages, gpu would clock higher


hmm your bus interface has pretty high load, thats unusual since direct storage is not yet available
possible reason:
you dont have enough VRAM, so GPU runs from system RAM (thus high pcie load)
your PCIE runs at low version, check that with GPU-Z while gaming
imTTKGW.png
 
Solution
Again, PerfCap reason is why the GPU doesn't want to boost further, it has nothing to do with GPU utilization or how the GPU actually performs except if the reason is due to thermals.

I don't think the hardware has an issue here. It may be a software configuration issue that may need things like reinstalling the drivers, the game, or whatever dependencies the game uses (which should automatically come with the game)

EDIT: If you're running games off a hard drive, some games may have a higher chance of hiccups because of their constant need to pull stuff from storage.
 
that voltage limit means that GPU is limited by voltage limiter....not much to render, high fps, maxed out gpu clocks
you could rise voltages, gpu would clock higher


hmm your bus interface has pretty high load, thats unusual since direct storage is not yet available
possible reason:
you dont have enough VRAM, so GPU runs from system RAM (thus high pcie load)
your PCIE runs at low version, check that with GPU-Z while gaming

You said my GPU is being limited because there's not much to render or I have too high of FPS already but I don't. I used to easily run CSGO at 350-400fps and now when I run it I barely get 100. Why is my GPU being limited if I'm barely getting any FPS as it is.

GPU-Z: View: https://imgur.com/a/b05YaBc

View: https://imgur.com/a/VTIIAjg
 
Again, PerfCap reason is why the GPU doesn't want to boost further, it has nothing to do with GPU utilization or how the GPU actually performs except if the reason is due to thermals.

I don't think the hardware has an issue here. It may be a software configuration issue that may need things like reinstalling the drivers, the game, or whatever dependencies the game uses (which should automatically come with the game)

EDIT: If you're running games off a hard drive, some games may have a higher chance of hiccups because of their constant need to pull stuff from storage.

I have made sure graphics drivers are up to date, even did DDU to do a fresh install, windows updates are good. I did chipset drivers for my motherboard and yet nothing has helped. I've probably factory reset the PC about 2-3 times now and nothing helps it. Al games are ran from an M.2 SSD
 
then something messed up in your bios settings, if your unfamiliar with how to configure your bios, then turn off your PC, switch PSU to OFF position, then few times press power button in front of case to discharge 5VSB from PSU capacitors (if your mainboard has power led, it will turn off once discharged), once done, located those 2 pins on your mainboard
yg05y7N.png

once located, use screwdriver to connect them (short them) for about 5 seconds
once done, switch PSU to ON position and turn on your PC, it might take while for PC to bootup, but once done (without editing bios settings), check gpu-z if bus lanes changed from x1 to x16
 
then something messed up in your bios settings, if your unfamiliar with how to configure your bios, then turn off your PC, switch PSU to OFF position, then few times press power button in front of case to discharge 5VSB from PSU capacitors (if your mainboard has power led, it will turn off once discharged), once done, located those 2 pins on your mainboard
yg05y7N.png

once located, use screwdriver to connect them (short them) for about 5 seconds
once done, switch PSU to ON position and turn on your PC, it might take while for PC to bootup, but once done (without editing bios settings), check gpu-z if bus lanes changed from x1 to x16

Did exactly what you said and this is the new GPU-Z page. View: https://imgur.com/a/HtjHm0O
.
 
x4 is better, still low
how does look bus interface load in sensors page? 3060ti doesnt have high bandwith needs so i wouldnt worry about not having x16 lanes available if bus load is low

Back to running at x1 as soon as I launch a game, View: https://imgur.com/a/iY7j7UP

Also, don't know if this is relevant but my whole PC just crashed and it has never done that before. I have a few warnings and a critical in event viewer. https://pastebin.com/SE4PWha8

Reliability History: View: https://imgur.com/a/ZGMF9gK