I thought this was ended when the obvious statement of "let's see how DX12 games turn out first" was made.
Seriously.
First, we have to wait until the big hardcore developers put the work in to actually utilize DirectX 12. You won't see small studios develop for it, not immediately. One reason why Mantle hasn't seen significant attention - it's not because AMD has a lower market share, it's because AMD has a lower market share COMBINED with the fact that it is difficult to develop closer to the metal. DICE is really the only developer that really crazy about the idea, Crytek still hasn't brought out the Mantle support they promised awhile ago.
With the code now getting into DX 12 and Vulkan, it still remains the big dogs promising support. Due to the complexity of developing bare metal (or closer to it), more games will use the big developer's packaged engines. UE4, Frostbite, CryEngine, and Unity will basically be the main players, for better or worse.
I doubt studios like CD Projekt RED will take to DX12 any time soon, likely not bringing RED Engine 3 to support DX 12 until at least a year or two after the fact, or more likely never. I'd love to see a TW3 Enhanced Edition with DX12 and see Cyberpunk 2077 launch with DX12 support, but I highly, highly doubt that will happen. I'd love to be wrong, oh boy would I.
The short of it is, I just don't see 12_1 being utilized much at all, not before the next generation of cards. I say this is someone who prefers Nvidia.
There might be some minor, very minor support of FL 12_1, but it's going to be very minimal and be more akin to Crysis 2's DX 11 patch-in, which is to say, inefficient and heavy-handed. It'll take a more focused and long-haul approach to get quality and efficient utilization of 12_1 features added to the major engines, likely the next-gen engines too, ones coded from the ground up with it in mind. I don't expect the current gen engines (Frostbite 3, for example) to get anything more than efficiency-based 12_0 features. Battlefield 5 or whatever is next for DICE after SW: Battlefront, will likely be carry the Guide On for Frostbite 4 with a focus on "native" support for 12_0 and 12_1.
I do suspect a Frostbite engine with efficient support for DX 12 will be first out of the gate, and more so, the first engine wholly developed focusing on DX 12 will be an engine from Dice. They helped pioneer Mantle, so they are the most prepared for DX 12. I don't think they can do much but basically code on top of Frostbite 3 to showcase some features and efficiency, but whether they code from the ground up or just more intuitively work in support into Frostbite 4, it'll be among the better examples on how to do it right.
Not saying that the engine or games based on it will be the greatest thing since sliced bread, or that netcode problems or other issues that BF games have faced will be absent and resolved, but I do suspect they'll have the lead in development.
Facebook
Twitter