[SOLVED] RTX 2070 problems with screen, and temperature issues.

[Lunanescence]

I started getting temperature issues when i found out that the GPU idles at 50C and under load it hits 88C. , and the performance was sluggish. I replaced the paste for the first time since buying it,
and then i started getting cooler temperatures for a while, until last week, it started idling at 50-60C again and still maxes out at 88C even on full fan speed.
(i used to OC the card too btw, but its been factory since then.)

Earlier today i hopped on Flight Simulator, and the GPU maxed out at 88C, and i saw artifacting for a split second, and then my screen went black and turned off.
When i turned back on my screen, it was just black, and then the computer just restarted itself.
After it restarted i opened up MSI to check the temps, without realizing for some reason, i had a unstable OC preset saved, and it was loaded. (before this whole thing happened, it was literally on factory settings)

During that whole time i didn't even realize it
I went to turn the preset off to default, AND IT WAS STILL DOING IT, at this point i thought i might have actually damaged the card somehow, but i was also trying to press keys to see if the screen would somehow turn on lmao.
but even after switching back to factory settings, it still kept doing it. I tried other titles, even down to the basic ones like Minecraft and it still kept doing it.
At one point, my screen had like a massive tear in the center and it looked like it was screen tearing really badly. It went away after a while, but i get a lot of screen flickering overall.

is my card on life support now? and i definitely don't have any warranty

system specs:
CPU - Ryzen 9 3900X
MOBO - ASUS ROG STRIX X570-E
RAM - G.SKILL TRIDENT Z 2X8GB (16GB) @ 3200MHZ
GPU - ASUS DUAL RTX 2070 8GB OC
PSU - EVGA SUPERNOVA G3 750W
and i have a soundcard under the GPU, SoundBlaster AE-7)
I'm also running at 1440p

 

[ak195]

Considering your PSU is fine and there's no voltage problem from its side and your CPU and Mobo is fine too.
Remove all GPU software and remove drivers also and install a fresh one and check it. Otherwise install clean Windows so that no preset, registry etc. remains.
If issue persist then you can try this GPU in another system.
If same thing happens there, then your GPU may become defective. If you want to keep it then use heavy cooling solution, but I think it will not work better for high end gaming.

 

[Lunanescence]

Considering your PSU is fine and there's no voltage problem from its side and your CPU and Mobo is fine too.
Remove all GPU software and remove drivers also and install a fresh one and check it. Otherwise install clean Windows so that no preset, registry etc. remains.
If issue persist then you can try this GPU in another system.
If same thing happens there, then your GPU may become defective. If you want to keep it then use heavy cooling solution, but I think it will not work better for high end gaming.

I've asked on numerous other forms, and let me also add, that now it artifacts and crashes the whole computer after 5 minutes from being on desktop.
someone else said that the memory module is busted.
however, i don't have warranty so i'll be kind of screwed

 

[Phaaze88]

What paste did you use? It actually matters more with gpu dies.


Your situation seems to be related to bad memory OC, or overheating memory. The latter is fixable, but the former? Doubtful.
So at some point, you enabled the option for Afterburner to apply your OC settings at startup? That's not on by default. [I don't know if that's different for a beta version...]
And the memory was overclocked in said profile, right? Memory OC is iffy - I don't do it anymore because of scenarios like this. Gpu core clock is far more forgiving.

Try ak195's suggestions.

 

[Lunanescence]

What paste did you use? It actually matters more with gpu dies.


Your situation seems to be related to bad memory OC, or overheating memory. The latter is fixable, but the former? Doubtful.
So at some point, you enabled the option for Afterburner to apply your OC settings at startup? That's not on by default. [I don't know if that's different for a beta version...]
And the memory was overclocked in said profile, right? Memory OC is iffy - I don't do it anymore because of scenarios like this. Gpu core clock is far more forgiving.

Try ak195's suggestions.

I had left-over corsair paste, and it seemed to do the job for a little bit and then it just went back to the same thing

also in terms of the MSI afterburner situation,
i was pressing random buttons when the screen was black, I guess i turned on a preset by accident and i wasn't even realizing it until i had looked over and saw it, and thought that might have been why it was crashing.
after i turned it off, it just still kept doing it.

Everything was in stock settings, no OC on anything.

i got the screen flickering, half of the screen was blue, and then at one point it would look like the middle of the screen was screen tearing really badly.

 

[Phaaze88]

I've not heard much about Corsair's paste.
-A lack of, or uneven mounting pressure, combined with thermal expansion... the more runny pastes end up shifting off the die and leaving hotspots. Thicker and stickier pastes are more resistant to this.
-Some pastes are runny by design, and others become runny from high(er) temperatures, where the components that actually conduct/transfer heat end up around the die leaving mostly oil(harmless) on it.

Again, I'm not familiar with the Corsair paste, so I'm not sure where it falls in this equation, but it likely wasn't the best paste to use here.

also in terms of the MSI afterburner situation,
i was pressing random buttons when the screen was black, I guess i turned on a preset by accident and i wasn't even realizing it until i had looked over and saw it, and thought that might have been why it was crashing.
after i turned it off, it just still kept doing it.

Everything was in stock settings, no OC on anything.

It might have been too late by that point, or a mere coincidence. I suggest a clean driver install with Display Driver Uninstaller if you haven't already.
-Download the latest Nvidia driver package - but don't install it yet.
-Download DDU from here: https://www.wagnardsoft.com/ and run it in Safe Mode.
-When you open DDU, select to uninstall the graphics driver, and then Clean and Restart.
-Once you're back in, install the driver package > Custom install w/o Geforce Experience > Clean driver install(restores all Nvidia Control Panel settings to default).
If at the very least, this doesn't help, then the hardware itself is to blame.
Heck, I'd even include trying the gpu in another PC if possible, as well as another monitor cable - if these don't help, it's definitely the card.

 

[Oxicoi]

I started getting temperature issues when i found out that the GPU idles at 50C and under load it hits 88C. , and the performance was sluggish. I replaced the paste for the first time since buying it,
and then i started getting cooler temperatures for a while, until last week, it started idling at 50-60C again and still maxes out at 88C even on full fan speed.
(i used to OC the card too btw, but its been factory since then.)

Earlier today i hopped on Flight Simulator, and the GPU maxed out at 88C, and i saw artifacting for a split second, and then my screen went black and turned off.
When i turned back on my screen, it was just black, and then the computer just restarted itself.
After it restarted i opened up MSI to check the temps, without realizing for some reason, i had a unstable OC preset saved, and it was loaded. (before this whole thing happened, it was literally on factory settings)

During that whole time i didn't even realize it
I went to turn the preset off to default, AND IT WAS STILL DOING IT, at this point i thought i might have actually damaged the card somehow, but i was also trying to press keys to see if the screen would somehow turn on lmao.
but even after switching back to factory settings, it still kept doing it. I tried other titles, even down to the basic ones like Minecraft and it still kept doing it.
At one point, my screen had like a massive tear in the center and it looked like it was screen tearing really badly. It went away after a while, but i get a lot of screen flickering overall.

is my card on life support now? and i definitely don't have any warranty

system specs:
CPU - Ryzen 9 3900X
MOBO - ASUS ROG STRIX X570-E
RAM - G.SKILL TRIDENT Z 2X8GB (16GB) @ 3200MHZ
GPU - ASUS DUAL RTX 2070 8GB OC
PSU - EVGA SUPERNOVA G3 750W
and i have a soundcard under the GPU, SoundBlaster AE-7)
I'm also running at 1440p

Sounds like a "doppelganger" problem.

1440p already makes this card run harder than it normally would on 1080p. The RTX 2070 is the best for 1080p. Running a resolution up would increase it by close to 10 degrees Celsius guarantee.

I had a RTX 2080 once... and still do. I once created a thread here and had the support of @Phaaze88. He suggested I get a certain thermal paste. That thermal paste fixed the mysterious and magical problem of the temperatures.

No matter what overclock or downclock I had, no matter what settings I lowered, no matter what system it was put in, it would still have the same temp issues. It would spike to blank temperature, that temperature, blah blah blah. It would rise immediately to the 70's and sound like a rocket in the 80's.

Now the temperatures gradually go up like normal instead of spiking like usual. Sure, it'll jump sometimes, but not spike. With that said, I propose a question...

Would you like to know that thermal paste?

 

[Lunanescence]

Sounds like a "doppelganger" problem.

1440p already makes this card run harder than it normally would on 1080p. The RTX 2070 is the best for 1080p. Running a resolution up would increase it by close to 10 degrees Celsius guarantee.

I had a RTX 2080 once... and still do. I once created a thread here and had the support of @Phaaze88. He suggested I get a certain thermal paste. That thermal paste fixed the mysterious and magical problem of the temperatures.

No matter what overclock or downclock I had, no matter what settings I lowered, no matter what system it was put in, it would still have the same temp issues. It would spike to blank temperature, that temperature, blah blah blah. It would rise immediately to the 70's and sound like a rocket in the 80's.

Now the temperatures gradually go up like normal instead of spiking like usual. Sure, it'll jump sometimes, but not spike. With that said, I propose a question...

Would you like to know that thermal paste?

If it can save it , please LOL

 

[Oxicoi]

If it can save it , please LOL

Introducing the wonders of the paste that was given by the savior and lord savior and thermal paste master and [you get it]

We haveeeeeeee...... drum roll

Spoiler: The Magic Paste Sauce

NT-H2! fireworks, cheering, farting (what?)

This is the gift of knowledge given to me by @Phaaze88. The question now remains...
Will you try it?

If so, do so with the X method. By that, I mean make an "X" on the GPU die. Do not put too much though.

 

[Lunanescence]

Introducing the wonders of the paste that was given by the savior and lord savior and thermal paste master and [you get it]

We haveeeeeeee...... drum roll

Spoiler: The Magic Paste Sauce

NT-H2! fireworks, cheering, farting (what?)

This is the gift of knowledge given to me by @Phaaze88. The question now remains...
Will you try it?

If so, do so with the X method. By that, I mean make an "X" on the GPU die. Do not put too much though.

definitely gonna check it out.

also ITS ALIVE

i tested it out on another system, and re-updated the drivers, and it just works fine now!

now its just the temperature problems,

Thanks for helping out

Also, my local micro-center only sells the thermal grizzly kryonaut stuff and some extra thermal pads.
might just go pick up the new pads tomorrow.

 

[Oxicoi]

definitely gonna check it out.

also ITS ALIVE

i tested it out on another system, and re-updated the drivers, and it just works fine now!

now its just the temperature problems,

Thanks for helping out

Also, my local micro-center only sells the thermal grizzly kryonaut stuff and some extra thermal pads.
might just go pick up the new pads tomorrow.

Lucky you for having a local micro-center. As for the grizzly kryonaut, it isn't bad. I never used it, so I don't want to suggest it.

Thermal pads are at your own risk as I do not know what the sizes are that are needed for your specific GPU. You'll have to reference by pictures or measure them yourself and get the correct ones and place them in the correct spots.

No problem. We all like to help (more me than others I suppose).

 

[Phaaze88]

I don't like Kryonaut because of its silly little quirk - burns out quicker with higher operating temperatures - but I do understand I'm not the target audience for it either.
Better options exist for the money.

Thermal pads can be more trouble than they're worth.
Different sizes, different texture(soft, firm), as well as there being no universally agreed upon testing medium for the ones produced.
If you haven't torn any, I'd suggest leaving them alone.