Tuesday, February 16th 2016

NVIDIA Talks Vulkan, Supports it on "Kepler" and "Maxwell" GPUs

NVIDIA talked Vulkan in its latest GeForce blog post, announcing that your GeForce GTX graphics card already supports the "industry forged" API. NVIDIA is offering Vulkan hardware-acceleration on its "Kepler" and "Maxwell" GPU architectures at this time, and on Windows 7 and above; PC Linux, and Android. NVIDIA is all praises for Vulkan's low-latency and high-efficiency pathways, which streamline the process of drawing graphics.

Vulkan makes its big mainstream debut with a major update to "The Talos Principle," by Croteam (the people behind the "Serious Sam" franchise). This update adds a Vulkan renderer to the game, and ships later today. NVIDIA has an driver ready with the Vulkan API, which you can download from here. Maintained by the Khronos Group, Vulkan is a successor to OpenGL, although it's built from the ground up, with a major chunk of its code being contributed by AMD, from its Mantle API.

Source: NVIDIA Blog
Add your own comment

20 Comments on NVIDIA Talks Vulkan, Supports it on "Kepler" and "Maxwell" GPUs

#3
Fluffmeister
Interesting the update is coming later today, I look forward to seeing the early results.
Posted on Reply
#4
Loosenut
btarunr said:
...with a major chunk of its code being contributed by AMD, from its Mantle API. Source: NVIDIA Blog
Not according to these guys. Apparently, it's all Nvidia o_O
Posted on Reply
#5
DigitalUK
AMD announced the release of there beta today
http://support.amd.com/en-us/kb-articles/Pages/Radeon-Vulkan-Beta.aspx

because
Today the Khronos Group announced immediate public release of the open standard and cross-platform Vulkan™ 1.0 graphics API

As a complement to OpenGL™, descended from AMD's Mantle, and forged by the industry, Vulkan™ is a powerful low-overhead graphics API designed for developers who want or need deeper hardware control over GPU acceleration. Vulkan™ gives software developers control over the performance, efficiency, and capabilities of Radeon™ GPUs and multi-core CPUs.
Posted on Reply
#6
ZoneDymo
Loosenut said:
Not according to these guys. Apparently, it's all Nvidia o_O

I think you have to read it again,
never does it say Nvidia did anything other then making sure it works on their hardware.


"NVIDIA has worked closely with the Khronos Group, the creators of Vulkan, throughout its development, and as of today all Kepler and Maxwell graphics card running Windows 7 or later, or Linux, are supported by Vulkan."

and obviously Nvidia would never mention let alone give praise to AMD, they are their enemy yo.
Posted on Reply
#7
birdie
Do not try to benchmark The Talos Principle yet:
Edwin,

engine design for Vulkan is basically consited of three major parts:

1)
Port. Make it work as fast as possible just by wrapping current engine design around Vulkan. Avoid all pitfalls and bottlenecks. This is what we did by now and released as patch for Talos.

2)
Use Vulkan for multi-threaded rendering. Our engine is designed really well for multi-threaded rendering, but we have only our wrapper for it - calls to graphics API (like Vulkan) are not multi-threaded. Yet.
That being said, this is the next step what we'll do. And probably release that also as patch for Talos. I tried to do that with Direct3D 11 long time ago (support for its deffered contexts), but it was too much pain and too little or even no gain. :( That's just one of reasons why we decided to stick with our own approach for MT renderer for that long. :/

3)
Redesign engine for Vulkan. This is the biggest step and can be split in two:

3a)
Precache all rendering states (which mostly mean materials in game) up front. This will make rendering calls much simplier and faster. So, instead of deciding at rendering time what is needed for a material to be rendered via Vulkan, do this at loading time and then when material needs to be rendered just give it to Vulkan, via one or two simple function calls.

3b)
Precache all geometry, material, textures, everything that is needed for rendering an object up front. This basically creates so called command buffer ready for Vulkan, and nothing extra needs to be set or created at render time.

3rd part of port is, obviously, the most complex one, and it'll take time to change engine design for it, step-by-step.

Hope I explained this well. :)
DEN
I mean you surely can benchmark but the game is not yet optimized for this API.

The Talos Principle Vulkan FAQ
Posted on Reply
#8
Loosenut
ZoneDymo said:
I think you have to read it again,
never does it say Nvidia did anything other then making sure it works on their hardware.

"NVIDIA has worked closely with the Khronos Group, the creators of Vulkan, throughout its development, and as of today all Kepler and Maxwell graphics card running Windows 7 or later, or Linux, are supported by Vulkan."

and obviously Nvidia would never mention let alone give praise to AMD, they are their enemy yo.
I stand corrected, I did read it again, thank you.
Posted on Reply
#10
the54thvoid
Nem was inappropriately touched by Nvidia at an early age. It left him damaged and lonely. Unable to communicate properly his pain is expressed in irrelevant attacks on the company that stole his innocence. Let's pray for him.

I have him on ignore but it doesn't work on links from the main page on mobile.
Posted on Reply
#12
chaosmassive
its seem that Nvidia actually have all tech in hand (read:GPU) they only waiting right moment for announcement and 'implementation'
when AMD announce Mantle, Nvidia simply keep silent, but now that Mantle handed to Khronos and reincarnate to Vulkan
Nvidia simply say our GPU support Vulkan
Posted on Reply
#13
ZoneDymo
the54thvoid said:
Nem was inappropriately touched by Nvidia at an early age. It left him damaged and lonely. Unable to communicate properly his pain is expressed in irrelevant attacks on the company that stole his innocence. Let's pray for him.

I have him on ignore but it doesn't work on links from the main page on mobile.
idk man, seems you are the one with some unresolved issues
Posted on Reply
#14
the54thvoid
ZoneDymo said:
idk man, seems you are the one with some unresolved issues
I have issues with brand zealots who exhibit irrational loyalty. This goes for both camps. My recent posts elsewhere recommend AMD over Nvidia for gfx options, so I'm not one of them. If you check Nem's post history you'll find he's a true red coat. Constant posts like his drag threads down in a technical way. Contrast with some others who show a bias who use reasoned arguments. I've got no issue with them. My unresolved issue would only be a lack of patience.
Posted on Reply
#15
TheGuruStud
Nvidia must feel threatened by being left out or they wouldn't even care. If AMD creates/contributes (not to mention open source), then it just so worthless and terrible that they can't be bothered. So, it's a little interesting to see them on board so early. Especially, since they flat out refused to even think about Mantle.

It sounds like there's a big API shift a-coming. And it sounds like even M$ won't be able to stop it this time (F you DX).
Posted on Reply
#16
idx
Loosenut said:
Not according to these guys. Apparently, it's all Nvidia o_O
"Vulkan is derived from and built upon components of AMD's Mantle API, which was donated by AMD to Khronos"

"Vulkan™ is an open standard and cross-platform Application Programming Interface (API) developed by the Khronos™ Group. Derived from AMD’s revolutionary Mantle API"


Also DX12 is another repackaging of AMD API:

Posted on Reply
#17
R-T-B
idx said:
"Vulkan is derived from and built upon components of AMD's Mantle API, which was donated by AMD to Khronos"

"Vulkan™ is an open standard and cross-platform Application Programming Interface (API) developed by the Khronos™ Group. Derived from AMD’s revolutionary Mantle API"


Also DX12 is another repackaging of AMD API:


That's never proven it's a "repack" only that the drivers error returns function similarly. Well no shit Sherlock, they are both low level APIs...
Posted on Reply
#18
Fluffmeister
The problem with open standards, is everyone gets to use them ladies.
Posted on Reply
#19
Musaab
R-T-B said:
That's never proven it's a "repack" only that the drivers error returns function similarly. Well no shit Sherlock, they are both low level APIs...
In fact it remind me with a story saying that mantel is rip off from DX12 and Microsoft was crazy mad that AMD did that without the partners permission and that explain why Windows 10 lunched two years earlier than what planed.
Posted on Reply
#20
efikkan
Loosenut said:
Not according to these guys. Apparently, it's all Nvidia o_O
Vulkan is a collaboration by Khronos (lead by Nvidia btw). Mantle as one of many sources of input. But the underlying API built around SPIR-V is completely new.

BTW, remember all the fanboys which claimed that AMD would rock Vulkan because it "was just Mantle" and Nvidia would struggle, well Nvidia managed to create the first driver which has been usable for months. Nvidia managed to create the first demo almost one year ago. And if Vulkan was Mantle then AMD would have showed a great advantage, an advantage which we now know for sure doesn't exist.

idx said:
"Vulkan is derived from and built upon components of AMD's Mantle API, which was donated by AMD to Khronos"

"Vulkan™ is an open standard and cross-platform Application Programming Interface (API) developed by the Khronos™ Group. Derived from AMD’s revolutionary Mantle API"

Also DX12 is another repackaging of AMD API
That is utterly untrue. As Nvidia said "Our work with Microsoft on DirectX 12 began more than four years ago".

Mantle was AMDs side project while waiting for Direct3D 12, the core features was planned in 2010/2011. Mantle was one source of input during formalizing Direct3D 12, which is the reason why some basic generic function calls are the same. But any programmer would know that this does not make the underlying API the same.

Only a fool would believe that AMD and Nvidia managed to add the right hardware features for Direct3D 12 years in advance without planning for it.
Posted on Reply
Add your own comment