2080Ti stuck on 300Mhz Core Clock under load

[Vya Domus]

Nvidia cards are known for this issue, it started happening ever since Kepler after they added heinous power sensing circuits and monitoring, basically for whatever reason the card senses that it's drawing too much power and throttles itself into the ground. There used to be a way to fix it but I don't think it works anymore, here you can try it : https://www.techpowerup.com/forums/...gpu-with-high-power-tdp-on-idle.223300/page-2

Another potential fix would be to try a shunt mod but probably it will do nothing.

 

[zeohi]

bit of an odd follow up since we're nearly half a year later, but i tried plugging the card back in and everything worked fine.... for a month or 2. but now it's back to stuck at 300mhz. can anyone give me any possible reason as to why sitting in a box for 6 months made the card work as normal for a little while longer?

 

[oldwalltree]

bit of an odd follow up since we're nearly half a year later, but i tried plugging the card back in and everything worked fine.... for a month or 2. but now it's back to stuck at 300mhz. can anyone give me any possible reason as to why sitting in a box for 6 months made the card work as normal for a little while longer?

this sounds like a driver issue? I would try and roll back drivers until i found one that worked.

 

[zeohi]

this sounds like a driver issue? I would try and roll back drivers until i found one that worked.

ok i can try this, but between it wroking to not working i never updated graphics drivers or anything

 

[trog100]

some mining software can do this.. a clean driver install will fix it.. i have two machine running nicehash miner..i cant run anything 3D on them without doing a clean driver install first..

trog

 

[Falkentyne]

bit of an odd follow up since we're nearly half a year later, but i tried plugging the card back in and everything worked fine.... for a month or 2. but now it's back to stuck at 300mhz. can anyone give me any possible reason as to why sitting in a box for 6 months made the card work as normal for a little while longer?

This might be an easy fix or you need to send the card to someone who can repair this stuff (there's someone on Hardforum (the old HardOCP forums) who can do this, I think his name is Razorwind, as he's been helping some other people on the video card sections. If it's a damaged solder trace or BGA rework, you need to get someone to fix it for you. But if it's just a very dirty PCB or foreign component on the PCB, you may be able to fix it yourself.

Buy some 99% isopropyl alcohol. You're probably going to need about a gallon of it. Half a gallon "may" be enough.

Buy some replacement thermal pads. Unfortunately you're going to have to do some research on what thicknesses you need. If you can't find this information, you should buy some Arctic (6 w/mk) 0.5mm, 1.0mm and 1.5mm pads (145mm * 145mm), and then do some jigsaw experimenting and pressure testing with temporary single drop of toothpaste or single drop of mayonnaise (on the GPU core), UNPOWERED, to check for contact pressure and correct pad spacing. The toothpaste or mayo is to save wasting thermal paste unnecessarily until you determine exactly the pads you need.

Disassemble the card completely, and clean all thermal paste and remove all thermal pads from the card.

Douse the entire PCB in 99% alcohol. Leave it submerged for about 10 minutes.

Remove the PCB, let it air dry (do not use a hairdryer, isopropyl is flammable), then reapply new thermal pads and thermal paste, AFTER you determine the correct thickness of each section of thermal pad area.

If there isn't a defective soldering joint or chip or IC damage on the card, this should fix it.

Thermal pads are here but the price keeps changing literally every hour.

https://www.amazon.com/dp/B00UYTTXSM/

Thermal paste recommendations (Highly suggested):

https://www.amazon.com/gp/product/B079DQS77Q/