Linux Gaming: Looking Back And Looking Forward

James Hills writes: "Gamespy.com has released their end of the year Linux retrospective , "Operating system historians will record the year 2000 as the year that Linux gaming began to become a serious prospect for both gamers and developers. While many things still need to be resolved for Linux gaming to seriously compete with Windows gaming, companies such as Mandrake, Red Hat, SuSe, nVidia, 3Dfx, Loki, Vicarious Visions, Tribsoft, Hyperion Entertainment, and thousands of programmers working on projects such as KDE, Gnome, and Xfree86 have begun to make gaming in Linux gaming a more mainstream concept. Through the efforts of corporate investors and individual netizens, the Linux gaming market experienced tremendous improvements in all areas last year and the year 2001 looks even brighter. ""

Comments

Re:I've re-installed my Windows partition

[John Carmack]

We are still supporting linux.

The only downside of the next product is that initially it will probably only work at full feature level with the Nvidia OpenGL driver, but after the first test gets out (still a very long time in the future), I will jump back in to the driver development to try and bring the other open source drivers up to par.

John Carmack

Cross platform design is the biggest issue

[MongooseCN]

The biggest issue that needs to be addressed in development is to design the games to be platform independant. You will not get any developers to create games specifically for linux, you will have to get them to create games for both linux and windows at the same time, or at least port their game from windows. The problem this leads to is that developers are not going to rewrite their game twice, hence the creation of companies like Loki, so the best thing developers need to do is to write their game for multiple platforms at the same time, using cross platform libraries(SDL is a very popular one) or better design techniques. There is alot of FUD about cross platform development (extreme performance loss, difficulty in designing...). Currently there is an interesting college research project going on that gives alot of good information about this topic. The server for research project goes up and down a lot so keep trying...

Closed vs Open Source

[1010011010]

For Linux Gaming to succeed, a number of things will be needed:

1. Linux users will have to overcome their desire for Open-Source. Gamers probably don't have this problem, but it bears mentioning. See also #3.
2. Distributiuons will have to adhere to standards. For instance, a game that uses C++ will probably require a special version for RedHat 7, because it's broken. Libc will have to be the same, or compaitble enough. XFree3 vs Xfree4. KDE Vs Gnome Vs whatever for menu icons or base libraries, etc; Etc.
3. Linux users will have to get used to the idea of paying for software. If the games are ripped off left and right, then there will be no money for the people producing the games. And then there will be no games.

- - - - -

Re:Closed vs Open Source

[Black Parrot]

> 1.Linux users will have to overcome their desire for Open-Source. Gamers probably don't have this problem, but it bears mentioning. See also #3.
> 3.Linux users will have to get used to the idea of paying for software. If the games are ripped off left and right, then there will be no money for the people producing the games. And then there will be no games.

You are assuming that "success" is defined by how many units can be sold. What if we defined it by "being able to play lots of nifty games under Linux"? In that case, "Open Source" and "free" might be advantages rather than disadvantages.

Take Freeciv as an example. It's open source, and it's successful in the senses that it has lots of dedicated players and it is still getting attention from programmers even after 5 years "on the market". It has also started moving away from being just a clone toward being an innovator.

How many commercial games still get bug fixes and enhancements after 5 years? In the commercial world, bugs get fixed exactly when businessmen think the fixes will contribute to the bottom line, which is rare indeed. Indeed, a steady stream of bug fixes might be seen as a conflict of interest with getting people to buy the next shoddy product.

Also, your favorite genre is at the mercy of a handful of businessmen. What if you love FPSs, and they decide this year that FPSs are declassé, and move on to the next big thing? You're screwed. No new titles, no bugfixes for the old ones. It has happened before, and it will happen again. With OS games, it's not a risk at all, because it only takes a handful of volunteers to keep a game evolving.

OK, so I named Freeciv. Granted, that's not much. But remember, Linux still has a very small share of the desktop market, and even a smaller share of the gaming market. As those shares grow, mindshare will grow too, and you'll see more games of Freeciv quality popping up.

Remember too that the PC world had a thriving shareware market long before Linux was born. The itch and the will to scratch has always been there. What happens as more and more of the shareware author type become aware of Linux and the joys of OS programming?

My brother used to tinker with shareware games, but he was stuck with QBASIC because he didn't have money for development suites, and of course it takes years for one person to develop a nice game if he can only work on it a few hours a week. But suppose he installs Linux on his system? He suddenly has more development tools than he knows what to do with. And between OS licensing and the internet, he can suddenly round up a handful of like-minded comrades to help make his game become a reality.

I think we are headed for an OS revolution in gaming just as we are in desktops, and just as we have already experienced in servers and supercomputing clusters.

We may even experience a catastropic collapse of the game production sector. Game programmers already work deathmarches for low pay, with a low probability of producing a hit. Any substantial spread of OS gaming is going to put pressure on a system that is already ill. (Did I say already ill? Look at the release-day pricetag of games that are coming out now, and compare that to what they sold for five years ago.)

As for scratching the itch... I quit buying games altogether. I bought a few for Linux, but every time I play one I find myself asking "Why should I play this, and deal with the bugs and screwball features, when I could write my own game, or take Freeciv and tweak it to suit my own tastes?"

If only 1000 people develop the same attitude, OS games will start popping up everywhere. And the more that pop up, the more mindshare they attract.

No, I don't think the desire for open source and the unwillingness to pay for certain kinds of software is going to hurt the "success" of Linux gaming at all. No more than they hurt the success of OS operating systems. The same factors still apply.

--

Re:Games can be open-source

[Datafage]

Games can be open source, yes, but this wouldn't work for all of them, and perhaps even the majority. MMORPGs piss me off, I want a nice FPS or Mechwarrior-style game. You're not going to be able to apply the subscription model to that and have me pay for it.

Multiplayer is where it's at, but not necessarily for a central commercial server. I never played Diablo2 on Blizzard's server after I saw the lag, I always play on a LAN. Same for UT. The purchase model is not going anywhere anytime soon, and bringing up cases where it could possibly work is not going to make everyone decide that commercial server games are the future.

-----------------------

I've re-installed my Windows partition

[maynard]

And am planning the purchase of a PS2. Frankly, gaming on Linux is a joke. How much do you wanna bet that ID Software isn't going to release Doom 3 for Linux, or allow a port? I'm doubting it.

I own Terminus, Unreal Tournament, and about ten Loki games. I'm pleased with Vivarious Visions (and actually like Terminus)... but as many already know I'm NOT pleased with Loki, or their attitude towards this customer. So I don't buy their products any longer. Where is Alpha Centauri, or any reasonable explanation for this mess? Why do they continue to accept multiple pre-orders on their web page when they know it will hold up the delivery of other games purchased in the same order to their customers?

I'd love to see gaming take off under Linux, and think the software infrastructure should be in place once XFree-4 w/GLX and MesaGL, Linux-2.4 w/DRI, and a standardized base between distributions comes along. I'm guessing this will really fall into place by Redhat-8.0, which means about a year from now or so. Until then games authors are going to be forging a new Linux market and eating the support costs associated with the various incompatibilities between Linux distributions. Until this happens though I've given up. After buying twelve or so games for Linux in the hope that I could completely wipe out Windows in order to enjoy a few good games I've now come full circle and re-installed a Windows partition in order to game again. It's easier than dealing with this mess. Frankly, I'll be buying a PS2 ASAP (God knows when they'll be on store shelves) and am thinking of dumping PC gaming altogether.

Cheers!
--Maynard

The ladder to success for linux gaming

[jhol]

What we have here is a classic case of Domino ladies and gentlemen...

For linux gaming to become successful linux itself has to go mainstream, and it's falling short on ease of usability. The average gamer/user is not experienced enough (or do not have the time) to install and learn linux properly and install the games that are available for linux RIGHT NOW.

So once the game companies starts to see that even ordinary gamers/users installs linux with ease and has no problems in learning it then they will finally develop games for the platform and sell them through retail. IMHO linux still has a very long way to go before that happens.

Re:Gaming Learning Curve

[DrEldarion]

Kids today are born and bred on computers

Close. Kids today are born and bred on WINDOWS computers. They're used to everything being handed to them in a nice little package. As the years go on, they think, "Hey, computers are easy!".

Then you show them Linux. A select few will say, "wow, this is damned cool." The rest will say "Why bother? There's no point. I already have my nice-little-user-friendly Windows box right here. Show me ONE advantage that Linux has for me (that I really give a damn about), and MAYBE I'll give it a chance." And chances are, there won't be any advantages to give them.

These kids grew up on computers running Windows. Most of them will stick with what they know.

-- Dr. Eldarion --

Re:I'll sum it all up.

[Black Parrot]

> Linux gaming on the retail scale is a joke.

And it didn't exist at all two years ago.

For the purposes of fortunetelling, it might be better to look at trends rather than a snapshot.

--

Why don't businesses think like businesses?

[KernelBloat]

If I were the president of a gaming firm (instead I'm the president of a free software producing company), I would seperate the code from the content and cut some costs.

Game development costs are ridiculous for the returns of mediocre selling games. Either you make a smash game or you sink. Costs could be cut a heck of a lot if there was more code re-use in today's gaming industry. It seems every little game has its proprietary 3D rendering engine. That's great, but why not focus your efforts where it counts (the fun of the game!).

If there was a free, cross-platform, very fast (as in cpu cycles) rapid game development platform that emphasised modularity, anyone could build games on their spare time using world-class tools developed by thousands of enthusiasts all over the world! All that would be needed is a game production company to want to cut development costs by not wanting to pay for a rewrite of mountains of code.

If the game development environment was to be composed of many interworking (but independant) parts, the useless parts could be discarded when not needed (stopping game developer's fear of bloat :)

Making the environment easy to use would encourage new developers who are only interested in making games (instead of computer programming) to make more games! It would also increase pressure on game production co's to make better games!(nice graphics are a novelty, crappy storylines make for crappy gameplay and no fun)

This only leads to increased prosperity and everything your heart desires. Thank you.

• Python as a scripting language
• Crystal Clear project
• Crystal Space

Re: Re:Gaming Learning Curve

[DrEldarion]

Woah, slow down there Mr. Zealot.

They moved away from DOS because everyone suddenly stopped supporting it. Do you forsee everyone abandoning Windows anytime soon? That's what I thought.

As for the advantages, note that a good portion of people won't care about stability if they don't know how to do anything. Price is almost not an issue, because they buy computers with Windows pre-loaded. They don't know any better. Basically your choices for computers without Windows (that the average Joe knows about) are: Build it yourself, Build it yourself, or.... Build it yourself! I'd say at least a good 99% of computer owners in this country couldn't do that if their life depended on it.

That last statement is getting less and less true as years go on. Windows is MUCH more stable now than before. My win2k box has been up since I installed 2k three months ago without *1* crash. I'm sure many other people have much better records too.

Pull the RedHat box out of your ass and look at the facts, Linux is NOT going to take over Windows in the Desktop market until Mr. Joe Smith can use it with the ease that he uses Windows, and that's not going to happen for a *LONG* time (if ever). That, added to the fact that his favorite software will most likely never exist for Lunix makes the likelihood of Linux as the dominant OS pretty slim.

-- Dr. Eldarion --

Games can be open-source

[xant]

I know, because I'm writing one. There's money to be made on games these days, and you don't have to sell the box the game comes in to do it. It's all about multiplayer environments these days. People want human interaction.

You know what that means don't you? Subscription sales. You give away the source to the game client (hey maybe even the server), set up your world (this should take the most time - this is where you add value, besides the actual server hardware itself) and invite people to play your free client in your world for a trial month. If they like it, pay .. $8 a month for the service. Make sure the world isn't always static (I'm not talking Evercrack here, they've made an obscene amount of money on a pretty crappy service) so players have a reason to come back month after month, year after year, and voila, you have an open-source game that makes business sense.

You don't lose anything by giving away the source; any potential competitors have to flesh out their own world and put up their own server hardware and offer the support that you will provide (because you're not Blizzard, you don't just leave your customers in the lurch when the game breaks down). And some other company did this, and their game looked great, hey I'd play it. Variety is always good, and their code will get back into your codebase. If someone uses your engine to create a fantasy game service, and someone else uses it to create a space game service, they've nailed 2 different, minimally-competing niches.
--