Pushing Linux Adoption Through Gaming 269
An article on CNet questions the viability of using games as part of a strategy to increase Linux adoption. It points out a blog post by Andrew Min which suggests:
"... Linux companies also need to start paying attention to the open source gaming community. Why? It's lacking. However, gamers can get excited about free games. They just have to be up to par with commercial games. The problem is, commercial companies pay hundreds of employees to build a game for several years, while many competing gaming projects only last several years before the developer moves on. It's time for open source developers to start getting paid for their jobs. Who better to pay them than the companies that benefit most?"
Paying the OS Game Developers (Score:4, Interesting)
What a bunch of wank (Score:3, Interesting)
What has actually been observed to increase adoption (citation needed) is fancy crap like wobbly windows and spinning cube desktops.
Maybe collectively the companies could make a "content light" face booking, im-ing spinning flashing version of linux and attempt to lock up the teen market, i think you might find that would be of more interest to more people than the marginally smaller "hardcore pc gaming" crowd.
Personally i don't really care.
Re:Open Source Games... (Score:4, Interesting)
Or as an alternative, working on a commercial game and releasing the engine code as Open Source (like id software does).
You don't actually need the games to be completely free to users. But if the code is available for Open Source developers to port to Linux with the existing art (that costs money), then Linux still gets a boost. Of course, porting a Direct X game to OpenGL is a pain in the behind that is going to make the release lag a little.
But a lot of people here only claim to have Windows installed to play games.
Re:Late to the Party (Score:3, Interesting)
Not so fast. Gaming on Linux doesn't have to be on the PC. The PS3 runs Linux well. It has decent hardware for gaming purposes already built in. If game developers were to develop games that would run on a PS3 Linux using open APIs, it could be easily recompiled to run on a PC running Linux. Now you've developed for two platforms at the same time.
What we need is very real and very serious (possibly commercial) support of Linux on at least one game console to make this work. Once game devs see that their competition is making more profit on a model like this, they will start to migrate to that platform. So perhaps the console manufacturers need to look at this. Sony did a half-assed attempt at this years ago, but if they (or a third party) introduced official support for a Linux OS on their console, they could make Linux gaming happen and bring more devs to their platform.
Re:Open Source Games... (Score:1, Interesting)
If you want to standardise parts of gaming (I don't understand really what opportunity for standardisation there is there, or what is stable enough for standardisation) then let's hear which parts you think are suitable.
A little bit more complicated. (Score:5, Interesting)
I'm all for getting games away from Windows, because I remember DOS, I remember games running fine on DOS because the OS wasn't trying to do all kinds of crap under them, and I remember just about every version ever Windows breaking games that ran just fine either on DOS or older, lighter-weigh versions of Windows.
And really, this is a big reason PC gaming sucks compared to the consoles. Consoles don't have to worry about whether they need to be doing all kinds of other crap at the same time; PC's running Windows do (and this is more true with each version). Same goes for Mac, and frankly, same goes for any mainstream distro of Linux.
So one big thing that needs to be overcome is how to optimize Linux so it's actually better for gaming than Windows or Mac. Do you strip it down and get rid of stuff games don't need, come out with a gaming-specific distro? Or do you work on making the internals as fast as possible in ways that matter to games? Or something else entirely?
Get Linux to the point where things run better on it than on Windows or Mac, on equivalent hardware (since it is equivalent nowadays), and you might attract more game development.
The issue of artists someone pointed out is the other big issue. You need to motivate the artists. And - especially if you want them to work for free - you need to give them something really compelling. That means something OSS that's better than what they have now. Something that beats DirectX, beats OpenGL, or whatever. I don't know whether adding OpenCL support like Apple is doing will help - that seems more aimed at offloading processing tasks to the GPU, not offloading graphics tasks to spare CPU cores.
But in both cases, I think Linux is going to have to be a clear "best choice" before game developers will flock to it. Make it outperform other OSes in game execution as well as graphics and multimedia, and make compelling tools or toolkits for developing games and the graphics and multimedia they need, and they will come.
I honestly don't see it happening, though. :(
Re:The way I see it... (Score:3, Interesting)
The way one plays windows games via Wine is to create a separate account with X that only spawns an xterm. One gets better performance than Windows does, doing this way.
Perhaps one could load a 2d Windows platformer via Gnome or KDE, but it'd probably be slow. oh well.
Re:Open Source Games... (Score:5, Interesting)
Re:Open Source Games... (Score:4, Interesting)
On the other hand, nowadays there's less and less proprietary code in games. Everybody's licensing Havok's physics engine, for example. That's code that would either need to be released by the developer as open source (ha ha--when they make their money from selling the stuff, no way in hell will they give it away), or replaced wholesale (which is likely to not be as good, and would, at least in the case of gameplay-affecting physics, bork any hope of crossplatform multiplayer).
That also completely ignores the unlikelihood that any multiplayer code will be released, for fear of exposing vulnerabilities. Rather than dealing with paying to fix them, publishers would rather hide the code and hope that it works.
Re:The way I see it... (Score:4, Interesting)
I honestly don't know what you're smoking. I submit the following example:
Running World of Warcraft + Firefox (and system monitor) in Vista = 15 frames per second, over 2gb RAM usage
same hardware, ... WoW for x86 isnt really meant to run in OpenGL mode, and like I said, under WINE, and I'm getting four TIMES the performance?
WoW (under WINE, no less) + Firefox (and system monitors) in (a fresh install, no tweaking) of Ubuntu = 60 frames per second, only 700MB total RAM usage.
J
Re:Open Source Games... (Score:2, Interesting)
Personally, I feel that Linux's file system is even more of a sewer as the Windows file system, and until it takes a major jump up... perhaps it doesn't deserve to be everyone's desktop.
That having been said, an OSS or Creative Commons license texture, object, skeleton, animation, sound, etc repository might be helpful, especially if standardized around specific file formats. There would need to be some recommendation system to tell people which resources can go with what, as some OSS games take the "melting pot" theory of art (which just looks terrible).
Also, we should look to the sorts of evolutionary development that OSS games can do, but retail cannot. Which is to say, focus on making your engine and code as modifiable as possible, release a minimal chunk that shows the greatness, and shephard your players to help build. That's the theory, of course. While most OSS software is actually build by one or two people, MUDS have shown that if you put gamers explicitly into the role of creators you can get amazing community-built experiences.
Keep it scripable, keep it modular, explicitly let anyone play and build, and maybe something will come out that will make the windows users jealous.
Myst Online will be Open Sourced (Score:4, Interesting)
This is a start [mystonline.com]
What is the problem? Please get a move on. (Score:3, Interesting)
There are plenty of gamers who are pretty smart, and ready to make content using modeling and content creation tools, (www.racer.nl - huge libraries of fan-made performance cars and stuff imported from other games). Gamers are highly conditioned to not paying money for games... and ready to bit torrent anything they want at the drop of a hat. So I think all the ingredients for a OSS gaming revolution is there. What we need is a few killer projects to get things going. The few OSS games now aren't very good - they pale in comparison to some excellent indie games out there.
Re:Open Source Games... (Score:3, Interesting)
You hit the nail on the head, sir. It is not uncommon to see tons of artists and designers, but only 5 programmers in a commercial game development team.
I wonder though if one successful open source game - not just a quake3 mod, but an entire game including top-notch design and custom-made game art - would kickstart a wave of similar projects. At the very least, it would serve as an example how it can succeed.
Re:Open Source Games... (Score:3, Interesting)
Personally, I feel that Linux's file system is even more of a sewer as the Windows file system, and until it takes a major jump up... perhaps it doesn't deserve to be everyone's desktop.
What do Linux filesystems have to do with gaming?
I believe GP is talking about the linux file structure (/usr, /etc, ...).
Pretty much what I had to assume to make any sense of it. Especially since there are multiple file systems. I prefer ReiserFS! I don't care that there is little doubt that he had killed his wife... the filesystem is still, to use the words of Peter Griffin, "Freakin' sweet". After all, I accept as a fundamental part of my world view that products of a person's life are not all interconnected. A murderer doesn't _just_ murder. He does other stuff... and if he did a much better job at the other stuff, then why not accept it anyways?
In any case, I definitely agree that the whole /usr /etc etc stuff kind of sucks. Programs should be more self-containing and such... for a good example, see how Mac OSX handles things. Great idea. I understand the superficial benefits from having all your bins in the same directory, however that's just really worth it for CLIs, and stuff... when you're getting into non-command-line interaction... it's way better to have things encapsulated. It's like the difference between OOP and non-OOP... grouping things together doesn't just make so much sense that C programmers already do it anyways... but it also makes things manageable.
God, I must be tired... I'm rambling... Bed time for this girl, good night...
Re:Open Source Games... (Score:3, Interesting)
Have a look at taspring (http://spring.clan-sy.com), that's a great open source game with a huge amount of user contributed mods and maps.
Making windows users jealous isn't going to work, because open source games will inevitably be ported sooner or later anyway.
Re:Paying the OS Game Developers (Score:3, Interesting)
Too many people forget that if its not Microsoft its not DirectX/Direct3D.
OpenGL runs on the following platforms:
Apple - OSX
Sole 3D API of the PS3
Same for the Nintendo Wii
Google Android
Linux and other *NIX OS
Its also very competitive with D3D and is just as feature rich. Many people have the misconception that its more difficult to work with when in fact it is not. My brother has a degree in game development which pretty much means a CD degree in game development. He is writing his own turn based strategy engine for an RPG idea he has. And guess what 3D API its getting written in? Thats right OpenGL. I asked him why not D3D and his answer was pretty simple: "Its not any easier and it locks me into Xbox/Windows only. I could port it to my Android phone if I wanted."
But its not just API's that we have to worry about. Too many games today are shipping with DRM like starforce and securom. Both of which have been defeated with simple patched EXE's. Also Both have caused countless headaches with consumers. So hopefully publishers will see they are pointless wastes of money and embrace Linux as a potential platform without worrying that it cant be infested with DRM.
With the kludgy mess Vista appears to be Linux can really shine in places closed platforms cannot. Look at the debacle SLI is. I have been fighting for 2 weeks now to get two 8800GT's to run Crysis. I don't think windows is solely to blame as I place more of the blame on Nvidia and then on to Crytek. But Microsoft is responsible its API's and driver architecture play nice together. It doesn't appear to be doing that.
With ATI open sourcing chip specs and Intel to do the same with larabee, linux can develop a tight integration of all the hardware and expose the GPU to the developer like never before. Better management/scheduling of GPU's so multiple GPU applications can run at once. Physics engines written using OpenCL can balance the load between the GPU(s) and the CPU(s) seamlessly. OpenGL threading will allow multiple OGL windows. An example would be a full screen game on one monitor and another monitor with a Compiz desktop running on the same GPU with no performance hit to the Compiz desktop. Hopefully with things like OpenCL, OpenGL, OpenAL etc. Linux can give developers a real stable and flexible development environment.
Oh and one last thing! We need an IDE to compete with Visual Studio. Thats one thing my brother hates about Linux and why he isn't working with it yet. I am showing him Eclipse and he is fooling around on it with the Android dev kit but not too impressed with it. Give it time and hopefully it can compete.