GTX560Ti on 100 degrees

anonymous_86

Distinguished
Aug 24, 2011
33
0
18,530
So I had bought a Zotac nVIDIA GTX 560Ti 1GB around 2 years ago. Since the first day, it was a pain in the arse. The day I bought it along with other products (I was building my machine myself with help from a guy from here, Tom's Hardware), I tried running Modern Warfare 2 and it gave a large beep after 10 minutes of gameplay, then the monitor went black and it entered sleep mode. Even restarting didn't help. So I shut the PC down completely and tried again. But the bloody thing kept doing this every time I tried to play MW2. I didn't know I could RMA the card at that time, so I just sucked it up and moved on. I worked day and night to find a solution for this problem online for three days, and finally ended up learning underclocking the damn card to run it stable enough. This resulted in lower FPS in games, but I thought at least its stable now and I decided to stick with it. Nearly two years on, I formatted my PC and lost my OC settings along with everything else. But I managed to remember the settings and again clocked the card after the format. But things had gotten worse. My card would usually rocket up to 100 degrees while playing ANY game (excluding those low-requirement ones like Driver San Fran and Need For Speed Hot Pursuit 2010). I tried to reset to the default settings but even that didn't help. I tried cleaning my CPU up and reapplied thermal paste to be sure, but all in vain. My efforts seemed fruitless and futile. So I finally decided to RMA that piece of junk. Nearly 37 days and two warnings to Zotac (of suing them for failing to deliver my replacement card under the promised 30 days policy) later, today, I finally received my new card. I was excited and plugged the card in to run the tests. I didn't even put the lid on the case and fired up Modern Warfare 2. With MSI Afterburner on-screen HUD running in background, I got the card racing from 46 degrees all the way to 99 degrees in a matter of seconds. I was terrified. Exited the game at once and ran an MSI Kombustor test (1280x800, DX11, AA=4x) and got 97 degrees as the final result. Did a GPU Burn-in (1600x900 native res., DX11, AA=4x). Final result was 99 degrees in under 7 secs. I had to exit the burn-in test as well in a hurry. So now, I'm in a dilemma- is it that Zotac has been producing faulty cards all these years, or is just that I'm the receiver of all the faulty units from them. Or maybe, like I hope, there is some other issue I am not aware of? Because this card wants to race to 100 degrees faster than my previous card.
I've read through several google results and each of them threads ended in OP RMA'ing the card. Alright. Now what would you guys recommend in this case where the card has JUST been RMA'd and its still the same story? Oh, and mind you, I'm aware of the fact that RMA units are just the refurbished products.

PC Specs:
i7-2600k @ 3.40 GHz
8GB G.SKILL Ripjaws professional RAM
Asus P8Z68 V-PRO LGA1155 @ 95W
Corsair TX750W v2 Power Supply
nVIDIA Zotac GTX 560Ti 1024MB

Stock settings as recorded by MSI Afterburner:
Core Voltage = 1025 mV
Core clock = 822 MHz
Shader clock = 1645 MHz
Memory clock = 2004 MHz
Fan Speed = Auto
(Also, its winter time and the case lid is not on)

Driver: 301.42
When I first got my card, I tried different drivers and found this version works best with almost all games I play.

Thanks for taking the time to read.
 
Solution

RobCrezz

Expert
Ambassador
Certainly sounds like a junk cooler. Is the fan speed going up when the temp goes up?

Do you have good cooling in your case? If you have no case fans you could be suffering from lack of airflow perhaps?

My 560 ti with the Asus DCU II cooler, never goes over 60'c in games when manually set at 37%, so I think its the cooler that is no good.
 

anonymous_86

Distinguished
Aug 24, 2011
33
0
18,530
I have a stock fan in the back of the case. The second fan is the one on th mobo and the third one is installed in the video card itself. I haven't purchased any after market fans since I believed I shouldn't need them as I won't be OC'ing the processor/card or anything.
 

RobCrezz

Expert
Ambassador


I would recommend having at least an intake fan in the front of the case to bring in cold air. It doesnt need to be expensive, $10 buys a decent 120mm fan to bring in fresh air. Im not saying this is the cause of your temps, but it is certainly a good idea.
 

anonymous_86

Distinguished
Aug 24, 2011
33
0
18,530
Room temperature right now is 15 degrees. My side panel is not on and I have removed all other unused slots to maintain a good airflow. My case is already filled with holes on all sides and has 4 fans in all. I cranked the voltage down to 1 volt and made sure all the wires were piled in a corner as to not obstruct the airflow inside the case. Installed the latest drivers (excluding other bloatware) after running driver sweeper in safe mode in order to clean any and all residual of previous drivers and manually deleted all the other leftovers in program files and other system folders. Turned off hardware acceleration for browsers and toggled the PhysX to CPU. Cleaned the CPU once again thoroughly and restarted the system.

Ran the tests. It idles at 45 degrees and soon as I fire the Kombustor, it shoots to 70 degrees all by itself and starts rising at the rate of 2 degrees per second. In no time, it touches 100 and stays there. And I don't think another fan will help. It'll just burn another hole in my pocket. Sigh! This is hopeless.
 

RobCrezz

Expert
Ambassador




Ok thats not unusual for kombuster to do that to a card with a "not great" cooler. Try a game again, it wont load it as heavy as kombuster.
 
Solution

Prithwi2050

Honorable
Aug 14, 2013
246
0
10,710
It happens to some cards. In my old PC I got an XFX HD 4770 which gave me great service all these years. But for the last six months or so the temperature shoots upto 120 degree Celsius when playing games. At 0% load temperature hovers around 94 degree Celsius. But I still can play everything even though the temperature is that high and it rarely restarts(once in a blue moon). I feared that it would melt my PCI slot at 120 degree Celsius. But nothing happened in the past six months. And that PC runs almost 20 hours each day cause I use it mostly for downloading stuffs to protect my new PC. I even tried BF4 on it and it ran pretty fine but for the awful frame rates. But this is no solution. I just wanted to share my experience. And by the way its not about ypur cabinet or any air-flow issue. It's the card which is at fault. This happens when there is fault with the VRAM. So there is no solution to this except to get a new card..
 
Without seeing the card I'd say the root cause is poor heatsink contact or the use of cheap thermal paste.
Just to clear up a point: You say you've changed the thermal paste, did you mean the main processor (CPU) or the graphics card (GPU) cooler?
If you took the graphics cooler off, how did you clean the old stuff off, what new stuff did you use, and how much did you apply?
 
download afterburner and click the gear next to your fan speed at the bottom that says user defined, from there click the check mark to enable a custom fan curve, now adjust your fan speed to gpu temperature.

click ok and at the bottom of msi afterburner hit enable overclock at startup. you may want to save the profile as well by hitting on of the numbers on the left side and hitting save
 

anonymous_86

Distinguished
Aug 24, 2011
33
0
18,530
Alright so all the time I was absent, I tried setting a custom curve and tried again. But the Kombustor kept pushing it to 100 degrees. At one point, I thought screw it. There was literally everything I had tried so there was nothing more left to do. I've so far tried 3 games (GTA:IV EFLC, Saints Row The Third and Saints Row II) besides CoD: MW2. And the results were surprising enough:
GTA IV EFLC: Highest temp- 70 degrees
Saints Row II: Highest temp- 70 degrees
Saints Row The Third: Highest temp- 85 degrees

Its a good news. I, for one, am still baffled as to why MW2 pushes it to 100 degrees, but I've recently tried only my favorite games, each for non-stop 3 hours and have gotten so far satisfying results. Even the FPS is minimum 70 in all of the above games. Guess thats enough for me. Thanks a lot for everyone's advices. I'm really grateful for your inputs. In the end, I've decided to not pay attention to the temperature since if its gotta blow, its gotta blow and there's nothing more I can do about it. If it blows up one more time, I'll file another RMA. I've got 5 years warranty after all. Thanks all.

PS: And yes, I was looking at GPU temprature, and not usage.