Jump to content

is 78 degrees too hot for a rtx 2060?

aren332
Go to solution Solved by manikyath,

no.

 

and for goodness sake, you do not need to repaste your GPU.

I just bought my self a used rtx 2060 evga ko ultra, and then i overclocked it. In fullscreen 1080p res furmark, is 78 degrees too hot? Do i need to repaste my gpu?

Link to comment
Share on other sites

Link to post
Share on other sites

Tis further toward oasty, but you're in Furmark not actual load (Furmark is a power virus  - does its best to pull the most power possible, not to simulate any normal load) so it's likely higher than normal, and you're still below the 80-83C the FE cards run at fine (my FE 2060 Super ran ~83C under full load). I believe the max safe temp is a good bit above that.

Intel HEDT and Server platform enthusiasts: Intel HEDT Xeon/i7 Megathread 

 

Main PC 

CPU: i9 7980XE @4.5GHz/1.22v/-2 AVX offset 

Cooler: EKWB Supremacy Block - custom loop w/360mm +280mm rads 

Motherboard: EVGA X299 Dark 

RAM:4x8GB HyperX Predator DDR4 @3200Mhz CL16 

GPU: Nvidia FE 2060 Super/Corsair HydroX 2070 FE block 

Storage:  1TB MP34 + 1TB 970 Evo + 500GB Atom30 + 250GB 960 Evo 

Optical Drives: LG WH14NS40 

PSU: EVGA 1600W T2 

Case & Fans: Corsair 750D Airflow - 3x Noctua iPPC NF-F12 + 4x Noctua iPPC NF-A14 PWM 

OS: Windows 11

 

Display: LG 27UK650-W (4K 60Hz IPS panel)

Mouse: EVGA X17

Keyboard: Corsair K55 RGB

 

Mobile/Work Devices: 2020 M1 MacBook Air (work computer) - iPhone 13 Pro Max - Apple Watch S3

 

Other Misc Devices: iPod Video (Gen 5.5E, 128GB SD card swap, running Rockbox), Nintendo Switch

Link to comment
Share on other sites

Link to post
Share on other sites

15 minutes ago, Zando_ said:

and you're still below the 80-83C the FE cards run at fine

yeah, its "fine" but afaik thats also where most nvidia gpus throttle to or under base clock? hence if a card is really stuck there,  that's not actually a good thing. personally i get a crisis if my gpu goes over 60C... thankfully it does rarely...ex: ow2 1440p 120fps ultra... 55C... 

 

agreed about fumark its dumb... doesn't actually prove anything of relevance and can (still) damage gpus that are a bit dodgy already due to high temps/power draw.

 

its basically snake oil for sellers to say "look max usage, max temps, doesn't crash!" ... doesn't mean thats necessarily true for an actual load scenario, like, i dunno,  an actual videogame instead of this awful "graphics benchmark". ¯\_(ツ)_/¯ 

The direction tells you... the direction

-Scott Manley, 2021

 

Softwares used:

Corsair Link (Anime Edition) 

MSI Afterburner 

OpenRGB

Lively Wallpaper 

OBS Studio

Shutter Encoder

Avidemux

FSResizer

Audacity 

VLC

WMP

GIMP

HWiNFO64

Paint

3D Paint

GitHub Desktop 

Superposition 

Prime95

Aida64

GPUZ

CPUZ

Generic Logviewer

 

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

2 minutes ago, Mark Kaine said:

yeah, its "fine" but afaik thats also where most nvidia gpus throttle to or under base clock? hence if a card is really stuck there,  that's not actually a good thing. personally i get a crisis if my gpu goes over 60C... thankfully it does rarely...ex: ow2 1440p 120fps ultra... 55C... 

No. Pascal and onwards will drop like... 10Mhz every 10C increment after 50C, a number that matters to benchmark scores (thus overclockers insisting on running cards cool as ice) but not practical use unless you're trying to streeeeetch how long you can keep a GPU usable. My 2060 Super ran the normal 1995Mhz at those temps fine. They won't throttle to under base clock until they hit max temps and have to do so to save from overheating.

4 minutes ago, Mark Kaine said:

agreed about fumark its dumb... doesn't actually prove anything of relevance and can (still) damage gpus that are a bit dodgy already due to high temps/power draw.

 

its basically snake oil for sellers to say "look max usage, max temps, doesn't crash!" ... doesn't mean thats necessarily true for an actual load scenario, like, i dunno,  an actual videogame instead of this awful "graphics benchmark". ¯\_(ツ)_/¯ 

Like Prime95 smallFFT for CPUs, it does have some use. "Can my cooling handle a surprise power virus workload without going kaboom?". Amazon's New World RPG killed a couple 3090s by springing a surprise power virus load on them, IIRC it cause the VRAM to overheat and die. It's not unreasonable to want to make sure your cooling handles the worst surprises thrown at it.

Intel HEDT and Server platform enthusiasts: Intel HEDT Xeon/i7 Megathread 

 

Main PC 

CPU: i9 7980XE @4.5GHz/1.22v/-2 AVX offset 

Cooler: EKWB Supremacy Block - custom loop w/360mm +280mm rads 

Motherboard: EVGA X299 Dark 

RAM:4x8GB HyperX Predator DDR4 @3200Mhz CL16 

GPU: Nvidia FE 2060 Super/Corsair HydroX 2070 FE block 

Storage:  1TB MP34 + 1TB 970 Evo + 500GB Atom30 + 250GB 960 Evo 

Optical Drives: LG WH14NS40 

PSU: EVGA 1600W T2 

Case & Fans: Corsair 750D Airflow - 3x Noctua iPPC NF-F12 + 4x Noctua iPPC NF-A14 PWM 

OS: Windows 11

 

Display: LG 27UK650-W (4K 60Hz IPS panel)

Mouse: EVGA X17

Keyboard: Corsair K55 RGB

 

Mobile/Work Devices: 2020 M1 MacBook Air (work computer) - iPhone 13 Pro Max - Apple Watch S3

 

Other Misc Devices: iPod Video (Gen 5.5E, 128GB SD card swap, running Rockbox), Nintendo Switch

Link to comment
Share on other sites

Link to post
Share on other sites

78c is, well, not ideal, but not bad actually. A repaste will only drop 2 or 3c, depending on your thermal paste, so I don't think it'll worth the hassle if you're not used to it.

 

Is your fan spinning at maximum or full speed?

 

One way to reduce your GPU temperature is to open your side panel. Another option is to increase airflow of your front fans. PC Case is also pretty important here, so without a clear picture, I can't pin point if we can make any improvement.

 

But over all, You shouldn't worry if the temperature stay 78c and don't go any higher. Your GPU cooling is still adequate.

I have ASD (Autism Spectrum Disorder). More info: https://en.wikipedia.org/wiki/Autism_spectrum

 

I apologies if my comments or post offends you in any way, or if my rage got a little too far. I'll try my best to make my post as non-offensive as much as possible.

Link to comment
Share on other sites

Link to post
Share on other sites

24 minutes ago, Zando_ said:

No. Pascal and onwards will drop like... 10Mhz every 10C increment after 50C

yeah, thats how nvidia boost works.

 

24 minutes ago, Zando_ said:

My 2060 Super ran the normal 1995Mhz at those temps fine

hmm ok, but my 3070 will be at ~ 1915mhz at ~80C (i think more like 78 or so, some games it actually gets that hot without my undervolt) which may still be "base clock" but an actually noticeable difference to 2010mhz at 55C (even if it not that big of a difference)

 

i mean basically what im talking about is this , but unfortunately its for a 1000 series card, i suppose newer cards may throttle later, but probably not by much and its something most people would want to avoid obviously. 

clock_analysis2-4.jpg.a01f7eb43ffc5386d8ce22417fb6ae5e.jpg

 

theres a really hard cutoff at ~83C , wish i would find the same graphic for newer cards, i just always assumed it would be basically the same more or less 

 

24 minutes ago, Zando_ said:

Can my cooling handle a surprise power virus workload without going kaboom

yeah... the question is... do you want to know that... lol... i always feel really uneasy running p95, and i don't do it anymore,  it is actually useful for testing stability of cpu and ram overclocks in my experience though (although there are obviously a few alternatives) 

 

 

16 minutes ago, Chiyawa said:

One way to reduce your GPU temperature is to open your side panel. Another option is to increase airflow of your front fans. PC Case is also pretty important here, so without a clear picture, I can't pin point if we can make any improvement.

undervolting does wonders,  10-15C on my 3070, problem is, while simple in theory it can be quite challenging and needs some experience/ practice with Msi afterburner for sure.

 

19 minutes ago, Chiyawa said:

But over all, You shouldn't worry if the temperature stay 78c and don't go any higher. Your GPU cooling is still adequate.

yeah, i agree, its still ok, even though i personally would really rather want to avoid this, the cooler a pc runs the better  = ) 

The direction tells you... the direction

-Scott Manley, 2021

 

Softwares used:

Corsair Link (Anime Edition) 

MSI Afterburner 

OpenRGB

Lively Wallpaper 

OBS Studio

Shutter Encoder

Avidemux

FSResizer

Audacity 

VLC

WMP

GIMP

HWiNFO64

Paint

3D Paint

GitHub Desktop 

Superposition 

Prime95

Aida64

GPUZ

CPUZ

Generic Logviewer

 

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

21 minutes ago, Mark Kaine said:

hmm ok, but my 3070 will be at ~ 1915mhz at ~80C (i think more like 78 or so, some games it actually gets that hot without my undervolt) which may still be "base clock" but an actually noticeable difference to 2010mhz at 55C (even if it not that big of a difference)

Never really saw an improvement from similar clock jumps (1995Mhz to 2100Mhz) on my 2060 Super in actual use, and it was less stable in stuff like Star Citizen that really doesn't like OCs even if they're stable in other games, so I've never kept an OC running. Point less being about overclocking, more that that clock difference is not that big. Getting like 8 more fps doesn't really help unless you were struggling to get 60fps to begin with (in which case you probably need to adjust your settings or get a better GPU).

 

And again, this is under FurMark, game temps should be a good bit lower, OP shouldn't have anything to worry about.

37 minutes ago, Chiyawa said:

Is your fan spinning at maximum or full speed?

^^^ may be more fan curve to be had if you don't mind the noise.

Intel HEDT and Server platform enthusiasts: Intel HEDT Xeon/i7 Megathread 

 

Main PC 

CPU: i9 7980XE @4.5GHz/1.22v/-2 AVX offset 

Cooler: EKWB Supremacy Block - custom loop w/360mm +280mm rads 

Motherboard: EVGA X299 Dark 

RAM:4x8GB HyperX Predator DDR4 @3200Mhz CL16 

GPU: Nvidia FE 2060 Super/Corsair HydroX 2070 FE block 

Storage:  1TB MP34 + 1TB 970 Evo + 500GB Atom30 + 250GB 960 Evo 

Optical Drives: LG WH14NS40 

PSU: EVGA 1600W T2 

Case & Fans: Corsair 750D Airflow - 3x Noctua iPPC NF-F12 + 4x Noctua iPPC NF-A14 PWM 

OS: Windows 11

 

Display: LG 27UK650-W (4K 60Hz IPS panel)

Mouse: EVGA X17

Keyboard: Corsair K55 RGB

 

Mobile/Work Devices: 2020 M1 MacBook Air (work computer) - iPhone 13 Pro Max - Apple Watch S3

 

Other Misc Devices: iPod Video (Gen 5.5E, 128GB SD card swap, running Rockbox), Nintendo Switch

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, Chiyawa said:

78c is, well, not ideal, but not bad actually. A repaste will only drop 2 or 3c, depending on your thermal paste, so I don't think it'll worth the hassle if you're not used to it.

 

Is your fan spinning at maximum or full speed?

 

One way to reduce your GPU temperature is to open your side panel. Another option is to increase airflow of your front fans. PC Case is also pretty important here, so without a clear picture, I can't pin point if we can make any improvement.

 

But over all, You shouldn't worry if the temperature stay 78c and don't go any higher. Your GPU cooling is still adequate.

fan is not spinning at max speed

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×