GPU clock stuck at 300Mhz

[Saint Anne]

My laptop is a Dell G5 5590 with and RTX 2060 and it’s running games on 10fps that used to run just fine before. Ran heaven benchmank with GPUz open and found out the clock speed is stuck at 300 when the minimum expected should be around 960 for my graphics card… I’ve tried everything from updating drivers, messing with power settings, and even resetting my laptop and nothing works, someone pls help!

[Bad 💀 Motha]

It could be that the wall adapter PSU is faulty

[Illusion of Progress]

Saint Anne eredeti hozzászólása:

Illusion of Progress eredeti hozzászólása:

Are you using the default power adapter (PSU) it came with, or a third party replacement one, by chance?

That wattage used while being at 300 MHz is really interesting.

I haven't changed anything about it, I don't really know much about computers. Also yeah it's weird because before the gpu driver reinstall it was using like 50w and then it jumped to 215w.

Yeah, 200W on a laptop GPU seems... suspect.

I'd be questioning if either the PSU (be it the battery or charger in this case) or the GPU is bad at that point. If you've reinstalled Windows and drivers, and temperatures are good, then I'm not sure what else would lead to it being locked at 300 MHz (and reporting insanely high power draw numbers).

I asked about the power adapter (or "charger") because if you ever use a different one, if might not always suffice.

[Saint Anne]

ooh okay, I'll take it to a repair shop because I'm all out of ideas, but I'll let them know all the stuff I tried and the suggestions I got. Thank you all!!

[Bad 💀 Motha]

Komarimaru eredeti hozzászólása:

Unfortunately, you need to install the DELL drivers, not nvidias.

https://www.dell.com/support/home/en-vn/product-support/product/g-series-15-5590-laptop/drivers

Download BOTH
-NVIDIA GeForce GTX 1650/1050/1660 Ti and RTX 2060/2070/2080 Super Graphics Driver
-Intel UHD Graphics Driver

Disable all your online connections after this.

Run DDU(download it first if have not), remove Intel and Nvidia drivers that you installed for Graphics, and enable the option to prevent windows from installing drivers. Don't reboot before removals, until both are gone, then reboot.

Once rebooted, install the Intel UDH driver and restart the system.

Install the Nvidia driver and restart the system.

See if this works.

You only need to use the DELL supplies driver if you are updating the existing driver. As it will not let you use the Intel/Nvidia stand-alone drivers to do an update. However once you wipe all of the Dell supplied drivers out with DDU app under Windows Safe Mode, then you are free to use the stand-alone drivers direct from Intel and NVIDIA.

[DoctorWeed]

Saint Anne eredeti hozzászólása:

ooh okay, I'll take it to a repair shop because I'm all out of ideas, but I'll let them know all the stuff I tried and the suggestions I got. Thank you all!!

Have you managed to solve the problem? I`m having same issues. Same laptop as you.

[Bad 💀 Motha]

Most likely is a faulty GPU.

I this happen on a Dell XPS laptop before. GPU clocks stuck in 2D Mode no matter what.

I also had an Alienware R11 that had this happen to the CPU which was stuck at like 1.8Ghz no matter what. Without it boosting properly to 2.8-3.4Ghz, playing Games was mostly a no go in the same manner. Luckily the Alienware was under warranty at the time; they had to replace the Motherboard in order to replace the CPU + GPU.

[Effulgent]

Hello, are there any solution to your problem was found in repair shop?

[Hyoh]

Got almost the same problem, fixed it by applying thermal pad on the gpu vrm.

Apparently gpu vrm gets super hot from delivering power and throttle the power to cool itself down.

[OffendedPotato]

Hyoh eredeti hozzászólása:

Got almost the same problem, fixed it by applying thermal pad on the gpu vrm.

Apparently gpu vrm gets super hot from delivering power and throttle the power to cool itself down.

Reallly brother???? That fixed the problem????

[Bad 💀 Motha]

Hyoh eredeti hozzászólása:

Got almost the same problem, fixed it by applying thermal pad on the gpu vrm.

Apparently gpu vrm gets super hot from delivering power and throttle the power to cool itself down.

OK but how old was your GPU at that time?

Unless you've been rocking a GTX 9xx or 10xx series since release time, that wouldn't be a problem with newer GPUs as they wouldn't be old enough to have this kind of an issue. I've rarely ran into a scenario where a GPU has needed thermal paste or pad replacements it's entire life; before it got tossed aside and replaced anyways. If you have a need to replace thermal pads on your GPU, then they were problem damaged during tear-down at some point or the GPU already had a defect from the start, from the factory.

[_I_]

Bad 💀 Motha eredeti hozzászólása:

Hyoh eredeti hozzászólása:

Got almost the same problem, fixed it by applying thermal pad on the gpu vrm.

Apparently gpu vrm gets super hot from delivering power and throttle the power to cool itself down.

OK but how old was your GPU at that time?

Unless you've been rocking a GTX 9xx or 10xx series since release time, that wouldn't be a problem with newer GPUs as they wouldn't be old enough to have this kind of an issue. I've rarely ran into a scenario where a GPU has needed thermal paste or pad replacements it's entire life; before it got tossed aside and replaced anyways. If you have a need to replace thermal pads on your GPU, then they were problem damaged during tear-down at some point or the GPU already had a defect from the start, from the factory.

maybe gpu sag warping the pcb might separate the mosfets from the cooler enough to tear the pads
never seen it, but it might happen
pcb can flex quit a bit and still work, and the heatsinks are not very flexible

[Bad 💀 Motha]

I'd like to see a real-world example of a GPU sag ever actually killing a GPU.

In all my years, have never used a GPU sag-bracket and have yet to see a GPU die or cecum to any issues or damages because of it. A majority of users seeing GPUs die, get physically damaged, or the Motherboard get damaged or simply due to them not handling their hardware with care; and also just going it the wrong way when moving the PC tower around, shipping a PC, or moving the PC via a vehicle.

While I refuse to get a 4090; I have used all of the lower models, or older models of NVIDIA GPUs; yes even the big ones like 4080 and 3090. They are a bit heavy, yes but even on Motherboards that didn't have a steel bracket on the PCIE slot, or ever using a GPU anti-sag-bracket; not once have had a problem with such GPUs.

And again I've done thousands of PC builds over the years; as I sell, resell and do custom builds for others over the years as well, not just for myself, family, friends.

The last time I've seen a physically broken (personal experiences, not stuff seen Online) GPU or Motherboard was back when PCIE was fairly new; or back in the days of AGP.

[_I_]

not saying it has happened, just that it is possible

just flexing the pcb can be enough to separate them

[Bad 💀 Motha]

Yes I suppose under some condition it is possible; PCB flexing + heating up + cooling down + something breaking as a result.