I recently bought a GTX 970 SC ACX 2.0 and have also been having the same issues after about 30 minutes to an hour of gameplay on Far Cry 4 and Battlefield 4. With the consensus apparently being to get a replacement card i guess i will be doing that too  
Although, I will be getting a replacement from Amazon because I bought it there and its still under the warranty. Would anyone recommend an RMA directly to EVGA vs. a replacement form Amazon?

Try the card in an other system, if possible. this is by far the best way to check it. it is possible that there are software issues causing major bluescreens "blackscreens" but they are unusual. 
with changes hundreds of settings you will in the end just waste your time. 

eVGA seems to have a good rma policy, faulty cards can happen, i can wait 2-3 days. 

Hello,
 
In case there are new readers, that are having the same issue as mentioned above. I used this forum as a troubleshooting guide. I did everything mentioned by Morphe and Malboja. In short, my experience was this: Played Dragon Age: Inquisition and would have video crashing within 1hour of game play. I would have no video display and have to hold down my power button to cold-restart my computer. After a day or two this was becoming an issue an multiple builds and every config with BIOS and NVIDIA drivers known to man. Oddly enough, nothing else (that I tested) was effecting my 970 in the same way, only DA:I. I could play demanding games like Far Cry 4 and not so demanding like WoW without any problems at all.
 
My DVI port didn't work (worthy of an RMA) and I was getting some board noise when using headphones. (not 100% sure it was the card at the time, but tested with my 660ti without an issue), and finally, the video crashing. I decided to give it one last go, and see how long it would take my game to crash while I had a day off of work. Within 5 minutes, everything crashed. Which scared me since it was so sudden compared to the 1hour windows I was getting before. Only this time, I wouldn't get any video even after the cold-reboot. I cried for about 2 minutes and replaced my 970 with my old 660ti.
 
After RMAing the card, and having to wait over both Christmas, and New Years for my RMA to be finished (over 3 weeks sadly, but totally my fault for not taking action more quickly), two of three issues SEEM to be resolved. No more crashing for Dragon Age (so far), and no more dead DVI! However I am still getting some feedback noise, which only started when using my 970 card. My 660ti is fine. But that I can deal with, since it's not terrible.
 
I wanted to post on this thread because I wanted to let other readers who might use this like myself, that an RMA did in fact resolve the issue.
Only thing I can come up with is that we are all customer who purchased the same card around the same time, and have 1155 socket CPUs. Coincidence?

I purchased my gtx 980s mid February and BOTH cards exhibiting this behavior . Bought mine from newegg . Nothing works to fix the issue . Swapping cards pci ...switching to pcie 2.0 ....down clocking ...nothing ...every time I play Guild wars 2 It will be fine for a length of time then my tv will just stop receiving signal ...pc will still be running but no signal to pc . I'm still in game as confirmed by my friend I play with and can even type by hitting enter ..typing ....and hitting enter he sees me type ....it's been a running gag for my guild ...I spend 1100.00 on two gtx 980s and can't even play a 2 year old game . Thinking of trading up to two titan x in sli and see if that resolves my issue if evga could hurry up and release the darn things

how is it not? my two xfx r290x cards ran flawlessly before i replaced them with these ....

from what ive seen folks who decide to rma the card have had mixed results but most (read like 80 %) have reported that the issue has been resolved with an rma ... hopefully mine will be as well ... then like i said ... if evga gets on the ball with their Titan X card i plan on trading up for two Titan X's in SLI and hopefully these cards do not present the same issues 

smallsbig
the truly funny part is that i can run firestrike no problem but when i boot up gw2 maybe 2 minutes ... maybe 30 ... but the thing will crash 

I had issues with my GTX 460 SLI in CoH2 and even after I bought a GTX 970, I still had the issue at least one.  The issue was that the drivers would crash and send me into a bluescreen which would be black because of the drivers crashing and it took me about a year to figure out it was a bluescreen.  One day, it happend and I left my PC under the black screen for about 30minutes to go eat, came back and windows notified me about sending the report to Microsoft which I assure you I never would.  But I know its not my PC because I have looked at the dump files using Microsofts SDK program and they all point to NVIDEA drivers crashing and going haywire.  So I know there drivers are not solid so to speak.  Luckily I don't really play the game anyways so it's not dear to my heart that I fix it or ask for it to be fixed.  To say NVIDEA drivers don't have issues would be a mistake though and I am willing to defend that they do have several flaws w/ stability.  
 
Overall, what im saying is I would not be surprised if it is not a driver related issue if you can stress test and play other games fine but this praticular game you play crashes soon after launch(10-20min). 

smallsbig
I purchased my gtx 980s mid February and BOTH cards exhibiting this behavior . Bought mine from newegg . Nothing works to fix the issue . Swapping cards pci ...switching to pcie 2.0 ....down clocking ...nothing ...every time I play Guild wars 2 It will be fine for a length of time then my tv will just stop receiving signal ...pc will still be running but no signal to pc . I'm still in game as confirmed by my friend I play with and can even type by hitting enter ..typing ....and hitting enter he sees me type ....it's been a running gag for my guild ...I spend 1100.00 on two gtx 980s and can't even play a 2 year old game . Thinking of trading up to two titan x in sli and see if that resolves my issue if evga could hurry up and release the darn things

I had the same issue as the OP.  I had the card (980 FTW) not even 1 day and it crashed and crashed and crashed.  I too bought it from Newegg and did not return it to them for fear I would get another of the same batch; it had the old BIOS based on a nVidia message I saw on their forum from back in November.  I think Newegg probably still have some old stock of these cards.
 
Just an FYI.  I did not do any diagnostics.  I refused to waste my time.  The only diagnostics I did was remove the 980, reinstall the 2x 680s and call EVGA for a replacement.  My 2x 680s ran smoothly, so I knew it was the card.  I contacted EVGA on Friday, March 13, 2015 and I had my new card on Friday, March 20, 2015.  They took care of me and I now have a card that has been flawless so far - max boost 1502.  I am glad I stuck with EVGA cards this second go round as I've always heard that their customer service was top notch.  Thumbs up to them.
 
Al

I'm actually starting up the step up process right now due to the RMA card being sent to me doing the same thing only not at the PC crashing level, sadly I cannot get the GTX 980 SC card and have to get the normal version in ACX while the card i'm using is a GTX 970 SC edition, I really wish they would let us upgrade to the card we want since we're paying the difference in the first place.

MrDonoman
I'm having problems as well with the 970 ftw acx 2.0.  I played skyrim for a while without the mods, had good fps constantly, and no problems.  Now I have some mods that are putting my card to the test, and I'm at around 50-60 fps, but now I randomly lose the HDMI signal to my TV.  This has never happened to me before in any other game ever since I got this 2-3 weeks ago, and now it's starting to act up.  Is there anyone that has solved the problem without returning the card?  It's my only card and I need to be at WoW raids every week, so RMA would be my absolute last option :'(.

You can cross ship RMA your card, you have to pay up front for the replacement card they send it and then you send yours back and they refund what you paid.

Reading this (helpful) thread again it looks almost certainly a hardware issue, I already started the RMA process after a WHOLE lot of testing.
It looks as if that mainly SC ACX2.0 are affected by this. My personal guess is that it has to do with on-board components of the cards, possibly the voltage regulators getting stressed too much (?) in some scenarios. Best example certain scenes in Heaven Benchmark. It goes black, then the PC reboots with black screen, the card "vanishes" from the system until hard-reset.

Wow! I was so happy with my card, Mine is the GTX 970 SC acx 2.0 model, It does run games just fine, But after running Unigine Heaven I got the infamous black screen! I ran 3dmark after, and it did fine until the sky diver bench then the black screen. But when I downclock it does fine on both benchmarks! You know this is absolutely ridiculous that so many cards are doing this! When a company makes a overclocked card they are supposed to make sure they use the better chip to make sure it stays stable! But apparently they just threw these cards together and overclocked them without caring a bit about quality! This makes me so mad that I am going to have to pay for shipping during a RMA when they are the ones at fault! First no backplate and now this! I can assure this will be my last EVGA if they make me pay shipping! 
 
EDIT: going to try a clean windows install first, But with the way it acted fine in unigine after downclocking it, I doubt this will help, But it might be possible even though I ran DDU, it was where I had a AMD gpu installed before this one. 

Ok she is fine! I got Windows 8.1 re-installed, And I installed the new driver. and I ran 3DMark Sky Diver and Unigine Heaven and it ran flawless! And a better score! (Thank God) So whatever it was, Seemed to be making it crash after the card would down clock from GPU boost. Not sure what it could have been, But this has to be a driver conflict of some kind. Now I have to re-install GTA V and War Thunder and Watch Dogs, LOL,, But I suppose it is better than a RMA.

staticus712

flexy123
Increase your FAN SPEED.
 
It's a problem with components (VRMs, NOT your GPU) overheating.
 
I can avoid black screens running my fan at 47%. But obviously this should not happen at stock speeds.

 
I have had this problem with mine for a few weeks and I can verify this works, although this is something you shouldn't have to do at all.
 
Setting the fan speed to Auto causes the black screen crashes, but if I force the fan speed to 40% or above it doesn't crash, or at least it hasn't for the last few hours.

You are 100% correct, It is something that should not have to be done, However my temps never got high, Mine barley got to 70c, And that was why I was so confused, I figured it out though, My problem was software related, It had to have been something left over from my AMD software Because after I did a clean format and re-install of windows 8.1, The card now longer crashes. I figured it out after I noticed it was crashing not when stressed, It was crashing after being stressed, as soon as the GPU boost would clock itself back down the driver would crash causing the blackscreen. Maybe you all should try a format/re-install, But if you are having temp issues, I would say it's a different problem, If your case has good cooling and it is still getting too warm I would RMA. I would have liked to figure out what was causing the conflict, But I am a impatient idiot, And I did not feel like spending hours on it. But it would have helped others.