My gpu has been increasing usage but decreasing clock speeds while charging during gaming

[Luxeno7]

I have recently been having a problem is that when I am playing any game when charging my gpu just increases gpu usage to 80%+ and decreases the gpu clocks which gets me huge fps drops and stuttering 
My laptop is Acer nitro 5 i5 9300h gtx 1650, And I have been trying to find the solution for so long I tried using different power plans nothing and tried changing the gpu power management mode still nothing and updated my gpu drivers but nothing
and tried changing the priority in task manager but that lead into more stutters and fps drops so I have got no idea what to do

[Tegneren]

My moms laptop kept throttling the CPU while the charging. What i found out was that the power brick that came with the laptop was too small (45W). When I got her a power brick with higher wattage(60W i think it was) everything was fine. So it might be a similar issue.

[Luxeno7]

I have a 135W brick and I have recently started getting this issue

[Luxeno7]

UPDATE: I have been testing and trying to fix this and now it only drops clock speed in Fortnite so what happens is it running on 1395 Mhz 40 - 71 usage on the gpu and then when plugged in it becomes 300 Mhz 91 - 100 usage and the cpu is the same 40 - 60 on 3.8 Ghz and the ram on is 6 - 8 gb so the only thing changing is the gpu and it's getting me alot of fps drops and I am not sure what to do

[Heliian]

How are the temps?  Maybe driver update? 

[Luxeno7]

8 hours ago, Heliian said:

How are the temps?  Maybe driver update? 

90 on gpu and 91 on cpu but it's always the same in every game and gpu drivers are up to date

[Luxeno7]

Just now, Luxeno7 said:

90 on gpu and 91 on cpu but it's always the same in every game and gpu drivers are up to date

but that still doesn't explain why it boosts to 80%+ if it's gonna throttle anyways

[Luxeno7]

And its also happening in valorant but this time i am just getting lower of but not fps drops and stutters like in fortnite

[unclewebb]

18 hours ago, Luxeno7 said:

90 on gpu and 91 on cpu

Acer lowers the Intel specified 100°C thermal throttling temperature on many of their laptops to 92°C. If the CPU is running steady at 91°C, it is likely thermal throttling and continuously bouncing off this limit. 

 

90°C is also right around the thermal throttling temperature that many mobile Nvidia GPUs use. Try running GPU-Z to see what the GPU throttling temperature is set to. You can run ThrottleStop to see what your CPU PROCHOT (processor hot) thermal throttling temperature is set to.

 

Open the ThrottleStop Options window and check the Nvidia GPU box. Turn on the Log File option on the main screen of ThrottleStop and go play a game for 15 minutes or so. Make sure you experience some FPS slow downs when testing. When finished, exit ThrottleStop so it can finalize your log file. The log will be in your ThrottleStop / Logs folder. Attach a log to your next post so I can have a look. 

 

You should not expect smooth game play if both the CPU and GPU are bouncing off the thermal throttling temperature limits. When was the last time you opened up your laptop and cleaned it out? Never?

 

When the CPU or GPU start to throttle and slow down, the GPU has to work harder. A GPU at 500 MHz has to work twice as hard compared to the same GPU at 1000 MHz. An increase in GPU usage is a good sign that the GPU is overheating and throttling. It is a mobile GPU. It could also be power limit throttling. Run a GPU-Z log file and it might show the type of throttling your GPU is doing. 

[Luxeno7]

When on battery.txtWhen on AC or Charging.t

Every drop in mhz is every time the gpu spikes to 100% usage and fps drop and I don't know how to check the PROCHOT in gpu-z

[unclewebb]

@Luxeno7

If you want a specific person to notice your reply, remember to include their name in your reply. 

 

 

ThrottleStop makes it easy to see that the CPU thermal throttling (PROCHOT) temperature is indeed set to 92°C. This is much lower than the Intel recommended value of 100°C. I could never figure out why Acer does this but now that I think about it, because the Nvidia GPU has a low thermal throttling temperature, Acer likely set the CPU throttling temperature low so it does not overheat the GPU. That makes sense. Including a properly sized heatsink and fan for the CPU and the GPU would make more sense but a bigger heatsink costs money. I guess it is cheaper to low ball the thermal throttling temperatures. 

 

The 9300H has a 45W TDP rating. In a properly engineered laptop, you should be able to run a 45W CPU at 45W indefinitely without it overheating. Look at the screenshot you posted. At only 27W, the CPU is already overheating. That indicates a problem with your laptop cooling. When was the last time you opened up your laptop for some routine maintenance? Most users do not give maintenance a second thought until there are problems like you are having.

 

I would open up your laptop and blow out the dust in the heatsinks if you have not done that in recent memory. If that does not solve your poor cooling problem then you will need to also replace the thermal paste. This procedure is usually not that difficult. Watch a few YouTube videos to see how it is done. Just blowing out the dust can solve some overheating problems so try doing that first before replacing the thermal paste. 

 

There is usually a tab in GPU-Z that will specifically list the Nvidia GPU throttling temperature. It is not too important if you cannot find that info. The log file makes what is happening obvious. 

   DATE       TIME    MULTI   C0%   CKMOD  BAT_mW  TEMP   NVIDIA GPU     VID   POWER
2022-07-07  15:47:20  21.00   44.3  100.0       0   89     300    91   0.7175   11.9
2022-07-07  15:47:21  21.00   85.4  100.0       0   89     300    91   0.7151   11.7
2022-07-07  15:47:22  23.73   89.4  100.0       0   90     300    91   0.7092   14.6
2022-07-07  15:47:23  21.00   46.4  100.0       0   89     300    91   0.7092   12.0
2022-07-07  15:47:24  21.00   55.0  100.0       0   89     300    91   0.7151   12.1
2022-07-07  15:47:25  25.11   56.7  100.0       0   89     300    91   0.7151   14.6
2022-07-07  15:47:26  21.00   44.3  100.0       0   89     300    91   0.7174   11.9
2022-07-07  15:47:27  24.25   88.8  100.0       0   92     300    90   0.8370   14.6
2022-07-07  15:47:28  27.37   88.7  100.0       0   90     300    90   0.7184   16.2
2022-07-07  15:47:29  21.00   56.5  100.0       0   89     300    90   0.7157   11.9

 

The GPU reaches a maximum temperature of 91°C and just sits there at that temperature. That is thermal throttling in progress. The GPU slows down to only 300 MHz but it is still running at 91°C. It cannot increase its speed because it is already running too hot. At times your 45W CPU is only at 12W yet it is running at half of its rated speed. It too is bouncing off the thermal throttling limit. Your laptop needs much better cooling so both the CPU and GPU can run at their full rated speed. 

 

If you are going to replace the thermal paste, find something that works well in laptops. Ignore any Google reviews where they are testing desktop CPUs. A mobile CPU does not use an integrated heat spreader like desktop CPUs use. A paste that works well on a desktop CPU might work poorly or only last a week or two on a mobile CPU before temperatures start going up and up again. 

[Luxeno7]

@unclewebb
Alright but why does it do that when it's charging and not while on battery

[Luxeno7]

25 minutes ago, Luxeno7 said:

@unclewebb
Alright but why does it do that when it's charging and not while on battery

And I also do have my cpu undervolted by 100.6 mV on cpu core, intel gpu and cpu cache and my cpu is normally capped at 4.1 ghz but I capped it at 3.8ghz

[unclewebb]

23 minutes ago, Luxeno7 said:

why does it do that when it's charging

When a laptop is charging more heat is being generated inside the case. This throttling problem still happens when running on battery power. It just takes a while for the GPU to get up to 91°C. The battery log file shows throttling just starting to happen near the end of the log file. Your plugged in log file shows that the GPU is already at 91°C at the start of the log file. 

 

Mobile GPUs use lower power plans when running on battery power. That might also be why it takes the GPU longer to heat up. 

[unclewebb]

@Luxeno7

I just noticed. You did your battery testing first. This heated the GPU up to 91°C. As soon as you plugged your laptop in, it had already reached the GPU throttling temperature. 

 

There are no magic settings to fix this problem. You need to fix the cooling issues. That is the bottom line.

[Arkafreak]

On 7/1/2022 at 12:28 PM, Luxeno7 said:

I have recently been having a problem is that when I am playing any game when charging my gpu just increases gpu usage to 80%+ and decreases the gpu clocks which gets me huge fps drops and stuttering 
My laptop is Acer nitro 5 i5 9300h gtx 1650, And I have been trying to find the solution for so long I tried using different power plans nothing and tried changing the gpu power management mode still nothing and updated my gpu drivers but nothing
and tried changing the priority in task manager but that lead into more stutters and fps drops so I have got no idea what to do

This is exactly whats happening with my rtx 3050 laptop from acer nitro 5.... its rated at 75 watts but suddenly from today it srarted running ar 22 watts or 20 watts and gpu clock is stuck to 210mhz and memory clock is stuck to 840mhz and is not moving what so ever.... and my gpu is getting 99% utilization but is wayyyyy slower in performance and its temperature is around 67°C and cpu is very normal and running fine..... this issue is getting fixed by reinstalling graphics driver or by disabling discrete gpu from device manager and re-enabling it. But the probelm stays the same after consecutive cold or warm booting no matter what booting... it again becomes same and gpu power gets stuck at 22watts and core voltage gets stuck at 0.6 volts and again i have to disable enable thingy and then it works for that session.... please give a solution for the fix