Jump to content

Recommended Posts

Posted

Here's a confusing, but logical aproach. ;)

Depends on what you plan to do with the data.

 

Shut down to prevent damage? Stop O/C to prevent damage?

- Be paranoid enough and trust the higher number.

 

Want higher speeds and higher O/C - doesn't care about damaga?

- Trust the lower number.

 

Normally the datacollector from the GPU-specific software should collect it from the correct place.

Other more 'generic software' may try to use other and incorrect conversion methods. In this case Smart doctor is the 'generic software'.

- Assume ATI knows what they are doing, it's not ASUS product and temp-probes.

- Trust ATI temp-probe.

 

ATI temp-probe wins with 2 vs 1.

The mind is like a parachute. It only works when it's open | The important thing is not to stop questioning

Posted

More likely the hotter of the two. :) I've been struggling to keep my GPU heat just under 70C during game play (Nvidia 470 GTX). I'm mostly successful, but last night something got borked with the game and my GPU was stuck at 100% usage the whole session. Spiked the heat up to 81C! Started to chug quite a bit. It stayed like that till I restarted too. Exiting and restarting the mission didn't fix it.

Posted

I would trust ati software more than asus for gpu.

 

But to double check it, install rivatuner, that will give you the temps and better control of your video card.

Posted
Mine goes up to 88 when fully loaded when weather is hot. These things can take it to almost 100 and still work.

 

My old GTX 280 used to simply die at 75C which it hit a lot due to bad cabling. Now I have a new case and what everyone agrees to be a much hotter running card but I'm still afraid to break that 70C barrier. heh

Posted

75oC ???

 

Nvidia gpu's dont slow down/throttle themselves til they hit 100 - 120oC if I recall correctly, so it shouldnt of died at 75oC.

 

Even at 100oC all that would of happened is the card would of underclocked itself which would then reduce the temps until it got below the safe temp margin and then it would of went back up to stock clocks.

 

The only reason I can see that happening is if the heatsink to it wasnt put on correctly, but on the whole nvidia cards dont throttle themselves to protect from overheating til atleast 100oC, what it could of been was maybe the memory chips on the card were overheating since they just have passive heatsinks on them, so maybe the ram chips were overheatign and causing your problem ?

Posted

Usually it's overheating of the memory on the card that causes it to crash, rather than the GPU getting too hot.

 

Long story short: temps are irrelevant. You just want to keep it as cool as possible. If you're suffering instability due to overheating, it's time to upgrade your cooling.

[sIGPIC][/sIGPIC]

DCS A-10C: putting the 'art' into 'warthog'.

(yes, corny. Sorry.)

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...