False GPU Overheat Reading?

jeredgordner

Distinguished
Oct 31, 2011
4
0
18,510
My rig:
ASUS Crosshair V Formula AM3+ AMD 990FX
ASUS EAH6950 DCII/2DI4S/2GD5 Radeon HD 6950 2GB 256-bit GDDR5 PCI Express
AMD Phenom II X6 1100T Black Edition Thuban 3.3GHz, 3.7GHz Turbo Socket AM3 125W Six-Core
(I don't think the rest is relevant for this thread)

My problem:
I built this desktop about a month ago and it's been chewing up and spitting out everything I can throw at it. I've been running all sorts of high end games on the highest settings with no issues or slightest hiccup, no overclocking at all, everything stock settings. A couple days ago I started playing Dead Rising 2, seemed to hiccup a little but I associated that with the hundreds of zombies on screen at any given minute. So I turned down the zombie details and shadow details a bit and it seemed to fix the stutter.
This morning I opened up my machine just to tie some wires together to make it easier to manage inside and I changed nothing hardware or wire setting. I started to play Dead Rising today and the fans to my graphics card seemed to spool all the way up at high speed before the menu screen even loaded. Once the menu was up, the fans kept making sounds like an engine revving up and settling down. I quickly Alt+Tabbed my way out, opened up Asus SmartDoctor and it said my GPU was overheating at 104C. It normally sits around 45C with no game running, 70C with a game running. The second I turned off my game, it instantly dropped back to 45C and the fan went back to normal. So I tried it again with dual monitors so I wouldn't have to Alt+Tab again and sure enough, it INSTANTLY went from 45 to 104 with the fans revving up and down. It doesn't make any sense to me how it could double in temperature so quickly
I can't figure out if this is some sort of software issue with Asus SmartDoctor or if something could've happened with my graphics card.

Thanks in advance for any input!
 

clutchc

Titan
Ambassador
My first thought would be a failing graphics card. Or maybe the thermal compound between the GPU and cooler isn't applied correctly (??). But if that were the case, it shouldn't have worked well for a month. Just tossing out ideas. Have you checked that nothing was restricting air flow to and from the card? Could anything have worked its way under the card's cover to restrict cooling?
 

jeredgordner

Distinguished
Oct 31, 2011
4
0
18,510
@clutchc
I've inspected the card and absolutely nothing is restricting airflow. It has all the same clearance it's had for the past month and nothing is physically in it's way. I looked under the cards cover and every looks copacetic under there as well.

@greghome
There's definitely nothing on the fan, it's completely open. I suspected the same about the one game (even though I've been playing it for about a week now with none of this nonsense before) so I tried three others, same issue. I suppose you could be right about the thermometer going faulty...I certainly hope not.

Update: I've tried uninstalling the drivers and all related programs and still had no luck. Asus's technical support has been anything but helpful (either there's no technicians available or I keep getting transferred on the phone) so I'm about to try formatting my hard drive and re-installing everything from scratch. But please, keep the ideas rolling!
 

jeredgordner

Distinguished
Oct 31, 2011
4
0
18,510
@greghome

As per your prior post I downloaded GPU-Z and it gives me the exact same readings. ASUS tech support had me pull the battery to clear the CMOS and that didn't work so now they're having me download the newest motherboard chipset and re-download drivers for the MoBo and video card... I'm doubting that's going to work.