[SOLVED] RTX 2080 Gigabyte temps go up too high

[Oxicoi]

Hi,

I have a decently old RTX 2080 from Gigabyte, the typical Windforce. When first purchased, it never used to act out of the ordinary. However, about a year ago, it started to lose control of its' temperatures. When idle, it'll be around 40-50C (which is a bad sign). Then when using it with full utilization, it will jump way above 80c, passing 85c. The fans, I noticed, jump well above 100% fan speed to cool it down, but it doesn't. So, whatever is using the GPU I have to manually force close on Task Manager.

I was told to reapply thermal paste as it could be the issue, but I have done so and the issue still continues. I've seen others say it may not be making contact with the heat sink, but when opening it, the paste does spread.

Not exactly sure what to do here. Should I just RMA or get a GPU waterblock or something?

 

[Oxicoi]

I reapplied thermal paste again. I noticed this time it takes way longer to get up to 85C, but still gets there. When the fans boost up, it goes fast enough to keep it at 85C. So does this mean the fans are bad? The paste?

 

[Oxicoi]

So, in another post I've made, I asked for what a good GPU paste could be. Hopefully this will solve the issue, but if it doesn't, maybe someone will know? Who knows...

 

[theonerm2]

How are you testing? 85C is a normal temperature for a GPU to reach. But under normal conditions the exact Graphics card you're using could be anywhere from 68c to 80c under normal use. It all depends on a few factors. 1 being blockages. Dust can block your airflow and keep your graphics card from being cooled. Also the case can be another issue. If your case doesn't get good airflow it can cause everything inside to get hotter than it should. It probably isn't the fault of the thermal paste as long as there is enough and there isn't a huge gob of it under there. As long as you know how to put thermal paste on or followed some kind of guide you should be good on that. Also the fans could be going bad that's another common issue that can cause the part to overheat. Usually the problem isn't the paste although it can improve temperatures to put good quality or even better than factory quality paste on the GPU die.

 

[David0ne86]

Hmmm out of curiosity did you try undervolting?

 

[Oxicoi]

Hmmm out of curiosity did you try undervolting?

I do not know how to do that. I have used MSI Afterburner many a'time, but never have I seen a negative percent on voltage. Any option for that?

 

[Oxicoi]

How are you testing? 85C is a normal temperature for a GPU to reach. But under normal conditions the exact Graphics card you're using could be anywhere from 68c to 80c under normal use. It all depends on a few factors. 1 being blockages. Dust can block your airflow and keep your graphics card from being cooled. Also the case can be another issue. If your case doesn't get good airflow it can cause everything inside to get hotter than it should. It probably isn't the fault of the thermal paste as long as there is enough and there isn't a huge gob of it under there. As long as you know how to put thermal paste on or followed some kind of guide you should be good on that. Also the fans could be going bad that's another common issue that can cause the part to overheat. Usually the problem isn't the paste although it can improve temperatures to put good quality or even better than factory quality paste on the GPU die.

Tested on games, tested on mining, tested by itself. Idle seems normal (sorry if I said it wasn't before) after testing it many times. There is nothing normal about this GPU. A slight task will spike the temps up to 50-60C. Usually, a GPU would gradually get there.

No blockages such as dust, that's been dealt with. Don't have a case, I'm open chassis (which I have much better temps with, I have tested). I don't know about the fans, but they do spin and when putting them to 100% on MSI Afterburner, they do speed up like any other ordinary GPU.

I could be pasting it wrong, but I do the X way in a timely manner so I don't put too much, too less.

I wish I could just send it to someone to make them understand this annoying problem lmao

 

[hotaru.hino]

Some things of note:

• Idling at about 50C is expected if you're not running any kind of fan controller software (e.g., MSI Afterburner with a custom fan profile). A lot of cards now employ a zero-fan mode and the GPU will get to about 50C idling.
  • If you have more than one monitor plugged in, if they're not set up with a refresh rate divisible by each other (e.g., 60Hz and 120Hz or 72Hz and 144Hz), the GPU will idle at a higher power state
• 85C is a normal ceiling for a video card.
• I believe the most recommended TIM application on bare dies is to spread it yourself using a plastic spudger or credit card. Although if you can verify whatever you're doing is actually spreading the paste completely, then it shouldn't matter.
• How sudden does spike happen? If it's a few seconds, then that may be expected (although I need to figure this out with my own card)

 

[Oxicoi]

Some things of note:

• Idling at about 50C is expected if you're not running any kind of fan controller software (e.g., MSI Afterburner with a custom fan profile). A lot of cards now employ a zero-fan mode and the GPU will get to about 50C idling.
  • If you have more than one monitor plugged in, if they're not set up with a refresh rate divisible by each other (e.g., 60Hz and 120Hz or 72Hz and 144Hz), the GPU will idle at a higher power state
• 85C is a normal ceiling for a video card.
• I believe the most recommended TIM application on bare dies is to spread it yourself using a plastic spudger or credit card. Although if you can verify whatever you're doing is actually spreading the paste completely, then it shouldn't matter.
• How sudden does spike happen? If it's a few seconds, then that may be expected (although I need to figure this out with my own card)

It doesn't idle at 50C thankfully. It idles around the 30's like my 2070 Super does. The fans are always at 100% as I like the coolest possible. I do not have multiple monitors.

85C is a normal ceiling? Sure, but I don't want to keep hearing a leaf blower every time I play a game when it's trying to cool itself. It shouldn't get up there for every GPU-tasking program.

The paste does indeed spread on the entire die.

Less than a second for the spikes, sometimes longer. Usually lower though. Say I open GPU accelerated programs like Discord or Blizzard, it'll spike it, but then lower.

 

[hotaru.hino]

It doesn't idle at 50C thankfully. It idles around the 30's like my 2070 Super does. The fans are always at 100% as I like the coolest possible. I do not have multiple monitors.

85C is a normal ceiling? Sure, but I don't want to keep hearing a leaf blower every time I play a game when it's trying to cool itself. It shouldn't get up there for every GPU-tasking program.

I'm confused. You want to keep it at 100%, but you don't want the computer sounding like a leaf blower?

 

[Oxicoi]

I'm confused. You want to keep it at 100%, but you don't want the computer sounding like a leaf blower?

It goes more than 100% when at 85C. It goes even faster than that.

 

[hotaru.hino]

It goes more than 100% when at 85C. It goes even faster than that.

Fans at 100% tend to sound like a leaf blower, is what I'm implying.

 

[Oxicoi]

Fans at 100% tend to sound like a leaf blower, is what I'm implying.

Of course. This GPU, however, likes to go beyond that lmao

 

[Oxicoi]

Could this at all be any reason for high temps?





The pads are messed up somewhat and dust is on them.

EDIT: All the heat majorly comes from the die itself, or in that general area, so wondering if the thermal pads are bad. I just applied NT-H2 though, thanks to a helpful person from another post, and was able to get the temps even lower than it was before. I got a bad batch of NT-H1 as it was hard to push out and was very stiff looking. H2 helped a ton though. Currently running a mining program and it seems to settle at 40's, when previously it was in the 60's.

EDIT 2: This is what it does all the time (it'll jump from 50% to 100%, then 50% to 100% [this is with mining]):



EDIT 3: After testing the GPU with Cold War with all graphics on the highest possible, RTX on, native 1080p but 100x upscale, it stayed in the 70's and fans don't go at 200%. Am I good now? Finally?

 

[David0ne86]

Those temps look perfect to me. It's classic cool gpu temps (70s). Yeah i think you did it. 😁 If you're looking to shave a few more, I'd look into undervolting. But those temps are perfectly fine.

As far as utilization during mining i have no idea I'm no mining expert. So I'll leave that to others more adept to it

 

[Oxicoi]

Those temps look perfect to me. It's classic cool gpu temps (70s). Yeah i think you did it. 😁 If you're looking to shave a few more, I'd look into undervolting. But those temps are perfectly fine.

As far as utilization during mining i have no idea I'm no mining expert. So I'll leave that to others more adept to it

Could please tell me how to under-volt? I would definitely love to shave more off, in technical terms (not physical) lol.

 

[Phaaze88]

Could please tell me how to under-volt? I would definitely love to shave more off, in technical terms (not physical) lol.

Imma just copy-paste this from another thread:
Undervolting is TEDIOUS to do. It's going to be easier to just open Afterburner and run with a lower power limit... but that also means lower performance.
I've been playing around with it for a few weeks on my 1080Ti. Here's the 'short version'.
-Msi Afterburner and Gpu-Z(for the Sensors tab) are needed.
-Afterburner has save slots for profiles. Make use of them.
-With Afterburner and Gpu-Z open and running in the background, play your games. Simple as that. The goal is to find out how much voltage(max) your gpu is asking for.
That will be your 'starting point'. Whatever that number is, write it down or memorize it.

-Go to Afterburner's Curve Editor. A window will pop up containing the Voltage/Frequency curve for the gpu. You will see several points on said curve. This is all the voltage control Nvidia has allowed for.
-You can left click any point and lock the voltage there with the L key, as well as raise/lower the frequency from that point with the Up and Down keys.
-Recall the max voltage requested by the gpu, and find it on the chart, then select and click on the point to the left of it. [You can go to points further left if you wish.]

You'll notice as you go further left, the boost frequency gets lower. That's not the worst of it though.

I could just end it right here and it would still keep things somewhat simple: Curve Editor, experiment with points lower than the max requested, lock it, click Apply on the Main HUD, and go game with a little less performance...
But you probably want to offset some of the lowered boost frequency so as to reduce some of the performance hit, and this is where it becomes a pain in the butt - and Gpu Boost doesn't help, because it 'whines' over every little thing:
-Every 5C or so warmer that the gpu runs(excluding thermal limit)? "Imma throttle the boost frequency back a little."
-Don't have enough voltage to run at a particular frequency? "Imma throttle the boost frequency back a little."
-Keep bumping into the power limit? "Yeah..."


-If you're going to increase the frequency, do in increments of 10-15mhz.
-Test the frequency bumps for stability across as many games/benchmarks as you are willing. The more samples you run, the better.
You can use Unigine Heaven, but honestly, it is out of date - it doesn't even have DX12 support.
You can run Unigine Superposition, but it is very limited with the free version. If you don't want to spend extra, the first suggestion for testing with as many samples as possible is the best.
EXCLUDE Furmark and Kombustor. They do not run like actual games, and are better suited for testing the capability of the gpu cooler - they're like Prime95(small FFT, all AVX off), but for gpus.

-You basically keep raising the frequency - if you even get that far - until stuff starts crashing, and you dial it back to the last stable increase.

~And, I think that's it?
Unfortunately, I don't believe any set numbers exist, because of silicon lottery and Gpu Boost.
Memory clock? Nah, leave that one alone. Frequency is all the control we have on this one. Lowering it does not help temperatures: the voltage is fixed, and it doesn't pull much power to begin with.

 

[David0ne86]

Just watch a couple of YouTube tutorial. But yeah it's a tedious trial and error process. I wouldn't do it since as i said your temps are perfectly fine now, but again it's your rig. Just do some browsing and some video research on yt. With an AMD card would be definitely easier.

 

[Oxicoi]

Imma just copy-paste this from another thread:
Undervolting is TEDIOUS to do. It's going to be easier to just open Afterburner and run with a lower power limit... but that also means lower performance.
I've been playing around with it for a few weeks on my 1080Ti. Here's the 'short version'.
-Msi Afterburner and Gpu-Z(for the Sensors tab) are needed.
-Afterburner has save slots for profiles. Make use of them.
-With Afterburner and Gpu-Z open and running in the background, play your games. Simple as that. The goal is to find out how much voltage(max) your gpu is asking for.
That will be your 'starting point'. Whatever that number is, write it down or memorize it.

-Go to Afterburner's Curve Editor. A window will pop up containing the Voltage/Frequency curve for the gpu. You will see several points on said curve. This is all the voltage control Nvidia has allowed for.
-You can left click any point and lock the voltage there with the L key, as well as raise/lower the frequency from that point with the Up and Down keys.
-Recall the max voltage requested by the gpu, and find it on the chart, then select and click on the point to the left of it. [You can go to points further left if you wish.]

You'll notice as you go further left, the boost frequency gets lower. That's not the worst of it though.

I could just end it right here and it would still keep things somewhat simple: Curve Editor, experiment with points lower than the max requested, lock it, click Apply on the Main HUD, and go game with a little less performance...
But you probably want to offset some of the lowered boost frequency so as to reduce some of the performance hit, and this is where it becomes a pain in the butt - and Gpu Boost doesn't help, because it 'whines' over every little thing:
-Every 5C or so warmer that the gpu runs(excluding thermal limit)? "Imma throttle the boost frequency back a little."
-Don't have enough voltage to run at a particular frequency? "Imma throttle the boost frequency back a little."
-Keep bumping into the power limit? "Yeah..."


-If you're going to increase the frequency, do in increments of 10-15mhz.
-Test the frequency bumps for stability across as many games/benchmarks as you are willing. The more samples you run, the better.
You can use Unigine Heaven, but honestly, it is out of date - it doesn't even have DX12 support.
You can run Unigine Superposition, but it is very limited with the free version. If you don't want to spend extra, the first suggestion for testing with as many samples as possible is the best.
EXCLUDE Furmark and Kombustor. They do not run like actual games, and are better suited for testing the capability of the gpu cooler - they're like Prime95(small FFT, all AVX off), but for gpus.

-You basically keep raising the frequency - if you even get that far - until stuff starts crashing, and you dial it back to the last stable increase.

~And, I think that's it?
Unfortunately, I don't believe any set numbers exist, because of silicon lottery and Gpu Boost.
Memory clock? Nah, leave that one alone. Frequency is all the control we have on this one. Lowering it does not help temperatures: the voltage is fixed, and it doesn't pull much power to begin with.

Yeah, never mind. Technology is just getting too far isn't it?

Just watch a couple of YouTube tutorial. But yeah it's a tedious trial and error process. I wouldn't do it since as i said your temps are perfectly fine now, but again it's your rig. Just do some browsing and some video research on yt. With an AMD card would be definitely easier.

You're right on that. In the future I'll do that when I get my degree. Off I go to leave this GPU do its own thing.