Linux Gaming Thoughts

[Michael]

From a reader...

Originally Posted by Scott

When will we have real gaming on Linux?

This is, I suppose, a rhetorical question, but if there is any new news, I'd love to hear it. I've been a Windows user for years, solely because of games. If Linux could just gather its forces and address this one issue, I could cross over.

However, to the best of my knowledge, Linux isn't getting anywhere with gaming, and that's due to Linux's main undermining flaws: 1) lack of focused effort, and 2) lack of desire on the part of developers to be a mainstream solution. (The whole Cedega/Wine affair is not a proper solution, by the way. It's baling wire and duct tape.)

It would seem that the present time would be excellent for a concerted effort to supply game developers with a tool package that would facilitate the dual development of games for Linux and DirectX 10. I know that AMD/ATI has contributed one part that might help, a HLSL2GLSL tool, but I imagine a lot more has to be done, including hand-holding of developers.

I believe that if Linux could get games running natively, with push-button installation / de-installation and all the bells and whistles, then Linux could really take off. It seems to me that with online retailing, broadband connections for downloads, automatic updates, dual core computing, virtualization (?), better graphics drivers on Linux, and so on, this must be the best time there has ever been for this type of endeavor.

And if the Linux community isn't interested in this, I wish they would just pack up and go home. I know that sounds harsh, but by chronically doing nothing they are doing more harm than good. People think Linux is going to provide an alternative to MS - I hear it all the time - but if it won't address the concerns of the mainstream, then it isn't ever going to be a credible alternative. All it is doing is reinforcing the MS monopoly by sucking up the time and interest of people who do want a credible alternative.

Sorry if that sounded a bit like a rant.

There's no need to answer this email, but if the idea sparks an article, editorial, or a good remark in the right ear, then I've fulfilled my purpose.

[niniendowarrior]

Although I admire the person's intent, I highly doubt things will change. I have very extreme views about the gaming industry regardless of the platform and the one thing Linux never seems to have is some top quality games. Sure you have the minesweeper things and sokoban, etc. But you don't have original titles that are highlights of the system. You do get Linux ports that are often than not extremely delayed which is still appreciated, but does have a dilluted impact overall.

I'd like to see Linux gaming go beyond the emulation rings and Wind-nux ports. SDL is a great step forward. Wind-nux ports are very good steps towards recognition. Those aren't enough.

We need original content. And content that will have heads turning. That's the solution I see. Easier said than done.

[Fragadelic]

What about Sauerbraten, Warsow, Nexuiz, Enemy Territory,etc.

All have native Linux ports.

It isn't the Linux community or MS Employees that creates Windows games. They are made by gaming houses and this is where he/she needs to send that email and not to the Linux community.

There is still a fairly small percentage of overall users that game, even in Windows. This is why you don't see many gaming computers sold in retail stores. You could hardly play any decent games on the off-the-shelf hardware you can buy today.

I'm actually starting to reserve my game playing to consoles these days. I've almost completely stopped playing games on the PC altogether.

Just because gaming is important to someone who games, doesn't mean there is enough market share to warrant the money required by the gaming houses to employ developers to port or create for linux.

The entire Linux community still doesn't have enough clout to get hardware manufacturers to create gpl'd drivers for their stuff. Sure some of them provide specs to kernel devs and such but those are not the actions of a company who seriously considers Linux as an important share of their sales.

[niniendowarrior]

Sauerbraten, Warsow, Nexuiz, Enemy Territory,etc.

I know those games. They don't exactly make me thrilled. Enemy Territory from that list is id Soft's Linux port and I appreciate their effort, but it does not change the picture by a whole lot. It is also because of id Soft's more open stance when it comes to GPL-ing code and giving out Linux SDKs that help bring a lot of first person shooters into Linux. Too many, if you ask me. id Soft helps and I applaud their efforts.

On the appeal to the Linux community on gaming, the Linux community needs to get things right. Firstly, it should have a good API development platform. In my opinion, that's SDL. Secondly, good drivers are also necessary and that is somewhat provided. Thirdly, is to not give people a difficult time developing. Too many OS flavours, too many drastic kernel changes and you could have a very unpredictable platform to work with. I've heard quite some complaints about this in the past, although I cannot draw something concrete.

The whole call for more games on Linux is a chicken-and-egg problem, but that's why I think the Linux open source developers should step up and really make significant effort. I'll be frank, I tried VDrift and I wasn't exactly thrilled with how the game worked. I also test drove Trigger Rally and while I thought it might work, at its current state, I'm not impressed. The last game I played from start to end on Linux was Doom 3 and I really forced myself to wait for the Linux client. We can't have that. And furthermore, it's just a port. We need native Linux games. We need good content. We need originality.

[1c3d0g]

Unreal Tournament 3 is the only thing I'm waiting on right now. All other games are worthless to me. I will do whatever it takes to get that thing to work under GNU/Linux. UT3 shall be the best thing since sliced bread!

[Michael]

Unreal Tournament 3 and Quake Wars may get me back into gaming (at least for a while).

[joshuapurcell]

The PS3 has done it for me, even if mostly for the hardware it has. Quake Wars is the next PC game I'm looking forward to, and I'm not positive but I would think that it will be running on Linux.

I feel pretty good about Linux gaming in general... it's way ahead of what it was even two years ago, and it continues to get better every month. There's no way to know what gaming options we will have after this year.

[Janusz11]

Originally Posted by joshuapurcell

There's no way to know what gaming options we will have after this year.

I totally agree here. Concerning Linux and Open Source in general I have the feeling that this year could become very interesting. And concerning Linux and games- well, I think there's still some work that has to be done before Linux becomes a serious gaming platform. But who knows where we are at the end of this year-

[b15hop]

Originally Posted by Michael

Unreal Tournament 3 and Quake Wars may get me back into gaming (at least for a while).

Yeah depends on the quality of the games and if they run at least par. Oh and if the drivers at least work. If it runs better, drivers are good, then it will be like a fresh start all over again.

[yoshi314]

However, to the best of my knowledge, Linux isn't getting anywhere with gaming, and that's due to Linux's main undermining flaws: 1) lack of focused effort,

there are lots of frameworks and opensource 3d engines for linux. go ahead and pick one. some of them are even crossplatform.

single focused effort is just so not-linux. linux is all about diversity. there are almost always a couple of software solutions for a certain problem, be it kernel, libc, editor, desktop, graphics system, or something else. they compete with each other, and at some time they gain a following and eventually set a standard.

lack of desire on the part of developers to be a mainstream solution.

that's just so wrong, i won't even bother with a response.

It would seem that the present time would be excellent for a concerted effort to supply game developers with a tool package that would facilitate the dual development of games for Linux and DirectX 10. I know that AMD/ATI has contributed one part that might help, a HLSL2GLSL tool, but I imagine a lot more has to be done, including hand-holding of developers.

rewriting a game from DirectX into other platform independent solution is no easy task. it's an expensive undertaking.

since the linux gamers community is small, it just does not pay off. besides you have to maintain two codebases if you port a game from directx to ... someting else that also does not pay off.

if you write it to use something crossplatform from the very beginning - that's a different story.

as for that crossplatform game development package - ever been to http://www.ogre3d.org/ ? there ARE commercial games developed on top of this tool. it has at least an opengl and directx backend.

I believe that if Linux could get games running natively, with push-button installation / de-installation and all the bells and whistles, then Linux could really take off. It seems to me that with online retailing, broadband connections for downloads, automatic updates, dual core computing, virtualization (?), better graphics drivers on Linux, and so on, this must be the best time there has ever been for this type of endeavor.

let's start with better graphics drivers first. the rest is mostly ready. this one problem is mainly out of the reach of the community. for now.

And if the Linux community isn't interested in this, I wish they would just pack up and go home.

i wish people who think linux is/should be a 1:1 windows clone would "just pack up and leave". i hate that. why does everyone expect linux to become a windows clone?

I know that sounds harsh, but by chronically doing nothing they are doing more harm than good.

i'd like to see some proof of that "doing nothing". and doing "more harm than good".

People think Linux is going to provide an alternative to MS - I hear it all the time - but if it won't address the concerns of the mainstream, then it isn't ever going to be a credible alternative. All it is doing is reinforcing the MS monopoly by sucking up the time and interest of people who do want a credible alternative.

linux is trying to get into the mainstream. just when it catches up, new problems arise. mostly in the form of closed source drm-ish solutions, which are supposed to help certain vendor gain more market share. that's pretty unfair, don't you agree?