Nvidia hosted a Vulkan Developers Day which sounds awesome

There is too much secrecy going on for my taste. Oh and delays, too many of those too

I do worry at times that Vulkan will end up in a similar situation to OpenGL with developers using special Nvidia extensions and performance stuff which Intel and AMD may not have

This!!!! and unfortunately they more and likely will.

Bet they make sure async compute aint used too much too

edit:

This too.

There is too much secrecy going on for my taste. Oh and delays, too many of those too

Last edited by pete910 on 20 January 2016 at 8:30 pm UTC

I would be very surprised and disappointed if some game developers still choose DirectX 12 over Vulkan, but we'll see. I don't doubt that Microsoft will try to FUD as much as possible, but developers should now know how much Microsoft can be trusted.

Corporations do not have the public's interest at heart. This is why everyone should be concerned about having NVIDIA, AMD, etc on the board.

I have no problems with them been on board, what bothers me is the fact the president of the kronos group is working for one of them !

What "performance stuff" are you talking about nvidia in this case? They could create proprietary extensions, but as long as their hardware isn't capable of all the new features... well, people would buy it still, I guess. Fanboism :)

I do worry at times that Vulkan will end up in a similar situation to OpenGL with developers using special Nvidia extensions and performance stuff which Intel and AMD may not have. I hope I am wrong about this, as I would like to see a level playing field, but all the GPU people will be wanting to find a way to get people to use their chips and so I imagine we will see some form of it.

This is from the blog:

In the morning, NVIDIA engineers gave a series of lectures about the best ways to use Vulkan with NVIDIA hardware.

The developers will probably focus on NVIDIA. If so, it's business as usual. :(

This is from the blog:

In the morning, NVIDIA engineers gave a series of lectures about the best ways to use Vulkan with NVIDIA hardware.

The developers will probably focus on NVIDIA. If so, it's business as usual. :(

Of course they will give lectures about best ways to use Vulkan with Nvidia hardware. What did you expect, that Nvidia hosts an event to tell people how to program for AMD or Intel? Seriously, I don't get this whole conspiracy stuff, of course Nvidia, which has already told us that there will day 1 support for Vulkan from them, teaches developers how to use Vulkan with their hardware. Concluding from that that Vulkan has already failed is somewhat weird, putting it nicely. And, sorry if I have to say that, rejecting Vulkan based on that conclusion is outright moronic and would spell the end for Linux gaming.

Bloody hell. Can we stop with the conspiracies now please?

I would be very surprised and disappointed if some game developers still choose DirectX 12 over Vulkan, but we'll see.

They totally will, and they even will use it as a PR subject.
Game X - making use of incredible DirectX 12 feature Y and Z!

@TheBoss

You shouldn't be concerned about the extension mechanisms of OpenGL and Vulkan. Extensions have always been a major strength of OpenGL, allowing new new features to be used as soon as they are available. Features that developers want ( i.e. actually use ) then tend to find their way into the main specification.

In contrast, with D3D new hardware features only appear when Microsoft choose to expose them with a new D3D release. This meant that some Windows gamers paid for hardware features that were never used, because Microsoft would use D3D features as a business tool to play hardware manufacturers against each other.

Any good developer would want decisions about what hardware to use and how to use it left to them, not decided in the business interests of the OS manufacturer.

I would be very surprised and disappointed if some game developers still choose DirectX 12 over Vulkan, but we'll see.

They totally will, and they even will use it as a PR subject.
Game X - making use of incredible DirectX 12 feature Y and Z!

They will not choose Vulkan or DirectX because it's 'better'. Both are low level, so I don't think it's going to be that much of a difference. However, if they intend to port games, then using Vulkan is the only choice that makes sense.

Windows only games will probably still use DirectX 12. It's already out, and the various engines that companies use definitely support it. They make games for consoles, and Xbox One has the DirectX 12 API.

There are engines that will support Vulkan (Frostbite), even though they are not going to develop games for Linux. Seems like they (EA) will make smartphone and tablet games using that backend, there is no need for it otherwise.

This is from the blog:

In the morning, NVIDIA engineers gave a series of lectures about the best ways to use Vulkan with NVIDIA hardware.

The developers will probably focus on NVIDIA. If so, it's business as usual. :(

Of course they will give lectures about best ways to use Vulkan with Nvidia hardware. What did you expect, that Nvidia hosts an event to tell people how to program for AMD or Intel? Seriously, I don't get this whole conspiracy stuff, of course Nvidia, which has already told us that there will day 1 support for Vulkan from them, teaches developers how to use Vulkan with their hardware. Concluding from that that Vulkan has already failed is somewhat weird, putting it nicely.

It's the developers that decide what API they will use, and also everything else when it comes to developing. If they for some reason write code that perform better on NVIDIA, that doesn't mean that Vulkan itself has failed.

There is no reason at all to reject Vulkan. OpenGL isn't going to change, and Vulkan is of course the only way forward.

And, sorry if I have to say that, rejecting Vulkan based on that conclusion is outright moronic and would spell the end for Linux gaming.

When it comes to the future of Linux gaming, the only thing that matters is the number of games available. Lots of indies, and very few AAA games, is not what attract people to our platform.

The problem is that a lot of companies are not interested in porting AAA games due to market share. A new API doesn't really make a difference.

Today AAA games are mostly ported by Feral, Aspyr and VP. Then there's sometimes in-house ports released (like Dying Light, and Deep Silver games). With 1% market share, that will probably not change.

I don't know if it's that easy to make a translation layer for DirectX 12 to Vulkan, so maybe VP will not be able to release Vulkan games.

There are indie developers who did not port games because of OpenGL. They will hopefully start porting games when Vulkan is available.

It's the developers that decide what API they will use, and also everything else when it comes to developing. If they for some reason write code that perform better on NVIDIA, that doesn't mean that Vulkan itself has failed.

There is no reason at all to reject Vulkan. OpenGL isn't going to change, and Vulkan is of course the only way forward.

That's exactly what it could mean, how are you getting that logic? If a "standard" favors one vendor over another when it's supposed to be vendor-agnostic, it's either a) that vendor is doing it "right" and other vendors need to adapt to the standard, or b) the standard is a bad one.

The Vulkan API itself has to be vendor-agnostic, it doesn't make sense if it isn't. But obviously it's possible to write code that 'works' on a certain architecture (the good way), and then it's possible to write code that fully takes advantage of that architecture (the best way).

Hopefully all or most of the games being released will fully take advantage of all architectures.

I do worry at times that Vulkan will end up in a similar situation to OpenGL with developers using special Nvidia extensions and performance stuff which Intel and AMD may not have.

This is how standards evolve, actually. Let's say Nvidia introduces a new NV extension that developers like very much. Said developers begin to use it with great results, developing specific render paths for nvidia's GPUs. Nvidia's customers enjoy the results and the extension usage spreads, leaving AMD and Intel customers in the dust. AMD then adapt its GPUs to take advantage of the functionality and contextually publish a proposal for an ARB extension definition for the new feature, maybe in collaboration with some Nvidia engineer. Later the Kronos Group accept that ARB extension that becomes the new way to leverage the Nvidia invented functionality. Meanwhile developers begin to use the Kronos backed ARB extension, targeting both Nvidia and AMD users (Intel is still deciding what to do...)
Many years later the ARB extension becomes core functionality in a new version of Vulkan.

On the other end, Microsoft decides in complete isolation what you are allowed to use in D3D, which is not a standard.

Too much speculation here. I see it the positive way, NVidia tries to push Vulkan by doing Events like this for developers getting them closer to Vulkan, away from DX12. One of the main reasons we have so many D3D games is, that developers are familiar with it, and are not with OpenGL, especially for which things to use which methods and functions for optimal performance.

Extensions are not by any mean a bad thing, rather the opposite. It gives you the opportunity to tweak more, and it's up to the hardware vendors to support certain extensions. I didn't expect the end result to be 100 % hardware agnostic except for the base API, and it's pretty much up to the game developers which optimizations of which vendor they support. As long as they keep the extensions optional for their games and engines, I see no reason why they shouldn't use them to make their games shine even more if they want to do the additional work for it.


Last edited by STiAT on 22 January 2016 at 2:00 pm UTC