DXVK-Native 1.9.2a Released For Direct3D-Over-Vulkan With Native Games

[dragon321]

What is the reason for this being a fork and not included in the upstream since the dev of this is a dev of the base?

If I understood correctly this is probably because DXVK is build for Windows platform so it more or less depends on Wine while DXVK-Native gets rid of such dependency, replaces "Windows-isms" (quote from readme) with platform and framework agnostic replacements and provides development headers so you can use it in native Linux code without touching Wine.

[rabcor]

Why not just make new games using only vulkan?

Because many developers only know how to use DirectX, not Vulkan or OpenGL.

And the companies that hire them don't want to invest resources in having them learn a new API unless they deem they have a good enough reason, and linux with it's roughly 1% market share of gaming is not a good enough reason for most of them.

Android might be though. At some point.

[Pickup]

Because many developers only know how to use DirectX, not Vulkan or OpenGL.

And the companies that hire them don't want to invest resources in having them learn a new API unless they deem they have a good enough reason, and linux with it's roughly 1% market share of gaming is not a good enough reason for most of them.

Android might be though. At some point.

Maybe you're right, but Linux native games exist. For now.
What will prevent companies to "hey, who cares about these 1% weirdos, let's make our games windows-only and let the leenookz freaks rely on wine, dxvk and their luck"?
I don't feel it's good for linux users...

[Linuxxx]

Maybe you're right, but Linux native games exist. For now.
What will prevent companies to "hey, who cares about these 1% weirdos, let's make our games windows-only and let the leenookz freaks rely on wine, dxvk and their luck"?
I don't feel it's good for linux users...

Native Linux builds are only going to become more likely once SteamOS manages to surpass at least macOS' marketshare.

Should Valve make it easy enough to also serve as a general desktop OS, then this could actually happen sooner than we realize...

[rabcor]

Maybe you're right, but Linux native games exist. For now.
What will prevent companies to "hey, who cares about these 1% weirdos, let's make our games windows-only and let the leenookz freaks rely on wine, dxvk and their luck"?
I don't feel it's good for linux users...

Maybe steamdeck is gonne help push us there if it's popular enough, then again the steam machines were also such a hope once, but the steamdeck seems very promising, to the point where I'd say it's the most promising handheld gaming device I've ever seen and most of my friends (who do not even use linux) agree on that too. Is handheld gaming big enough? are there enough people who like to game on handheld devices and will they be willing to get the steamdeck instead of what they're currently using? Valve apparently seems to think so.

[jeoshua]

Maybe steamdeck is gonne help push us there if it's popular enough, then again the steam machines were also such a hope once, but the steamdeck seems very promising, to the point where I'd say it's the most promising handheld gaming device I've ever seen and most of my friends (who do not even use linux) agree on that too. Is handheld gaming big enough? are there enough people who like to game on handheld devices and will they be willing to get the steamdeck instead of what they're currently using? Valve apparently seems to think so.

I look at it this way: Smart Phones are basically Handheld Computers. What did the advent of Android do for the Linux community? Quite a lot, and also not that much. Everyone with an Android phone is running what basically amounts to a Linux computer in their pocket, but that didn't really change the Desktop experience much, if at all. There are some great things that happened on the backend, tho, mostly related to supporting the ARM architecture.

Maybe Steam Deck will do something similar, with respect to gaming and APUs? We can't say for certain, but Valve's efforts have already yielded a great improvement in Proton, and thus gaming on Linux. So I'm hopeful... but the idea that it will revolutionize all of gaming and lead to all companies making native Linux games is honestly a non-starter. Valve didn't put this much effort into Proton for nothing.

[birdie]

Why not just make new games using only vulkan?

Vulkan is akin to C while DirectX 11/OpenGL are akin to Java/Python/Ruby/C#.

Yes, applications written in C can be quite a lot faster than applications written in high-level languages, but such languages allow to write applications a lot faster, easier without wasting a ton of time debugging and solving issues. For simple (Indie) games there is almost zero benefit in using Vulkan/D3D12 - you'll end up writing 20 times more code while your application will run at 600 fps vs 500 fps using D3D11/OpenGL. Totally not worth it.

Vulkan/D3D12 also require a very deep understanding of the graphics pipeline while with D3D11/OpenGL you don't need to bother with such things.

Previous replies to your question in this topic are almost entirely incorrect.