Slashdot Mirror
WineX and the Future of Linux Gaming
SQLz asks: "I'm a Linux user and an avid gamer but unfortunately for me, I have a very limited selection of games to play without having to reboot into my 'Xbox partition' (a Win2k partition with only games). To supplement my aging collection of Loki titles, as well as UT2003 and a few Q3A mods, I use WineX to play titles like Battlefield 1942, SimCity 4, and Homeworld Cataclysm. Apparently this is bad, as many people in the community feel that Transgaming's WineX is discouraging developers from creating native Linux ports. Does anyone have any real proof of this happening? Do developers really point out WineX as a alternative to doing a native Linux port?"
While a 100% Native port would be cool I know lots of the Loki games and a fair few of the other ports from windows Use winelib.
From a Developers poing a view it's perfect. For as much as possible use winelib but if there are problems then port those area's. This means that with minimal effort you get a high quality game in Linux.
Instead of porting 100% just for the sake of it you can port 10% and get the same results. You get more games on linux that run better. As wine gets the games will run better with less porting. The less porting that is required the more likely a game will be ported. As more games get ported more engines will become cross platform to make it easyier for companies to port thier games to other platforms.
The engines are more impotant than the games. Look at http://www.garagegames.com/ about half the games run on linux because the engine supports it. Halflife is another good example. Lots of the mods start as windows only but as they gain in popularity the include linux Support.
The main thing to notice is that Wine is good for Linux as it gets more games on linux which means more engines will support linux (So they don't have the overhead of wine) which means you'll get more games on linux.
>What I am still wondering about is why no game company has created the selfbooting game.
A self-booting game would have to include all possible video and sound (and network, if netplaying) drivers that the player's system _might_ have, and autodetect them flawlessly. While I've had great luck with Mandrake doing this, it's still not perfect.
At least with a PC under Linux or Windows, the user has already gone through configuration hell getting things to work, and non-self-booting games can assume all systems okay and just use the API (DirectX or OpenGL).
That's also the advantage of consoles, actually-- you can self-boot because the hardware is exactly known.
A.
> Profitability. Or lack thereof. Same reason why there aren't more Mac games.
:-( The perceived gains, don't just the cost (of paying developers.)
IAAGD. (I am a game developer.) Take this with a grain of salt.
I disagree. These days you're game is usually running on a PC + console, unless you're a really small developer. (Consoles generate more money, so it's not financially wise to ignore them.) Now you could argue Win32+Xbox, but how many [game] developers are *strictly* tied to the Intel platform?
Do the gains justify porting? From an engineering point of view -- I'd say "Hell Yeah!" You find SO many hidden and potential bugs in your code-base. From a financial point of view, the attitude is "you developers want to take HOW much more time, to get a small percentage in stability?!? Ship it when it's supposed to be!"
You implied the real reason "but theres not a lot of money to be made in the Linux market." Carmack has experienced it, we've experienced, and anyone else who has ported their client** to Linux. The sad and unfortunate truth, is that the BIGGEST reason for the lack of Linux games, is a complete and utter lack of SALES. True, it's a chicken and egg problem -- there just aren't that many Linux desktop users compared to Windows users. Same for Mac users. If they would buy more games, we'd be more inclined to port to Linux. It also doesn't help when the Linux version ships 1 year later, after the Windows version.
I'd be curious in the buying decision when the Win32 and Linux client are seperate SKUs.
Well, that's my take on it. Feel free to debate it.
** There have always been more Linux (game) servers then clients. Standardized networking (TCP/UDP) plays a big part in this.