The card definitely needs to have its thermal paste redone.

What other temps are you experiencing in your rig? What is the ambient air in the case? Is this the only component getting hot?  

As long as you don't physically damage the card, you will be covered under warranty still.

That's the great thing about EVGA's warranty - they even let you replace the cooler (say, with an AIO water cooler or a water block) and it won't void your warranty.

http://www.evga.com/support/warranty/

Edit: this video should help you with repasting. It's not the exact same GPU, but it should be very a similar process. https://youtu.be/N-brze9p_9E

wmmills
" From what I can tell I run at 900 clock, 3500 memory clock, 1012V and it has 80C at 1800 RPM in Idle"   That right there says either Precision is false reading, the case your using doesn't have adequate airflow...especially if your using that sr2 and dumping all that heat in the case and running it through the gcards acx classified cooler. how are those fans setup in the case? Where do you live and what are the ambients?

Hey
Sorry thats old sig... I'm using i7 + gtx titan. I7 idles at 35-40C. I started using NVidia inspector to force idle on card with 4 screens connected, that droped the idle temps to 60C and 400 gpu/memory clocks. A bit better but still hot for idle. 
 
The case is quite well vented, I have 140mm water cooler corsair on top for cpu + 3 140 or 120mm fans around pomping air in/out. 

This your second GPU ? (two earlier "other threads" mention)
 
MB? Is it OC ?  does it crash when running stock speeds ?
 
4 screens -  what resolution ?
 
What happens with 1 monitor ? 2 Monitor ?
 
What software is running in the background ?
 
All drivers and software up to date ?  (sometimes older GPU drivers work better)
 
Looks like some pro software being used; is your hardware listed by publisher as compatible / recommended ?
 
- Reboot PC
    running stock MB & GPU settings, use OC software to force faster fan speed on GPU
    go offline - remove internet connection
    Only run 1 of your software programs, shutdown all other software even security software - what happens ?
 
Stock Titan X = Core Clock -1000 MHz Boost Clock - 1075 MHz & Memory Clock - 1753 MHz
techpowerup TitanX
 
techpowerup EVGA TitanX SC
GPU-1127, Boost-1216, Mem-1753
 
 
Your running the memory at 3500 MHz  ? = making a lot of Heat
 
 
Why Re - TIM ?
 
1) Sometimes the factory TIM application is less than perfect
 
2) Sometimes the TIM can get dry and less thermally conductive
 
3) You get to inspect / replace the thermal pads
 
4) Faster than RMA the card if you do it, have it done locally
 
 
My experience, on air the GPU should be around +35-50 C warmer than ambient air temp, depends on many things
 
 
I'm running my EVGA Titan X SC at GPU 1440 MHz, Mem 3300 MHz, mV 1274, with water blocks, 24/7 with 90% load - folding @40 C
 
 

Cool GTX
This your second GPU ? (two earlier "other threads" mention)

MB? Is it OC ? does it crash when running stock speeds ?

4 screens - what resolution ?

What happens with 1 monitor ? 2 Monitor ?

What software is running in the background ?

All drivers and software up to date ? (sometimes older GPU drivers work better)

Looks like some pro software being used; is your hardware listed by publisher as compatible / recommended ?

- Reboot PC
running stock MB & GPU settings, use OC software to force faster fan speed on GPU
go offline - remove internet connection
Only run 1 of your software programs, shutdown all other software even security software - what happens ?

Stock Titan X = Core Clock -1000 MHz Boost Clock - 1075 MHz & Memory Clock - 1753 MHz
techpowerup TitanX

techpowerup EVGA TitanX SC
GPU-1127, Boost-1216, Mem-1753


Your running the memory at 3500 MHz ? = making a lot of Heat


Why Re - TIM ?

1) Sometimes the factory TIM application is less than perfect

2) Sometimes the TIM can get dry and less thermally conductive

3) You get to inspect / replace the thermal pads

4) Faster than RMA the card if you do it, have it done locally


My experience, on air the GPU should be around +35-50 C warmer than ambient air temp, depends on many things


I'm running my EVGA Titan X SC at GPU 1440 MHz, Mem 3300 MHz, mV 1274, with water blocks, 24/7 with 90% load - folding @40 C

 
1st GPU I got RMA within a day, since shop is 5 min away they replaced it quickly. Titan X (might be SC I think it says SC in precision)
No OC, no CPU no GPU. I work on my system so I prefer it stable. I used to have 3.8ghz oc on CPU but that was via multiplier and I stopped it awhile ago.
Same issue with 1-2-3 monitors. Tried it in past. I also get interferences. I have attached a mockup screnschot of what it looks like. My monitors simply flash this way from time to time. At random over all of them. Tried different cables/ports/no help
Software can be any. But standard BG... avast, wacom, chrome
Drivers usually mostly up to date. My windows 10x64 is around 5months old without updates. There were some nasty updates lately introduces so luckily I missed them. Thank god to block windows 10 update apps lol.
Yeah my hardware is being listed as supported on softwares. My hardware is no different to 10 other workstations that are around me.
  

Cool GTX
- Reboot PC
running stock MB & GPU settings, use OC software to force faster fan speed on GPU
go offline - remove internet connection
Only run 1 of your software programs, shutdown all other software even security software - what happens ?

 
I get crashes with just runing chrome on its own in chrome. Its pretty random. Its why its so blody hard to diagnose. Shops usually just return it and charge me for false RMA because they just run stability apps and then blame it on something else.
 

Cool GTX
Why Re - TIM ?

1) Sometimes the factory TIM application is less than perfect

2) Sometimes the TIM can get dry and less thermally conductive

3) You get to inspect / replace the thermal pads

4) Faster than RMA the card if you do it, have it done locally


My experience, on air the GPU should be around +35-50 C warmer than ambient air temp, depends on many things


I'm running my EVGA Titan X SC at GPU 1440 MHz, Mem 3300 MHz, mV 1274, with water blocks, 24/7 with 90% load - folding @40 C
 

 

I'll try to replace paste when I get EVGA 1080ti either FTW3 or SC1/2 I don't want to have down time. I need my workstation hehehe Other than that I never seen those clocks/temps on my Titan X SC. I think I tired to OC it within 2 weeks of getting it. I believe after +25mghz boost on core it was crashing and then I give up. Just though it was a bad batch of chip for OC.
 
 
 
 

Dave3d
It sounds to me like the fans arent kicking in.
And, trust me, you need the fans at top speeds on these titans.
If not, they burn out early.

1, Get rid of evga precisionx. Just do it. Uninstall and reboot. Its junk software.
2.Install nvidia drivers, ONLY what you need. Dont install the crapware that comes with it.
3. After reboot, set video settings set power management prefer maximum performance.
4. Install MSI Afterburner. Set STEEP fan curves. These titans get hot, and the vrms and memory need to get cooled, and the regular setting isnt good enough. 70% start is good, ramping up to 100% fan speed. Woulsd you rather have noise, or a messed up card? Dont forget to set OSD > More > Custom Direct3d = On. Reboot, and whatever else you need for your software.
5. Dont forget, that every time you update drivers, you need to uninstall the msi afterburner (or evga precisioncx if thats what you want), then install drivers, then install msi/evga.

As for multiple monitors, you are on your own. These cards are bad enough with just 1 dispayport running, let alone more than 1. :(

Hey
1. Will give it a go thanks!
2. Yeah I do it already
3. Will think about it. I prefer to be more "green" than power...
4. Will give it a go. Damn this card is loud then !
5. Never know that thanks! will do it in future.
Nvidia inspector power managment seem to fixed the issue. It force GPU to low speed, I'm down to 60C.

Thanks all for your time info !
 
 

My advice would be if you're planning to keep the card for a bit, upgrade to "at minimum" a hybrid cooler if not a a full cover waterblock. I learned with my 780 Tis that air cooling is just not for me anymore. Too much noise, stability issues in the summer due to heat (even when not overclocked), etc. and with elite cards like these, you want them to be properly cooled. I've been running two Hydro Copper Titan Xs since Summer of 2015 and they've been rock solid stable @ 1300MHz in SLI. I've never had a crash. If you think you're tired of the Titan, then just dump it and pick up one of the new 1080 Ti models, preferably a hybrid model or one with upgraded cooling over the stock cooler.
I love my Maxwell Titan X cards - ready to upgrade now as I've been GPU limited for sometime, but these have been great cards. Wish you the best my friend, hope you get things working properly.

Dave3dI disagree, and here is why: I had a stock Titan X, it lasted 6 months, spent another 7 months getting it to work until it just kind of gave up and slowly died.
I then rma'ed, and got a Corsair kit with an h90, and although the die temps are fantastic, the vrms and memory still get too hot, even on my extra fan speed curve and its been going out slowly after 3 weeks of getting the new rma, the same way the old one did. Seriously thinking about rma'ing this one also, as it exhibits the same thing the last Titan X did.
My 780GTX runs the 2560x1440 144hz gsync monitor ten times better than the titan x does. There is something wrong with these crads, no matter what you do to them. They bend easily at the power cord also, and need something to keep them level. These are just bad cards.
My advice? RMA it, sell it, and buy a 1080TI instead.

If you had a stock Titan X, I'm honestly not too surprised. It's common knowledge that the stock cooler is significantly insufficient for the application. I remember reading the reviews on both the Maxwell Titan and the Pascal Titan and reviewers said with the stock cooler, the cards would begin throttling clocks back almost immediately under heavy 3D load due to overheating. I knew after that if I ever upgraded to one of the Titan cards it would have to be liquid cooled, and now I liquid cool all my equipment no matter what for noise anyway and it's worked out well. Sorry to hear about the issues you had.

Dave3d
Well, I do have my 2nd card on water, but not a full water block, just the gpu.
And its still dying.
:(

Strange stuff - is it the GPU that's dying or other parts of the card? I've popped the VRMs on previous cards in times before full cover waterblocks did a good job of covering them.