The news doing the rounds right now is that the Nintendo Switch, the new gaming device from Nintendo, will use Vulkan. People are getting rather excited and thinking it will mean more Linux ports, but right now it won't.
For one thing, the Switch hasn't even been released yet and it remains to be seen if it's even successful. It seems obvious, but people aren't even thinking about that.
The second most important thing to remember is that this is a brand new API, it's not proven itself just yet and not that many developers are actually using it. It's been out for nearly a year and so far on Linux only two games use Vulkan.
For the record: Using an open API is amazing for the success of the API. I think this is a great thing for it, but I don't want people to be unrealistic about what this means for Linux gaming. I also want to state for clarity I am not being negative here, but trying to help people be realistic for now.
This could push Vulkan forward some more, because it will be in the minds of more developers and hopefully more will end up using it. This is good for the stability of the API too, since more feedback will be sent off for the drivers and so on. For the API itself, it's going to help it. If more games eventually come to Linux and use Vulkan, it may mean we get a more stable experience too. What it doesn't mean is that by using Vulkan more games will come to Linux.
There is far more to a game than a graphics API. Sure, it gives developers a lower barrier for entry, but when has that alone suddenly meant more Linux ports? Not often at all. We are still to this day dealing with tons of developers using Unity that don't want to bring their games to Linux, for example.
You still have to worry about:
- Vulkan itself
Vulkan is more complex than both OpenGL and earlier versions of DirectX, it will take quite some time to learn.
- Development for every other bit of the puzzle
There's still tons of middleware that doesn't support Linux, for example.
- Testing for the above
People like to claim distribution fragmentation isn't an issue, but I see a lot of support requests of games not working on certain distributions for a variety of reasons.
- Post-release fixes
No game is really finished at release
- Marketing (if they actually want to make any money at all)
Just being on Steam doesn't make a game sell any more.
The biggest issue however, is the same as always: publishers and our market share. We still have that small market share to think about, so do the publishers.
To wrap up all of the above: It's good for the API, everything else people claim about it meaning more Linux ports are speculating.
What can we do about it? We continue on as we always have without getting too hyped about things that, right now, don't really concern us directly.
Buy Linux games from legitimate stores (Steam, GOG, itch.io, Humble, directly from developers), as that helps Linux gaming directly. Don't buy games before they are released on Linux (be sure your money counts!), and make sure developers know you want their games on Linux.
Also, make sure developers know to get in touch with us directly, since we have a rather big reach nowadays. Article taken from GamingOnLinux.com.
For one thing, the Switch hasn't even been released yet and it remains to be seen if it's even successful. It seems obvious, but people aren't even thinking about that.
The second most important thing to remember is that this is a brand new API, it's not proven itself just yet and not that many developers are actually using it. It's been out for nearly a year and so far on Linux only two games use Vulkan.
For the record: Using an open API is amazing for the success of the API. I think this is a great thing for it, but I don't want people to be unrealistic about what this means for Linux gaming. I also want to state for clarity I am not being negative here, but trying to help people be realistic for now.
This could push Vulkan forward some more, because it will be in the minds of more developers and hopefully more will end up using it. This is good for the stability of the API too, since more feedback will be sent off for the drivers and so on. For the API itself, it's going to help it. If more games eventually come to Linux and use Vulkan, it may mean we get a more stable experience too. What it doesn't mean is that by using Vulkan more games will come to Linux.
There is far more to a game than a graphics API. Sure, it gives developers a lower barrier for entry, but when has that alone suddenly meant more Linux ports? Not often at all. We are still to this day dealing with tons of developers using Unity that don't want to bring their games to Linux, for example.
You still have to worry about:
- Vulkan itself
Vulkan is more complex than both OpenGL and earlier versions of DirectX, it will take quite some time to learn.
- Development for every other bit of the puzzle
There's still tons of middleware that doesn't support Linux, for example.
- Testing for the above
People like to claim distribution fragmentation isn't an issue, but I see a lot of support requests of games not working on certain distributions for a variety of reasons.
- Post-release fixes
No game is really finished at release
- Marketing (if they actually want to make any money at all)
Just being on Steam doesn't make a game sell any more.
The biggest issue however, is the same as always: publishers and our market share. We still have that small market share to think about, so do the publishers.
To wrap up all of the above: It's good for the API, everything else people claim about it meaning more Linux ports are speculating.
What can we do about it? We continue on as we always have without getting too hyped about things that, right now, don't really concern us directly.
Buy Linux games from legitimate stores (Steam, GOG, itch.io, Humble, directly from developers), as that helps Linux gaming directly. Don't buy games before they are released on Linux (be sure your money counts!), and make sure developers know you want their games on Linux.
Also, make sure developers know to get in touch with us directly, since we have a rather big reach nowadays. Article taken from GamingOnLinux.com.
Some you may have missed, popular articles from the last month:
The comments on this article are closed.
Quoting: liamdaweIt's pleasing to see you do at least understand and actually agree with me, which is weird considering your other replies (mostly on reddit).
Market share is the key point that is almost always goes down to. Time and time again we see developers not doing ports due to it.
Again, you missed my point there, so I didn't understand what you were arguing then either. I never said developers will reconsider their stance on market share because of less technical issues. I said what I said - less technical issues = more potential ports. Because unlike some thick skull backwards thinking publishers, some already don't view Linux gaming market as insignificant, but it doesn't mean they don't face any potential technical barrier that they need to overcome when releasing for Linux. Proliferation of redundant APIs is one of them.
Quoting: liamdaweWeb development is an entirely different thing. The web is literally everywhere and can be easily accessed, Linux desktop gaming is completely different (again, market-share here).
That's without getting into how completely different it is deploying HTML to Vulkan-powered games to each platform. Anyone should be able to see how those topics don't really mix together.
I explained above how the parallel is obvious. Difference in details here doesn't change the parallel of open standard commonality. Something like Web Assembly can even make Vulkan be adopted for the Web as a graphics API.
Last edited by Shmerl on 20 December 2016 at 12:20 am UTC
1 Likes, Who?
As other people said, it's nice to see Vulkan active and actually used by game developers :) other than that there's nothing to be happy about... Nintendo doesn't even release their games for Windows xD
0 Likes
There is still a long term effect that might be beneficial for Linux.
In big game dev studios like Ubisoft, EA, etc. engines are often compatible with different graphics API. Of course, the less APIs the better, since it represents a lot of Q&A to verify that each APIs have around the same renderings for the same performances.
15 years ago, the default graphics API was OpenGL. But Khronos didn't improve this API for a long long time, and DirectX appeared and became more and more appealing to the developers. It was obviously easier to develop with DirectX than OpenGL (and its infinite number of extensions). Then the XBox was released and there were no real incentive to stay on the aging OpenGL API. Dev studios started to port there engines to DirectX 9, and since it was compatible with Windows and Xbox, no platform forced them to support OpenGL anymore. OpenGL became a "nice to have... eventually".
The PS2, PS3 and PS4 from Sony were released, but always with their own flavor of OpenGL (especially the shaders). The same goes for Nintendo with their GameCube, Wii and Wii U. So the dev studios still didn't have any incentive to port their 3D engines to the "true" OpenGL API, they already had to port it for Sony and Nintendo. And again, the less APIs, the better.
So: OpenGL is not interesting anymore + Linux's only API being OpenGL = no game on Linux.
Now, Nintendo announces that the Switch will run Vulkan. So porting an engine to Vulkan means that it is compatible with Windows, Linux, Android and Nintendo. It's starting to be appealing.
Nintendo is... nice. But I'm not sure the Switch will work. It's basically a big smartphone that can't phone. I think that the real announce that will have a big impact in the Linux world is if Sony declares that they support Vulkan. Whether it's for the PS4 or the PS5, that could change the balance in favor of using Vulkan as the default API. DirectX would just be the API that we have to use for the XBox's.
Of course, it would be nice if Apple could stop fooling around with their Metal API...
Last edited by Creak on 20 December 2016 at 12:37 am UTC
In big game dev studios like Ubisoft, EA, etc. engines are often compatible with different graphics API. Of course, the less APIs the better, since it represents a lot of Q&A to verify that each APIs have around the same renderings for the same performances.
15 years ago, the default graphics API was OpenGL. But Khronos didn't improve this API for a long long time, and DirectX appeared and became more and more appealing to the developers. It was obviously easier to develop with DirectX than OpenGL (and its infinite number of extensions). Then the XBox was released and there were no real incentive to stay on the aging OpenGL API. Dev studios started to port there engines to DirectX 9, and since it was compatible with Windows and Xbox, no platform forced them to support OpenGL anymore. OpenGL became a "nice to have... eventually".
The PS2, PS3 and PS4 from Sony were released, but always with their own flavor of OpenGL (especially the shaders). The same goes for Nintendo with their GameCube, Wii and Wii U. So the dev studios still didn't have any incentive to port their 3D engines to the "true" OpenGL API, they already had to port it for Sony and Nintendo. And again, the less APIs, the better.
So: OpenGL is not interesting anymore + Linux's only API being OpenGL = no game on Linux.
Now, Nintendo announces that the Switch will run Vulkan. So porting an engine to Vulkan means that it is compatible with Windows, Linux, Android and Nintendo. It's starting to be appealing.
Nintendo is... nice. But I'm not sure the Switch will work. It's basically a big smartphone that can't phone. I think that the real announce that will have a big impact in the Linux world is if Sony declares that they support Vulkan. Whether it's for the PS4 or the PS5, that could change the balance in favor of using Vulkan as the default API. DirectX would just be the API that we have to use for the XBox's.
Of course, it would be nice if Apple could stop fooling around with their Metal API...
Last edited by Creak on 20 December 2016 at 12:37 am UTC
4 Likes, Who?
Web development is a pretty good analogy here. Multimedia APIs situation is very similar to what the web looked a decade ago – everyone was targeting Internet Explorer, ActiveX, later with Silverlight, closed video formats for embedded playback, etc. This made a lot of it unusable on Linux and its open browsers, with a lot of stuff displaying “you need to use IE to properly display this website” or similar messages. Pretty analogically to D3D vs OpenGL situation.
Now, similarly to what happened to the web, broader adoption of Vulkan and OpenGL might convince more game devs to use those open APIs. And the more they use Vulkan (and/or OpenGL), the bigger chance that the next shiny engine will use it, that new tutorial “how to start writing your game” will be based on it, that books will get written on it, and in effect much higher portion of all devs will write using Vulkan/OGL.
Will they automatically release for Linux? No. But it will be easier for them than rewriting the whole DX stack, they won’t have a mental blocker about working with hated technology (some people really hate OpenGL, so porting D3D into it is painful for them), and the whole market will become a bit friendlier.
Also, when Vulkan becomes popular (gets used in games ported to Windows from Nintendo), those Vulkan games for Windows will work better on Wine. And games working well on Wine will convince a few more people to switch to Linux, which will make the Linux marketshare bigger…
So, sure, Nintendo choosing Vulkan won’t automatically save us, it won’t bring a lot of Linux games next month. But it’s not true it is not beneficial for Linux gaming and for open standards – it is, and every new adoption of those standards on every single closed platform with funding is a good thing, as it leads the industry out of MS vendor lock-in, makes open standards more viable, and removes stigma of being useless from them. And those changes are needed for Linux gaming to progress.
Last edited by silmeth on 20 December 2016 at 1:07 am UTC
Now, similarly to what happened to the web, broader adoption of Vulkan and OpenGL might convince more game devs to use those open APIs. And the more they use Vulkan (and/or OpenGL), the bigger chance that the next shiny engine will use it, that new tutorial “how to start writing your game” will be based on it, that books will get written on it, and in effect much higher portion of all devs will write using Vulkan/OGL.
Will they automatically release for Linux? No. But it will be easier for them than rewriting the whole DX stack, they won’t have a mental blocker about working with hated technology (some people really hate OpenGL, so porting D3D into it is painful for them), and the whole market will become a bit friendlier.
Also, when Vulkan becomes popular (gets used in games ported to Windows from Nintendo), those Vulkan games for Windows will work better on Wine. And games working well on Wine will convince a few more people to switch to Linux, which will make the Linux marketshare bigger…
So, sure, Nintendo choosing Vulkan won’t automatically save us, it won’t bring a lot of Linux games next month. But it’s not true it is not beneficial for Linux gaming and for open standards – it is, and every new adoption of those standards on every single closed platform with funding is a good thing, as it leads the industry out of MS vendor lock-in, makes open standards more viable, and removes stigma of being useless from them. And those changes are needed for Linux gaming to progress.
Last edited by silmeth on 20 December 2016 at 1:07 am UTC
8 Likes, Who?
Well, I only buy games that has linux version. And as I'm lenient I include android/ios games too lol. (I'm sucker for mobile time wasters lol.)
But yeah I understand that linux version is basically a fan service for any company. And I'm very grateful for it. My ultimate goal is to buy every linux games.
But yeah I understand that linux version is basically a fan service for any company. And I'm very grateful for it. My ultimate goal is to buy every linux games.
1 Likes, Who?
If AAA developers and publishers bring their game to the Switch and are forced to learn and use Vulkan, it is a huge benefit for everyone as they may use Vulkan for their PC releases too.
Last edited by t3g on 20 December 2016 at 2:23 am UTC
Last edited by t3g on 20 December 2016 at 2:23 am UTC
0 Likes
Quoting: t3gIf AAA developers and publishers bring their game to he switch and are forced to learn and use Vulkan, it is a huge benefit for everyone as they may use Vulkan for their PC releases too.Right now, even with Nintendo using Vulkan, DirectX is still more interesting. Having a graphics engine based on DirectX ensures to have a game compatible with Windows and XBox. It's like 50% of the market. The rest is mainly Sony with its PS4. That's why Nintendo announcing they're supporting Vulkan just isn't enough. But if Sony announces it, it might change the situation.
Last edited by Creak on 20 December 2016 at 1:55 am UTC
1 Likes, Who?
Quoting: CreakThe rest is mainly Sony with its PS4. That's why Nintendo announcing they're supporting Vulkan just isn't enough. But if Sony announces it, it might change the situation.
It would be awesome if Sony added Vulkan support to the PS4 or the PS5 and above. The PS4 is still killing the Xbox One in market share (even if the XO has outsold the PS4 in the past few months) and having Sony in the Vulkan camp would be beneficial.
1 Likes, Who?
This article is overly pessimistic. More developpers needing to use Vulkan to support a new very popular platform is going to mean more developpers learning Vulkan which in turn mean more developpers using Vulkan, and this cycle goes on forever. And it will also means more developping tools, more improvements...
If we get so much games on Linux today it's not because of SteamOS, SteamOS is dead, it's because of what Valve and SteamOS did, because of it a lot of developpers learnt to develop games for Linux and now they are available on the market, companies are less afraid to develop for Linux anymore because their people have the skills required. Having more and more developpers working with Vulkan out there will mean having more and more people capable of making Linux native games, it's one thing less that companies will have to worry about if they consider having a Linux version, one thing that could make a difference between having it or not.
Last edited by Nyamiou on 20 December 2016 at 2:32 am UTC
If we get so much games on Linux today it's not because of SteamOS, SteamOS is dead, it's because of what Valve and SteamOS did, because of it a lot of developpers learnt to develop games for Linux and now they are available on the market, companies are less afraid to develop for Linux anymore because their people have the skills required. Having more and more developpers working with Vulkan out there will mean having more and more people capable of making Linux native games, it's one thing less that companies will have to worry about if they consider having a Linux version, one thing that could make a difference between having it or not.
Last edited by Nyamiou on 20 December 2016 at 2:32 am UTC
3 Likes, Who?
Patreon
PayPal
See more from me