Hi,
There has been a lot of press releases and discussions about both computer and API performance jump in the next generation. Do you believe it and why?
I am asking both in philosophical and technical thoughts on these issues.
(Overdue if true, in my opinion)
Personal life and your private thoughts always effect your career. Research is the intellectual backbone of game development and the first order. Version Control is crucial for full management of applications and software. The better the workflow pipeline, then the greater the potential output for a quality game. Completing projects is the last but finest order.
by Clinton, 3Ddreamer
I am in the DX12 EAP... and yes, a correct D3D12 usage could give a very nice - critical - performance boost in certain (not so uncommon) usage scenarios. Keep in mind also that current state of drivers, API and OS are fare away from be considered final. Unfortunately most of technical details are still under NDA, but I guess more will be unveiled at the upcoming GDC ^_^
There are a lot of "small" changes that are being hyped as major advancements - IMHO
I'm a bit "behind" the times - the last thing I actually cared about was Java 1.8
I cannot remember the books I've read any more than the meals I have eaten; even so, they have made me.
~ Ralph Waldo Emerson
I'm pretty sure there is not much philosophy in performance.
I am asking both in philosophical and technical thoughts on these issues.
"I AM ZE EMPRAH OPENGL 3.3 THE CORE, I DEMAND FROM THEE ZE SHADERZ AND MATRIXEZ"
My journals: dustArtemis ECS framework and Making a Terrain Generator
. 22 Racing Series .
Philosophically, the main PC API's have been long overdo for some lipposuction to get rid of alot of that fat. Looks like its shaping up to be a real dozy with whats in the works.
There has been a lot of press releases and discussions about both computer and API performance jump in the next generation. Do you believe it and why?
This exact quote could be repeated every few years since the mid-1970s.
Yes I believe it, because of about 40 years of history.
Yes, DirectX12/Mantle/glNext will ring in some huge improvements on the PC side -- 50% performance improvement *easily*, 100% improvement or more in renderers/games architect ed to take full advantage of it, due to the much-improved multi-threading situation and how that impacts the simulation architecture and timing in typical games today. This essentially means that games that were once CPU-bound, due to the architecture required to get the most from GPUs, and necessary synchronization, are no longer so -- the "rendering thread" is no longer a 'long-pole' holding up progress on 3 or more other CPU cores.
On the Visual Fidelity side of things, PCs being unfettered by draw call numbers is a big deal, but perf improvements as a result of this will mostly owe, I think, to obsoleting all the nasty hacks and tricks that are required today to make a few thousand draw calls look good. The lower draw-call overhead is great--one could, in theory, keep jumping through hoops, use < 3k draw calls, and simply take the free perf given by lighter draw call overhead--but I think greater gains are to be had by rethinking draw-call strategy in the new paradigm. But it we do start using 10x or 20x or 50x draw calls as we do today on PC, we'll eventually eat up the CPU again, we'll just look prettier doing it.
And of course GPU-bound scenarios will mostly remain so -- D3D12 might gain some ground due to finer grained resource management and re-use on the GPU side, and it might alleviate some scenarios that currently require extraneous buffers/copies or round-trips to the CPU and back, but at the end of the day not even a secret pact with Satan will squeeze any more FLOPS from the shader cores.
What I don't agree with is the many fanboys postulating that D3D12 is going to bring similarly large performance benefits to Xbox One, which is something I see all-over places where armchair developers congregate. It'll be positive for many reasons, some of them (smaller) performance reasons, even -- but D3D12 won't be the end of PS4's performance advantage.
throw table_exception("(? ???)? ? ???");
You'd need some awesome API design for making up for that hardware difference :D
What I don't agree with is the many fanboys postulating that D3D12 is going to bring similarly large performance benefits to Xbox One, which is something I see all-over places where armchair developers congregate. It'll be positive for many reasons, some of them (smaller) performance reasons, even -- but D3D12 won't be the end of PS4's performance advantage.
"I AM ZE EMPRAH OPENGL 3.3 THE CORE, I DEMAND FROM THEE ZE SHADERZ AND MATRIXEZ"
My journals: dustArtemis ECS framework and Making a Terrain Generator
You'd need some awesome API design for making up for that hardware difference
Yep, no matter what happens, PS4 simply shipped with better specs 
What I don't agree with is the many fanboys postulating that D3D12 is going to bring similarly large performance benefits to Xbox One, which is something I see all-over places where armchair developers congregate. It'll be positive for many reasons, some of them (smaller) performance reasons, even -- but D3D12 won't be the end of PS4's performance advantage.
It won't be the end of PS4's advantage, but hopefully it will help to close the gap.
GNM is an amazing API (as you'd expect), which makes D3D11 look like a Canyonero. The Xbone's D3D11.x can try and bolt on bare metal features all it likes, but it's doomed to just be The Homer. D3D12 will help the Xbone somewhat reduce the gap that it has with the PS4.
As for the fanbois, they probably don't realize that the Sony consoles have always had ridiculously bare metal APIs, and that Sony's GNM was probably the inspiration behind Mantle.
. 22 Racing Series .
