Friday, September 16th 2016

AMD Actively Promoting Vulkan Beyond GPUOpen

Vulkan, the new-generation cross-platform 3D graphics API governed by the people behind OpenGL, the Khronos Group, is gaining in relevance, with Google making it the primary 3D graphics API for Android. AMD said that it's actively promoting the API. Responding to a question by TechPowerUp in its recent Radeon Technology Group (RTG) first anniversary presser, its chief Raja Koduri agreed that the company is actively working with developers to add Vulkan to their productions, and optimize them for Radeon GPUs. This, we believe, could be due to one of many strategic reasons.

First, Vulkan works inherently better on AMD Graphics CoreNext GPU architecture because it's been largely derived from Mantle, a now defunct 3D graphics API by AMD that brings a lot of "close-to-metal" API features that make game consoles more performance-efficient, over to the PC ecosystem. The proof of this pudding is the AAA title and 2016 reboot of the iconic first-person shooter "Doom," in which Radeon GPUs get significant performance boosts switching from the default OpenGL renderer to Vulkan. These boosts aren't as pronounced on NVIDIA GPUs.
Second, and this could be a long shot, but the growing popularity of Vulkan could give AMD leverage over Microsoft to steer Direct3D development in areas that AMD GPUs are inherently good at - these include asynchronous compute, and tiled-resources (AMD GPUs benefit due to higher memory bandwidths). AMD has been engaging aggressively with game studios working on AAA games that use DirectX 12, and thus far AMD GPUs have been either gaining or sustaining performance better than NVIDIA GPUs, when switching from DirectX 11 fallbacks to DirectX 12 renderers.

AMD has already "opened" up much of its GPU IP to game developers through its GPUOpen initiative. Here, developers will find detailed technical resources on how to take advantage of not just AMD-specific GPU IP, but also some industry standards. Vulkan is among the richly differentiated resources AMD is giving away through the initiative.

Vulkan still has a long way to go before it becomes the primary API in AAA releases. To most gamers who don't tinker with advanced graphics settings, "Doom" still works on OpenGL. and "Talos Prinicple," works on Direct3D 11 by default, for example. It could be a while before a game runs on Vulkan out of the box, and the way its special interest group Khronos, and more importantly AMD, promote its use, not just during game development, but also long-term support, will have a lot to do with it. A lot will also depend on NVIDIA, which holds about 70% in PC discrete GPU market share, to support the API. Over-customizing Vulkan would send it the way of OpenGL. Too many vendor-specific extensions to keep up drove game developers to Direct3D in the first place.
Add your own comment

111 Comments on AMD Actively Promoting Vulkan Beyond GPUOpen

#26
Ungari
rtwjunkieThey saw before Maxwell, and rightly predicted, that DX12 adoption would be super slow. And it has been. Why waste resources putting out products now that will do DX12 and perform worse on 11 if they don't have to?
Sure, because they can make even more money by delaying Volta so they could sell more of the surprise Paxwell cards, then obsolete them before owners get a single year of usage.
With 1080Ti coming out so late, you have to wonder if Volta will be delayed further.
Posted on Reply
#27
FordGT90Concept
"I go fast!1!11!1!"
wiakive followed vulkan a little bit and there is more ways to convert a dx11/opengl titles

* use opengl code and simply run it with a vulkan wrapper (this is probably what was done with talos)
* create native vulkan code and run it (this is probably what doom uses)

aka

* Talos runs like crap because it it uses opengl code behind it
* Doom runs like butter because its close to 100% vulkan code
Looking at what the Talos developers said, I didn't get the impression they wrapped it because they've put a lot of work into it and have updated several times. It's still rather buggy.
Posted on Reply
#28
bug
the54thvoid... AMD are just as much the reason Nvidia do what they do as AMD haven't stepped up to the plate. Their long game is so long that people get bored waiting and move on to something new. But then there is the fact that AMD is so focused on providing for the new consoles that they neglect the PC side. It seems fairly obvious now that Polaris was co-designed on the PC side as it was being produced to go into the PS4 Pro.
The story is much more complicated than that. AMD hasn't fallen behind because they wanted. First ATI was bought by AMD. Then AMD was backstabbed by intel and lost huge potential revenue, going in the red. Their video cards division delivered most of the time, but they simply do not have the cash to keep up. And thus they need to focus more than Nvidia.
RejZoRI've had like 6 generations of Radeon cards and I don't know where people find all this crap support and awful performance on DX11 and OpenGL. I've never had performance issues with any Radeon and I've played UT99 via enhanced OGL renderer as well as all id Software games. Which were all OGL as well.
OpenGL was always problematic on Linux, for example. Even now with their new, open source driver, OpenGL performance is still poor.
As for DX11, you can see that the same hardware can deliver more performance under DX12. You may choose to believe it's because the hardware wasn't used to its fullest under DX11 (not necessarily a driver problem) or you may choose to believe DX12 is putting pixie dust in the cards. I'm not going to try to convince you either way.
Posted on Reply
#29
qubit
Overclocked quantum bit
rtwjunkieNo. What NVIDIA have been doing is because they can afford to, and that is make the best card for the current most popular API. Nothing wrong wiith that. They saw before Maxwell, and rightly predicted, that DX12 adoption would be super slow. And it has been. Why waste resources putting out products now that will do DX12 and perform worse on 11 if they don't have to?

As to Vulkan, apparently this situation doesn't apply as much to NVIDIA, since it works pretty well on the current gen.
Yes, buying a high end NVIDIA card allows me to get top performance in current DX11 titles rather than some vapourware "DX13" or whatever in five years time when the card is long obsolete. This allows me to enjoy the optimum benefits of my reassuringly expensive investment right now, rather than at some promised vague future date which is unlikely to materialize. A terrible, terrible situation I tell you. I'm so distraught about it, can you help? :laugh:

@Ungari Why do you keep calling it "Paxwell"? It's Pascal or are you trying to make some sort of "protest" because you don't like NVIDIA?
the54thvoidYou can always tell the balanced posters because they bitch about both sides being culpable for the current state of affairs.
I see what you mean... ;)
Posted on Reply
#30
phanbuey
RejZoRFalse hope. Gets 30fps more than DX11/OGL on same hardware... What kind of strong weed are you smoking?
30 percent gain in one OGL game because they fixed a broken OGL driver...

Don't get me wrong it's awesome - just not reading into it until I see it happen for majority of games
Posted on Reply
#31
RejZoR
"Broken OGL". Hahahaha, sure. That moment when last gen Fury X runs almost as fast as current gen top of the line GTX 1080. That must be one hell of an OGL fix.
Posted on Reply
#32
crazyeyesreaper
Not a Moderator
FordGT90ConceptI want a push to Vulkan for the simple reason it works on other operating systems too (Linux/Mac). That said, what I've read from developers is that making Vulkan games is much more work than DirectX 12 games because Vulkan does less for you. Developers have to do more of their own optimizations.


GCN cards get worse framerates on Vulkan in Tales Principle compared to D3D11.
Problem with the Talos Principle at the time was it was still primarily OpenGL with Vulkan wrapper essentially. So still limited by old code it was posted by a dev somewhere but cant seem to find it now.
Posted on Reply
#33
Prima.Vera
Caring1More like they realised that horse is dead and pushing it wont help.
Is funny though, since on professional cards, some of them are faster than nVidia's similar on OpenGL renders and apps. But some of them are like 10x times slower, which make me think of a small conspiracy of some sort...
Posted on Reply
#34
phanbuey
RejZoR"Broken OGL". Hahahaha, sure. That moment when last gen Fury X runs almost as fast as current gen top of the line GTX 1080. That must be one hell of an OGL fix.
What moment is that?
Posted on Reply
#35
RejZoR
The one where you're saying AMD only fixed their broken OGL driver. Which to my experience of owning several generations of Radeons never existed.
Posted on Reply
#36
the54thvoid
Super Intoxicated Moderator
RejZoR"Broken OGL". Hahahaha, sure. That moment when last gen Fury X runs almost as fast as current gen top of the line GTX 1080. That must be one hell of an OGL fix.
Yes, an 8+ TFlop card matching another 8+ TFlop card..... Fury X should perform this well. This is the whole point of everything I post. It's the most over specced and (in DX11) under performing card. Nvidia cards do what they are meant to do in their given time frame. Fury X needs DX12 and Vulkan to work but those API's aren't yet the normal scene. By the time DX12 and/or Vulkan is the norm and DX11 is long forgotten we will be on what? Navi and Volta?

The current discussions on the future of API's is a philosophical error when we discuss contemporary cards.
Posted on Reply
#37
RejZoR
Because so many games on Fury X run at 10fps with DX11 or OGL...
Posted on Reply
#38
rtwjunkie
PC Gaming Enthusiast
UngariSure, because they can make even more money by delaying Volta so they could sell more of the surprise Paxwell cards, then obsolete them before owners get a single year of usage.
With 1080Ti coming out so late, you have to wonder if Volta will be delayed further.
I'm really curious where this almost immediate obsolescence is occurring, because my 980Ti still performs stellar as does the 980 before it that is now used by my better half. Indeed, it continues to perform at it's same original standard, probably better then at release, due to drivers optimizing performance. In that respect, it is no better or worse for it's category than my R9 380X, which has matured in its mid-priced category.
Posted on Reply
#39
phanbuey
RejZoRThe one where you're saying AMD only fixed their broken OGL driver. Which to my experience of owning several generations of Radeons never existed.
Whatever helps you sleep at night. Amd's OGL performance was poor at best and broken for many, especially in DOOM. A 30% increase on the same hardware is a direct result of that:

community.amd.com/thread/202636
community.amd.com/thread/200378
community.amd.com/thread/200351

Also I am still waiting to see this thing you're talking about where the fury X is almost as fast as a 1080... lol. I'd like to believe it - would sure love a bench.
the54thvoidYes, an 8+ TFlop card matching another 8+ TFlop card..... Fury X should perform this well. This is the whole point of everything I post. It's the most over specced and (in DX11) under performing card. Nvidia cards do what they are meant to do in their given time frame. Fury X needs DX12 and Vulkan to work but those API's aren't yet the normal scene. By the time DX12 and/or Vulkan is the norm and DX11 is long forgotten we will be on what? Navi and Volta?

The current discussions on the future of API's is a philosophical error when we discuss contemporary cards.
^ this. Im gonna go find some posts with RejZor raging about how his 4870 will be better than the 280 "when games catch up" or the 5870 vs 480 or the 6xxx vs the GTX 5xx... or how the 7xxx series had way more compute than the GTX 6xx and will crush nvidia when the devs pull their heads out of their green asses.

oh wait... it just keeps repeating.

Meanwhile nVidia fanboys can put on a blindfold and throw a dart at a list of recent titles and 95% of the time hit a title where nV cards dominate. :/ hmm.
Posted on Reply
#40
Totally
RejZoRThere is a bit of a falacy there saying it inherently works better on AMD because it was largely derived from Mantle. Mostly because that's not true. It works better on AMD because AMD has been pursuing closer to the metal API for years. They built GCN architecture (hardware) around it and have been enhancing it for years. They made architecture that is inherently better at operating via such API's in general, be it Mantle, Vulkan or DX12. DirectX 12 has nothing to do with Mantle apart from core idea and behold, AMD again way better at it than NVIDIA.
That would have been true had mantle been released first before any GCN cards hit the market but it arrived 2 years after the fact. You are saying that the API runs better because the cards were specifically designed to work on an API(s) that didn't exist. When it evident it was the other way around, Mantle was designed around their and for cards.
Posted on Reply
#41
RejZoR
By that logic, NVIDIA also had very shit OGL then (but people say NVIDIA's OGL is amazing!!!1111). Because it also has gains. Ever thought it's not because OGL was shit, but because Vulkan is simply superior (because lets face it, why would otherwise everyone jump on these API's)? I guess I'm just imagining things then...
Posted on Reply
#42
RejZoR
TotallyThat would have been true had mantle been released first before any GCN cards hit the market but it arrived 2 years after the fact. You are saying that the API runs better because the cards were specifically designed to work on an API(s) that didn't exist. When it evident it was the other way around, Mantle was designed around their and for cards.
You apparently think designs and technology planning happens 1 month before the launch of something new... Just try to remember how long AMD was planning the APU's and why they bought ATi later on. The whole planning and execution took years and it was all happening way before they bought ATi and before we actually got the APU's.
Posted on Reply
#43
Estaric
rtwjunkiemy 980Ti still performs stellar as does the 980 before it
and thats how it should be people are just seeing the big jump in performance from maxwell to pascal as it putting card like 980ti and 980 to rest, but when you actually look they are both still very powerful cards. same goes to my R9 Fury its a fantastic card and i will never say its slow by anymeans.:toast:


And what id like to know is why cant we all get along
Posted on Reply
#44
R-T-B
RejZoRThere is a bit of a falacy there saying it inherently works better on AMD because it was largely derived from Mantle. Mostly because that's not true. It works better on AMD because AMD has been pursuing closer to the metal API for years.
I know my throat is getting hoarse from saying this, but that's simply not true. This is more evidence that no one here really understands what words/phrases like "low-level" and "close to the metal" mean.

You don't optimize hardware to a low level api, you optimize software to a hardware exposed by a low level api.

At this moment, people are using exposed parts by DX12 to better optimize for AMD because frankly, there's a lot of optimizing to do compared to their DX11 renderer. There is some valid argument that async compute IS better supported on AMD's side, but it's not a valid argument for the way you are using it as NVIDIA also supports several things AMD doesn't:

Posted on Reply
#46
xkm1948
I see nothing wrong with it. Everyone benefits in the long run.
Posted on Reply
#47
Nergal
Meh

to me it is only becoming more and more clear that NV has the power, knowledge, money and position to create astounding new gen cards and to push the boundaries of GPU´s.

They - do - NOT

They prefer to let the little guy splonder about and pace themselves at its rate of average development.
Sometimes AMD gets ahead a little bit, but they quickly fix that in a matter of months (example: 1060 release)

Intel is doing the same thing. Putting down a roadmap to 2022 for 14nm...sigh

So I am all in support of AMD where possible and give due credit.
And it´s idea/tech now being promoted in a major deal is I think a good thing and should be viewed as such.
(=> I will not advise people to buy cr*p thou)

I won´t be surprised if NV suddenly pulls of cards in their next gen that give even better support of the new API than AMD.
They had it all along, but wanted to milk fanboys with the Maxwell first.


I had been playing fallout4 on fullHD with a 2007-gen CPU and a 2011GPU (Q6600+6850)
If you wanted to play new titles in 2000 with a PC from 1992;...though luck.

sidenote:
I should pick up on an old topic about 9800pro vs 6600GT.
Many were certain the 6600GT was better.
Posted on Reply
#48
bug
@Nergal You'd do the same thing in Nvidia or Intel's shoes. After all they're businesses and their goal is to make money.
Imagine you were the CEO of one of these companies that released the best possible product in 2016 (despite having no competition) and in 2017 you struggle coming up with something better and your profits plunge. You don't loose money, you just make way less than you did in the previous year. Will you still have a job after the shareholder's meeting?
Posted on Reply
#49
Captain_Tom
$ReaPeR$well, they should! the important part here i think is the fact that google made it their default API for android.
Anyone else remember that both AMD and Nvidia are bidding to supply the graphics in Samsung's next Smartphone APU's?

By making Vulkan the standard API of Android, AMD may have just secured a massive advantage in their bidding....
Posted on Reply
#50
bug
Captain_TomAnyone else remember that both AMD and Nvidia are bidding to supply the graphics in Samsung's next Smartphone APU's?

By making Vulkan the standard API of Android, AMD may have just secured a massive advantage in their bidding....
How come? AMD has no business in the mobile space at the moment while Nvidia has Tegra. And Pascal is already way more energy efficient than Polaris.
A lot will change by the time we see Samsung ditch Mali, but right now AMD has a (disputable, imho) software advantage while Nvidia has the hardware.
Posted on Reply
Add your own comment
Nov 21st, 2024 09:00 EST change timezone

New Forum Posts

Popular Reviews

Controversial News Posts