Most Read Technology Reporter For More Than Two Decades

Maureen O'Gara

Subscribe to Maureen O'Gara: eMailAlertsEmail Alerts
Get Maureen O'Gara: homepageHomepage mobileMobile rssRSS facebookFacebook twitterTwitter linkedinLinkedIn


Article

Mainstream Games on the Linux Desktop

The Realities of a Thriving Linux Games Market

  • Read Ian Bonham's plea for a Games-Based Linux Distro
  • Read yesterday's Slashdot thread on Ian Bonham's article "Is the Key to Linux a Games-Based Distro?"
  • Read more Gaming Round Table highlights here (cross-platform game programming and design)



    Linux Gaming Industry Round Table - Participants' Bios

    • Timothee "TTimo" Besset: Contractor with id Software, responsible for porting id games to Linux. Chances are that if you've played an id game on Linux (Quake, anyone?), you're familiar with his work.
    • Chris DiBona: Cofounder of Damage Studios, which is currently in development of an MMORPG (Massive Multiplayer Online Role-Playing Game) running Linux on the back end. The future includes a client for Linux, though not with the initial game's release.
    • Ryan "Icculus" Gordon: Former Loki freelance programmer working with Epic Games and other established game publishers to port their games to Linux, Mac OS X, and 64-bit platforms. If you've played Unreal Tournament 2004, America's Army: Operations, Serious Sam, or Medal of Honor (and this is just the short list) under Linux, you're familiar with his work.
    • Gavriel "Gav" State: Founder, co-CEO, and CTO of TransGaming Technologies, the people who bring you WineX, a re-implementation of the DirectX API under Linux. If you've played The Sims under Linux, you are familiar with Gav's and TransGaming's work.
    • Joe Valenzuela: Current games programmer, former Loki games programmer, and both the co-creator and current maintainer of OpenAL.




    What reasons do you feel the games industry has for not supporting Linux at this time?

  • Ryan Gordon: Financial reasons, and FUD.First, the market is tiny, that's a given, and that's enough to drive most companies away by default.Second, those who make decisions don't really understand Linux, and their brief interactions with rabid Linux zealots don't help. I get cc'd on all sorts of crazy e-mails Linux users send to game companies. Frequently I hear game producers say that they don't want to think about Linux because of the tech support issues. Heaven forbid they have to respond to a flood of people who have swooped in and inconvenienced them by purchasing their product! Seriously, at this point, you can train your tech support to deal with Linux by saying, "We don't support that"... People are happy to have a Linux port at all, and Linux users tend to be tech-savvy at this point... by the time Linux users are the AOLers and grandmas of the world and need tech support, it shows that you've actually got a large enough market to make it worth it. It's a self-fulfilling prophecy.

  • Gavriel State: Primarily the issue has been one of opportunity cost. Triple-A titles today can take years to develop, at a cost of millions upon millions of dollars. A high-end publisher or studio doing the calculations will often come to the conclusion that their resources are best spent going after the larger Windows market.

    This is not because there isn't money to be made in the Linux market, but because for them, given their expertise at writing original titles, porting to less-popular platforms such as Linux (or Mac OS X for that matter) dilutes their talent.

  • Timothee Besset: Because the market is still way too small? Producers and game studios are unlikely to commit resources to a Linux port because they don't see any appreciable short-term revenue. In most cases the question "Should we release a Linux port?" is raised once the game is already out and has some success. But at that point, the game studio is already busy working on the next project and won't fork out any resources to do it... 

     

  • Joe Valenzuela: Because there isn't a sufficient home market for Linux. There isn't a home market for a lot of reasons. To avoid getting too controversial, I'll point out that even if Linux had its desktop politics worked out, it would be very difficult to make significant inroads into the Windows market. Even mature, well-funded efforts have failed to make many inroads.

  • Chris DiBona: I think that this question is not correctly stated. The industry has completely embraced Linux on the server; you'd be hard pressed to find any online component of a game based on anything but Linux outside of Xbox and Microsoft-related games.

    On the front end is where Linux has traction issues. Why? There are a number of technical reasons that are interesting to consider, and would be nice to overcome, but they're really tertiary to the main issue - that there are not enough spending gamers who run only Linux, so there is no reason for a game studio to consider it a primary platform.

    A number of my colleagues in this roundtable talk about some of the technical difficulties that Linux represents and they're right to consider and attack these problems, but they're all surmountable if the platform is popular enough with the gamers.


    In what ways is Linux already prepared to take on mainstream games?

  • Gordon: I'm looking at UT2004 running in an X11 window while I type this. The technology is clearly there.

    Looking at http://gamespy.com/stats, of the top 20 online games (in this case, this generally means "first-person shooters"), about half of them run natively on Linux...several of the others are built on engines that allow for relatively easy porting of the game client, even if there isn't a Linux version at the moment.

    The greatest asset for Linux adoption has, in my opinion, been its usefulness for running dedicated servers. This has gotten the proverbial foot in the door more times than I can count. Also, every major game engine (Unreal, Quake, etc.) runs on Linux at least as a server if not a full client. This leads to easier porting of game titles built on these engines.

    The problems are frequently a matter of development and management...having developers that "speak Linux" and managers that are keen to keep their games portable during the span of the project. This is a social, not technical, issue. 

     

  • State: While there has been much progress with the 3D drivers in the past couple of years, there is still quite a ways left to go for some vendors.

    Sound support is both good and bad on Linux. There are a plethora of different sound drivers available for different hardware, of widely varying degrees of quality. We work with users all the time who have poorly configured sound drivers, and helping them choose the right drivers can be quite painful.

    Finally, and most important, until very recently there have not been enough Linux distribution players who are seriously focused on consumer-focused desktop Linux systems. This is still a significant impediment to Windows users thinking of taking the plunge. While they have some great choices now, the "big names" in the industry haven't been focused enough on desktop Linux. TransGaming believes that 2004 will see a sea change in that attitude, and we're working with some of the biggest players in the industry to help make desktop Linux a reality for consumers.

  • Besset: Still missing some of that userfriendliness maybe? Sure, the new Linux distributions are easy to install, and provide a user interface for most configuration tasks. But is that enough to convince casual PC gamers to switch to Linux? 

     

  • Valenzuela: Although I could point out that Linux itself isn't really ready for the desktop, I think that would be misleading. The approach that Loki took kind of sidestepped the relative immaturity of Linux package management, desktop software, etc. Installing games on Windows isn't always great either. Again, I have to say it's just a lack of users.

    I spoke with Firaxis once about the possibility of porting Sid Meier's Gettysburg! This was the first title they had the IP to make licenses for, and they were pretty excited about the possibility of doing one. But there was no way to dress up the dismal sales for commercial Linux titles. 

     

  • DiBona: The only one that matters is users, but I'd love to see stronger sound and device support. Plugging in a force feedback joystick without too much trouble would be nice, as would proper six-speaker sound assisted by a good driver interacting with some of the very cool audio hardware out there.


    In what ways is Linux not yet prepared to take on mainstream games?

  • Gordon: 3D drivers. NVIDIA's are great; ATI's are flat-out useless. I really hope this situation will change someday. Better codegen and faster compile times would be wonderful. Visual Studio just beats the hell out of gcc right now. EAX support on Creative's hardware would be nice, but mostly so we can say we've got it.

    It'd be lovely if the glibc maintainers would stop breaking binary compatibility, too. Not that they are particularly sympathetic to those shipping binary-only products.

  • State: Widely adopted desktop Linux is the key to expanding the market for games on Linux. Having very serious players such as Sun, Novell, and IBM working to equip consumers with Linux systems throughout the world will spur hardware vendors to redouble their efforts to improve their Linux drivers.

    Having a more vibrant market will also help convince the large third-party game publishers that Linux is a market to watch more closely. While most of these companies are not currently developing triple-A titles [bestsellers] for Linux, companies such as TransGaming are demonstrating that tremendous consumer demand for Linux games does exist. We hope that our work will serve to bridge the gap for consumers, and serve as a demonstration of the fact that Linux represents a large and growing consumer opportunity.  

  • Besset: I don't have anything specific about games to suggest. Linux on the desktop keeps getting better; users try to convince friends and neighbors to switch...

  • Valenzuela: Not sure if anyone else agrees with my estimation of the teeny Linux gaming market, but I'll defer in case anyone does. 
     

  • DiBona: Time and more users. If Half-Life 2 were only released on Linux, that would help, but until then, it is just a matter of time. If the PS/3 is based on Linux it'll help a little more, but I disagree with one of the other panelists' assertion that OS X will mean more games for Linux thanks to system similarities. It won't. It will drive those interested in a compelling Unix desktop to OS X and away from GNOME and KDE, something that is not good for native Linux gaming.


    What might the Linux community do in order to change the thinking of the games industry?

  • Gordon: Buy games. Seriously. That's the whole equation.

    Game publishers speak one language, and it's American Dollar. Many publishers want a guarantee of 50,000 units moved before they'll talk to you. The big, big names (you know who you are) frequently want more than that.

    In the short term, Linux gaming is going to continue to be as much a guerilla tactic as it ever was.

    With the exception of Loki, a lot of the big Linux games are extracurricular in nature. Some programmer spends nights and weekends porting his day-job project from Win32. Someone has an explicit need to get a Linux dedicated server running, and does the client work while he's there. Someone has a friend at a game company that'll do the introductions so he can get permission to port a game. Someone will cold-call a developer and ask if he can port the game for free.

    All of these things have happened, and happened for top 10-selling games.

    Another interesting area is gaming in the indie/amateur sector. I think it's safe to say that open source game development is, at this point, stillborn. Many games are started; few are anywhere near finished before abandonment. The ones that are finished are crappy clones of '80s retro games. Programmers find they can write great code but can't scrape together free art assets to save their lives. This is unfortunate, but that's how it has been, by and large.

    The (commercial) indie scene, though, tends to put out some impressive titles on miniscule budgets. These people fall into the [serious enthusiast] demographic more often than not. The ones who don't specifically embrace the platform can at least recognize the sales potential. You might not be able to move 500,000 units of Diablo 2 on Linux, but these houses are thrilled to see 1,000 sales, and this is financial motivation to move to any platform where a few people will show up with their wallets. These houses also tend to embrace cross-platform development strategies, such as SDL and OpenGL, sometimes, I think, by lucky accident; this gets them running on other platforms faster. 

     

  • State: It's very simple. Buy more games and tell the industry that you're buying that game to play on Linux. Whether that's a Windows-only title that can run using TransGaming's WineX technology, or a title that has a Linux ELF binary that can be used to suppliment a Windows boxed title, let those publishers know that you're buying the game to play on Linux.

    Let your hardware vendors know that you're using their hardware on Linux, and give them feedback on how well you think their Linux support is doing. For those vendors whose support is not yet what you'd like it to be, write them and tell them nicely that you appreciate the efforts they've put in so far, and that you're looking forward to future revisions with bugs corrected.

    When we talk to people in the industry and tell them the number of users who are currently running WineX to play their Windows game on Linux, they're frequently surprised to hear that they already have thousands upon thousands of Linux customers. When they hear that, they begin to realize the potential of the market, and they begin to think more clearly about how they can better serve those customers.

  • Besset: Keep harassing them, and buy the Linux games that are already out there!

    There are several reasons why a game company should support Linux though. The main one would be that doing a Linux port and a Mac OS X port are not much different nowadays. And there are customers for games on the Mac. Clearly not a big amount compared to Windows, but still a lot more than Linux. Another reason is to make your code more robust. Your source will have to be cleaner, and you'll have to avoid ugly OS-specific hacks. Some bugs are easier to track down on one system or the other. Releasing your game on several platforms is a proof of technical expertise. 

     

  • Valenzuela: Mac OS X has shown that in order to even have a "niche" in the PC market, you have to be twice as good as the competition. If Linux had a similar group of people who would regularly buy six or seven titles a year - even a small group, say 50,000 people - it would be possible to sustain a market similar to the Mac gaming one.

    Even today, many triple-A titles find themselves on Linux. The difference is that these seem to be labors of love as opposed to money-making efforts, so future prospects are subject to the vagaries of whim. But since Ryan is probably responsible for 95% of these I'll defer to his thoughts on the subject and won't speculate. 

     

  • DiBona: Well, have patience. With time and enough hard work, more people will use Linux on the desktop, and that will bring the developers with it. Also, they should take a hard look at some of the cool games already out there for Linux.


    State your position on technologies that allow Windows games to run under Linux, as opposed to games that run natively under Linux.

  • Gordon: Honestly? I don't lose sleep over it. I don't believe it is preventing games from coming to Linux...there is anecdotal evidence that Game House X heard about WineX and decided that there was no reason to support Linux directly. I say these people never had any intention of thinking about Linux, so WineX makes a nice scapegoat.

    I think everyone, including TransGaming, can agree that a native Linux game is better than an emulated Windows binary. The sticking point, really, seems to be if the emulation is acceptable in lieu of a native version. 

     

  • State: Using TransGaming's WineX technology, Linux users can run games that they would otherwise have no access to whatsoever. From that perspective alone, we believe that we're doing the market a significant service, and we've got thousands upon thousands of customers who agree with us.

    Our technology is frequently classified as an "emulator," but that is no more the case than Linux being a "Unix emulator." WineX works by providing a real, highly portable implementation of the Win32 APIs required by most game titles, including the various DirectX APIs. WineX has itself been ported to a number of different hardware platforms, including the PowerPC-based Mac OS X, the MIPS-based Sony PlayStation 2, and a variety of other hardware.

    On x86 platforms, TransGaming's WineX is capable of directly loading and executing Windows PE format binaries. But the only way that this loading process is significantly different from loading a Linux ELF binary is that the Windows PE binaries are frequently more highly optimized due to higher quality compilers available for Windows.

    On the performance front, things vary depending on the title. With some of the technology we have in the lab right now, we're seeing games that perform absolutely 100% identically to Windows. In other cases, when we've done Linux-specific optimizations on games such as TimeGate's Kohan series, we've seen performance that's significantly higher than the same code running on Windows. We're working hard toward making that happen for all the titles we support and we're also working more and more with game developers and publishers to help release Linux-optimized versions of their products.

    Ultimately, seeing a thriving market for Linux games helps to wake up the rest of the industry to the potential of bringing their titles to the platform. We're in talks with a number of companies and want to reach out to more in order to help them realize the potential of the Linux market. If they don't have the expertise or the resources to deploy their games on Linux, we are happy to work with them directly to bring their content to the platform. 

     

  • Besset: I'm not a big fan. It is a valid business model; it gets you to sell licenses of your emulation software right here, right now. I'm sure that's reason enough for a lot of people.

    As a game company, is it a good thing to do all the development for a Windows platform, then expect Linux and OS X users to play it emulated? You might get a few sales without too much effort, but you won't have the technical benefits of doing the development on Linux.

    As a game buyer, should I buy the Windows games and run them emulated? They will run a little slower, is that good enough for me? If everyone is buying the Windows games and running them emulated, companies won't release native ports anymore. Guess what? At this point you have a Microsoft-like company controlling gaming on Linux through their emulation technology. Not mentioning the fact that the more game companies working with Linux there are, the more pressure on hardware vendors to release good drivers. 

     

  • Valenzuela: If I thought Linux had a long-term chance for a sustainable gaming market, I'd say "100% against." For the moment I'll say I'm 70% against.

  • DiBona: Bah. If you want to run Windows apps, use Windows; if you try to fit a square peg in a round hole you'll always get caught on the sides. I'll admit that for some this is a decent solution, but I do think that having emulation bring Windows games to Linux will delay making Linux into a game platform with any kind of parity with Windows or OS X. That's not a bad thing if you just want to play a game on Linux, but it is if you want Linux to be a premier platform for gaming. I'm just glad Icculus [Gordon] and the other panelists are here making Linux safe for gaming. I like playing games, after all.

  • More Stories By Dee-Ann LeBlanc

    Dee-Ann LeBlanc has been involved with Linux since 1994. She is the author of 12 books, 130 articles, and has more of both coming. She is a trainer, a course developer - including the official Red Hat online courseware at DigitalThink - a founding member of the AnswerSquad, and a consultant.

    Comments (18) View Comments

    Share your thoughts on this story.

    Add your comment
    You must be signed in to add a comment. Sign-in | Register

    In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.


    Most Recent Comments
    Gerry de Caires 06/08/06 06:42:59 AM EDT

    I think that one of the primary concerns for the gaming industry is quality control. MS users, by nature, are accustomed to problems ranging from hardware, software and bad design and as such, they will spend time after buying a product, just trying to get it to run. Take for example, the Civilization 4 game. When this was first released, it was lapped up by many Civ gamers, myself included, only to for us to find out that the game was buggy and caused serious issues with memory leaks and crashes. We all awaited the forthcoming patches that would let us fully use our $49.99 game. Translate that into Linux and Open Source in general and the criticism would be much higher on the manufacturers and I would bet that the "users" who purchased this game, having access to the code, would hack it to make it work and distribute the fixes to the community. Open Source users in general demand a higher level of work than most MS and indeed proprietary software users. Once the gaming industry realises that the Bazaar model used in Open Source can only benefit them, they will come around. Their games will have longer shelf lives and mods which always inspire those who are bored to keep playing.

    Pointless 07/19/04 12:26:38 PM EDT

    The simple reason why my friends and I don't use linux for gaming, is that the frames per second are poor. The data flow between X and the video card are just not up to par. It lacks adequate hardware acceleration and load balancing between the video card and processor. Even my Vodoo 5500 in linux only uses only one of its processors. I have spent hours upon hours trying to figure out how to make it work. But, the final result seems to be, the fps will not be there for about another 1-2 years (maybe sooner since the focus of linux has seemed to change). Who know what Novell and IBM are going to do. For awhile, I was doing windows and linux dual boot to get to my games. Now, I only need windows for some work related functions. I went ahead put all my games into storage (yeah I had a book shelf full of them). I have spent more money many times over on my PC based games.

    Simply, optimize the video on linux til it is compariable to windows and many gamers will switch to linux.

    For now, my friends and I will have to settle for Xbox (yeah I am not a linux zealot), PS2, and gamecube.

    Jay Swackhamer 03/19/04 12:42:46 AM EST

    There is a growing market for Linux games. Linux Game Publishing is helping. The work done at icculus.org is helping, Quake/UT2003/UT2004+ many others give more visibility.

    Most of the machines that go out of my doors are dual-boot, so that people can play their existing games under Win, and do everything else under Linux. By pre-loading the open-sourced games, and Enemy-Territory, America's army, +others, less and less people are using their Win installations after a while.

    The Activation issues with Hardware upgrades are enough to make some people just not bother with Win any more..........since gamers tend to actively update their systems with new hardware as it becomes available......

    If a game doesnt run on Linux, I wouldnt buy it, and as the new games get released, I'll be adding more and more titles to my shelves that are Linux-Only, or include the Client(UT2004,UT2003), but none that are Win-Only.

    Frank Earl 03/18/04 06:57:06 PM EST

    JonathanWizartar comments on a lack of a standard API like DirectX...

    Well, you're wrong in that regard. (In fact, you partially shoot yourself down with your comments later on...)

    SDL is a standard API for handling input devices, basic sound playback, 2D framebuffer operation, and 3D access through relatively seamless dynamic loading of OpenGL support through Mesa or comparable on the machine...

    OpenGL happens to support the latest graphics technology, contrary to what you claim...

    OpenAL happens to provide enhanced sound support and positional audio- for Windows, Linux, and MacOS...

    There's not a standard for network play- YET. Right now, I'm kind of tied up trying to get the network multiplay bugs caught in the beta cycle for Ballistics fixed so we can ship it, worrying about demoing some AMD64 based games at GDC and giving a 30 minute talk at GDC about the market for Linux Games, and helping Epic Interactive work on their multiplayer support for one of the games they're porting that we're publishing. As soon as some of my time frees up after GDC, one of my priorities is to do whatever I can to resolve the standard multiplayer layer issues (may end up being two or so of them, depending on what kind of game you're doing...). Front runners for the initial codebase for this effort is OpenPlay and ANet (Do a Google search for both...). Both network APIs have been extensively used in multiplayer games on Windows, MacOS, and Linux in the past.

    Video playback is something of an issue, but I suspect that there's some solutions in the works that are better options than smpeg (which is the current "standard" and is lacking because it's only MPEG1 compression...).

    Oh, on the subject being too hard to install, that depends on whether or not you do some careful research on what to purchase- which is actually NO different than with XP. Do you HONESTLY think it's simple or easy to get XP up unless you've picked something like an NVidia video card and a comparable sound card, etc? Besides, I've got a stack of live CDs sitting next to me here that don't seem to have a single shred of problems detecting and setting up for the hardware that's known to be fully and properly supported under Linux. No intervention needed in the case of Dynebolic or Knoppix- just pop the CD in the drive and GO.

    The same can be said for Supercomputer, Inc.'s little bootable disc that is supposed to be shipped with every Athlon 64 motherboard- it's a 64bit Live CD that boots up, and runs America's Army with NO user intervention in 64-bit mode. We're working on our own version of this for some of our games.

    JonathanWizartar 03/18/04 08:11:06 AM EST

    The biggest problem to game development is a lack of a standard up to date API like DirectX. No matter how much you think Linux kicks Windows ass, when it comes to this type of thing Windows is miles ahead. There isn’t enough support from the Vendors in creating such an API under Linux either. It’s taken Microsoft about 6 years but its now quite good to use. SDL is okay and so is OpenGL but they’re behind the times with it comes to all the latest GFX technologies. There also needs to be unification in the desktop, even if this means creating a new “Games” originated desktop used for primarily for games, while leaving KDE, GNOME, XFree, etc to there own devices without game support.

    Also Linux is still a bitch to install on most systems. Offices have the technological know-how to support Linux on the desktop. But your average home user wouldn’t have a clue how to do most of simple administration tasks. Too many problems with library versions and lack of a simple as in automated way of keeping everything in sync with other programs. But this is starting to get a bit better, but not much.

    Building your own Kernel! That’s what 99.9% of game player’s market pays for when they buy or download a distribution. It’s meant to work out of the box- batteries are included. I know of only one distribution with a specific games kernel.

    anonymous coward 03/18/04 07:00:47 AM EST

    Anyone ever heard of knopix?....... Thought so. I have not had a system that it would not load on and run properly. What does this have to do with games? Every major game programmer out there focuses on installing their game on an existing os. Where is the imagination?????? IT IS POSSIBLE to make a completely self booting, self aware, self configuring cd that would load the os and all the proper drivers etc. for whatever game you would want to write. Don't tell me the cd cannot hold enough data, drivers DO NOT take up that much room. Think about it, the game would be universal, no OS dependence at all. Oh, by the way, linux is light years ahead of the any other comodity os in this area. A short walk on the web will reveal literally dozens of self booting linux versions, all the way from several dozen k to a few hundred m. No, not for everyone, but don't tell me there is a market limitation based on compatibility when in truth it is only a self imposed limitation.

    John G. 03/17/04 07:54:15 AM EST

    Uhm. Please don't start with dependancies on KDE or GNOME! I have NP with ie. GTK+ or QT. But not a whole DE. Or making it horrible on non-KDE or non-GNOME. Not everyone runs either of these, even Linux gamers don't.

    Johann Seidenz 03/17/04 07:48:28 AM EST

    I think a viable solution could be found in bootable dvds. There could simply be a linux distro similar to knoppix which is bootable from a dvd which contains just the basic os needed to run the game + the game itself. Then you basically slip the dvd into your computer just like you would a gamecube game in a gamecube.

    Danni Coy 03/17/04 06:43:22 AM EST

    I am what you might call a casual gamer - I might buy say one title a year (2 at the most) Its not that I don't like games it is just that they take up too much time. I usually prefer games I can play socially... I like making things so I prefer games with a good deal of customisability.

    I am also a Linux user. These days if there is no Linux port I don't bother. Fortunately most of the titles I have been interested in have linux ports. Neverwinter Nights was my last purchase and I am looking forwards to Unreal Tournament 2004....

    I have two Issues with Linux gaming however. One is that fact that I can't walk to a shop and buy any kind of game that runs on Linux. When I ask the help just stares at me blankly. The other is that I would probably spend more time using the level editors etc than playing the actually game it is a pitty that these are not usually included.

    Anyways I wonder how many other people out there are like me.

    Quintesse 03/17/04 06:17:09 AM EST

    Doing the same but using Java and JOGL/JOAL (https://games-core.dev.java.net) I could use Xith3D (http://www.xith.org) as well, but I just like making my own framework.

    Simon Porter 03/17/04 05:23:45 AM EST

    Actually one interesting thing I've been looking at recently is developing games using C#. With the Microsoft .NET Framework and Mono it now means the possibility of being able to run your game on both Linux and Windows without having to make special versions for each. http://axiomengine.sourceforge.net looks to be an interesting C# based game engine

    Frank Earl 03/17/04 02:23:18 AM EST

    I have to beg to differ with D.Smith in some regards.

    Choice does present problems, yes. But it's not the main problem with Games on Linux. Nor is a lack of a "handle" or money directly involved with it.

    Ryan's pretty much pegged it in his comments in my not so humble opinion.

    You want games on Linux? Buy Linux games. Don't buy Windows/MacOS games. While Gavriel would have you buy them and use WineX (which is not a bad idea for older games you might have had in your possession), each and every purchase of those titles counts as a WINDOWS purchase, not a Linux purchase- so the publishers see nothing but Windows purchases. Which is the OS you choose to run under? Shouldn't you be buying software FOR it if you're buying software?

    You want better hardware support? Buy the fully supported devices and let the manufacturer know WHY you bought.

    D. Smith 03/16/04 11:57:50 PM EST

    The main problems with Linux are its main asset: choice and freedom. Most of the people that I see using Linux (I'm the only one that I know of within miles of me), most of them take some perverted pride in having hacked together their system using old garden hose and a gopher carcass plus the guts to a Sylvania TV set. The OS is not a choice made solely for security or low-cost. It's also a choice driven by intractable individualism. Thus, there are just about as many "distros" out there as there are users. The Linux community should be (and is, I hope) deciding upon a dead-bang set of standards for what might be termed "vanilla" Linux, and I don't just mean the kernel. I love KDE and GNOME, but the "us vs. them" mentality that those two interfaces generate often has me wishing that one of them would just die (not to mention the driftless morons who insist on running things from the CLI or using a "thinner" interface. Because of its mercurial nature, Linux lacks a handle for any software company to grasp. The gaming industry (like any entertainment industry) is run on money, and Linux runs contrary to that. You see, Linux encourages a cheaper alternative born of altruism that places functionality and human support above saleability and packaging. Linux gaming was at its zenith when Linus first pposted the kernel. Until Linux looks much more like its competition, not many are going to see Linux as competitive. An option is not enough, a closely similar option is needed. Ford doesn't compete with Maytag, it competes with GM and Chrysler and the Asian and European car manufacturers. Windows doesn't compete with Slack, it competes with OS X and possibly Red Hat (Fedora). In my opinion, Fedora should be mailed free of charge to every man, woman, and child in the US and Canada, and a serious effort should be made to give customer support to all of them. Then let the elite of them decide to go Slack or Gentoo or Debian aftef they've switched. However, I'm not naive. Unless they see substantial parity and spectacular savings, they'll probably make coasters from the disks.

    Byran Lee 03/16/04 09:12:01 PM EST

    People *PLEASE* get a clue. The reason linux users like me don't buy games is quite simple. There quite simply aren't any games on the PC platform that appeal to me. I abandoned the PC *YEARS* ago and bought a PS1 and then a PS2. It'll be a cold day in hell before I waste my money on a PC FPS title when I can find far better game titles on the PS2 like Grand Turmiso 3/4 or the Final Fanstasy Games.

    Jonah Hex 03/16/04 08:27:09 PM EST

    First a bit of Linux history reiterated, just to set the scene... Linux has gained the widest adoption in the server arena, probably the largest segment being dedicated Internet/Firewall boxes. Running without a GUI or even headless (without attached monitor/etc), configured via Web interface, even using "lower end" hardware. These are perfect canidates for adding more server software to, in this case we'd add a simple way to install and configure game servers.

    The advantages are obvious, we can sidestep the entire issue of "does Linux support the latest games" since most modern games come with a Linux server component www.icculus.org that is superior to running a server/client at the same time on your Windows box. (added server overhead usually only if you are serving the game of course)

    Considering the wide variety of dedicated firewall distros out there, my own favorite www.clarkconnect.org, and how easy they are for even Linux newbies to setup, there isn't much of a leap to create such a distro. One of the main issues of course is getting the needed copyrighted content from the game CD to the Linux box, something that is both a manual and painful process in many cases.

    I've discussed this project with a few friends who are familiar with Linux firewalls/game servers also, but unfortunately we haven't had time to do more than brainstorm on it. If we do end up putting together a basic distro like this you can be sure I'll submit it to www.slashdot.org games section. Anyone super interested can email me.

    Jonah Hex
    You have not brought any marijuana for me, or lavished on me the fat of your sacrifices. - God

    loco 03/16/04 07:45:24 PM EST

    I think that the focus of this article is misguided... Linux's potential to change the gaming industry isn't as just another desktop where you can run the same games that Windows/OSX can.

    Anyone who has used Knoppix, the most widely used Linux "live" CD, knows that Linux's most valuable asset is its ability to run on a large variety of hardware. If used as an "embedded" platform for games, Linux's open source roots would allow game companies to take advantage of more target hardware and develop games that could be run on a variety of devices. This hardware-agnostic approach would lead to the possibility of much more immersive games that could be enjoyed at any time with the hardware at hand (handheld, phone, desktop).

    Mike Wagman 03/16/04 07:43:18 PM EST

    I code. I've written a small game. It's in python and I've released a tgz for linux. I promoted that one some linux sites. I got what I feel was alot of traffic considering I'm a hobbiest trying to sell some shareware. I hate to admit it - but I am getting more windows downloads than linux from the referals I'm getting from linux orientated sites (names withheld to prevent emberassment).

    SiawnOverseas 03/16/04 07:26:46 PM EST

    Avid gamer of roughly 15+ years here. I'd love to put my money where my mouth is and buy Linux games...but not 6-12 months behind a Windows release. I have dual boot systems right now, Linux to do most of my business but have to keep Windows around for games. I jumped onto Neverwinter Nights when they committed to a Linux server and client...it was an awesome thing to see developers finally doing so. BZ to Bioware for that.

    If games were released at roughly the same time I think you'd see more sales. As it is, I'm already one of those folks you rely on so much...the got it have it yesterday type. Waiting is not an option.