I'd keep the 980. I think you will be fine that is a beastly card. I'm still rocking the gtx 680 and does alright on 1440p but it's starting to show its age quiet abit. Waiting to get the 980 TI first of the new year coming though.
I can only say one thing and this has me a bit baffled but star wars battlefront is still DX11.1 will be DX12 but is using the ASYNC Shaders in DX11.1 now and so far it is putting AMD in the lead but take that how you want we will have to wait and see how much more this comes to play when they unlock the DX12 mode of the game.
I have to say Nvidia themselfs admitted that Maxwell 2 AKA GTX 900 series / Titan X was designed for serial workloads and was not hardware oriented at compute performance while AMD GCN architecture was designed for years for parallel compute workloads very key for Async Shading and Async Compute. So this very well could be coming back and biting Nvidia in the rear hard we don't know completely yet but Nvidia did admit it as to being a serious issue and touted they would correct this with pascal.
Very good video on this
This one of the major things that give AMD an advantage with GCN Nvidia didn't design for parallel processing / parallel compute and ASYNC shaders are just that alot of tasks that can be done on a GPU / CPU in DX12 require a strong parallel workload even multi threading will require this allowing any GPU core to speak with any available CPU core.
A slight video on this is here to simplify it a bit
Nvidia can only allow a few requests / ques for the CPU to GPU communications at a time for preemption context switching while AMD R9 290x can do like 48 or more when it comes to preemption context switching again to put all this into perspective AMD in general just has much more computational horsepower when it comes to parallel tasks / workloads along with can communicate more with the CPU / GPU cores this is where the multi threading capability comes in it isn't all about CPU multi threading it is majorly about CPU + GPU multi threading together which comes in with DX12 these go hand in hand.
I will say this context switching is one of the most important parts of multi threading in general and for Nvidia to have issues with it is again very odd but again Nvidia was not after computational performance with Maxwell 2. So it is like having a V8 engine that runs on 4-6 cylinders every time you peg the throttle while AMD GCN cards just up and go and run with it like a proper setup V8.
Nvidia maxwell 2 or GTX 900 series / Titan X have issues with preemption context switching it drags Nvidias cards to a crawl and this is where CPU + GPU multi threading comes to life it is a hardware issue on Nvidia side that causes this no software / firmware can fix it it was simply Nvidia didn't design maxwell for compute or context switching little alone preemption context switching which is the core basis to make games multi threaded on both CPU and GPU's.
I will also say DX12 I have heard will use more Vram so whatever card you get you need a descent amount of Vram given alot of games are currently working up on 4gb of Vram usage at 1080p I would suggest something over 4gb of Vram for longevity.
I will say to the OP in general a GTX 980 TI is the better option currently for the money 6gb of Vram and current performance yes if current knowledge of DX12 is true it won't benefit much from DX12 but you wouldn't take a loss currently or a huge loss in the future over similar priced cards currently a GTX 980 TI is the most appealing card for both current and future.
i dont know why i did it or why i will do it again
GTX 980 doesn't have a "DX12 problem" and the cards are close enough in performance that the time and effort needed to exchange one for the other is going to be wasted, unless you have a very specific need.
...rightful liberty is unobstructed action according to our will within limits drawn around us by the equal rights of others. I do not add 'within the limits of the law,' because law is often but the tyrant's will, and always so when it violates the right of an individual. -- Thomas Jefferson