GPU overheating? Time to replace?

[NomarFachix]

Hey all,

So recently my MSI N460GTX GPU has been heating up to ~60ºC while playing low demanding games (WoW, specifically).. when it gets up around 55º, the screen blanks out and then appears again.. this happens a few times before coming back up and there is a constant sound that comes from the speakers and the screen is jarbled colors. It remains this way until I power off the PC.

I have the case open with an oscillating fan blowing inside, and the A/C keeps my apartment around 70ºF, so I know it's not an airspace problem. After the problem occurs, touching the copper tubes on the GPU is very hot.

Is this definitely a GPU overheating problem? If so, is this something that could be RFA'd (I bought the GPU last April/May..)? Are there any solutions? Any input that you guys could throw my way would be greatly appreciated.

Thanks!

 

[Homeles]

It's not overheating, it's dieing. My card idled at those temps.

So yes, definitely RMA or replace the card.

 

[digitaldurandal]

Those temperatures are fine.

Did you change anything about your system? Software or hardware? Can we know your specs (PSU)? Is the card still under warranty?

 

[NomarFachix]

Those temperatures are fine.

Did you change anything about your system? Software or hardware? Can we know your specs (PSU)? Is the card still under warranty?

I hadn't changed anything that I recall. For a while leading up I had been getting notifications that my drivers stopped responding.

PSU is an Antec 430W.

Warranty for labor is 2 years and parts is 3 years.. would those cover what I'd need? I'm assuming so.


-So after reading what I just wrote, I went ahead and uninstalled the drivers and then re-installed the GPU. I'll see if that makes a difference?

 

[NomarFachix]

Alright, so that most definitely wasn't it. Went into WoW, sat still with graphics on high, and got two blackouts and then a BSoD, which read something like "the video _____ failed" and that's all I could read before the PC restarted.

Frustrated.

 

[NomarFachix]

Downloaded Afterburner and lowered the core clock and memory clock to min, increased the fan speed to 65, and everything has been copacetic so far.. was able to play DA:O beyond the character creation screen for the first time.

 

[NomarFachix]

Final update before bed:

GPU came factory overclocked. Lowered the clocks and raised the fan, and was able to play several games for extended time without any problems at all. Had to lower graphics a bit to even out the framerate, but still.. uninterrupted gameplay.

But all of that said, still means that my GPU is shitting out, right? I guess I'll contact MSI tomorrow and see if I can get an RMA set up.

 

[Yuriman]

Yeah it's crapping out on you, but if it works fine I (personally) would be content lowering the clocks until it's stable. Reapplying thermal compound might help.

 

[Magic Carpet]

Clean the GPU and the area around the socket w/ isopropyl alcohol, reapply TIM. Let it dry.

OR, RMA if you have warranty.

 

[NomarFachix]

Just got off the phone with Nvidia (who apparently was the wrong person to call, MSI is who I need to get an RMA from) and the service rep was actually very resourceful and tried to help as much as he could..

Ended up saying that the power supply was not optimal for the card (I run a 430w, he says the minimum for the card is 460w) and that may be the problem. Told me to upgrade my power supply and try the card there, and if that doesn't solve the issue, to contact MSI about a PSU upgrade.

I'm not sure that's the problem, as the card ran perfectly for the first year and a half that I had the card in there.. would it suddenly stop getting as much power as it needed from the PSU? Should I skip the PSU upgrade and go straight to the MSI RMA, fibbing that I'd already tried the PSU upgrade?

Please advise.

 

[AnandThenMan]

I think you should replace your power supply either way it is marginal at best.

 

[fleshconsumed]

Your computer draws probably 300W max under full load, 430W should be perfectly fine. You can borrow one from a friend or something to test to be sure. Or perhaps buy this one as a spare/test

http://slickdeals.net/permadeal/784...d:+SlickdealsnetFP+(SlickDeals.net+Frontpage)

 

[Zxian]

PSU is likely not the culprit in terms of raw power supplied, but out-of-spec voltage rails can lead to odd behaviour. Video cards can be particularly sensitive to this.

If you can test your system with another PSU from a friend, do that. If the problem then persists, it's time for RMA.

 

[aigomorla]

1. have you tried replacing the TIM? Tim gets old over time, and with the constant heat and cool down cycles while gaming and then sleeping, it can go bad.

2. Have you inspected for dust bunnies inside the sink?

3. I see you prop'd case open and tried to feed your case cold air... however sometimes this doesnt help, as you can push the hot air more to your GPU.
Are you sure your GPU isnt pulling any hot air?


for me a card doesnt start feeling like its dying until i see artifacts.
Once i see artifacts, i start shouting loud cuz words, while trolling on the forums that this vendor sucks.... eh... so i guess u can tell im human...

 

[aaksheytalwar]

Isnt 60c cool for a 460?

 

[chimaxi83]

Isnt 60c cool for a 460?

It's a great temperature for any air cooled GPU.

 

[NomarFachix]

I sent this to MSI at the beginning of the month, and it's still "in replacement, testing, or repair process". Very frustrating, going to be about a month without a GPU.