XreaL: The Most Advanced Open-Source Game Engine?

[Raedwulf]

(no prior judgement intended) well, it is clear that the codebase is a mess, and raw GLSL code is scattered all over the place, resulting in badly maintainable code.

Take a look at the originally released quake3 source code, hey this is loads better.

And with only a bit of copy-paste-and-fill-in knowledge, it takes only 5 minutes to get a build system working for any project. The hard part is what comes next: automating unit tests, package installation, etc.

Unit tests would be a nice thing, but I've not seen them used in many other game engines either.

That was my point. There's no better in open stable drivers available, with Intel maybe an exception, but you can't play using those cards anyway.

Theoretically Intel cards could work, but they suffer from some bugs which cause the xorg server to crash when running XreaL.

With Mesa 7.2, it was possible to run the menu from XreaL, and hang when going in game.

With Mesa 7.4, the menu hangs.

As far as we (#xreal dwellers) know the software render path of Mesa works, but Intel drivers have bugs still.

I donno. The screens don't blow off my socks to be honest and this has nothing to do with the artwork. But maybe this is indeed a bit an unfortunate map to showcase who knows.

Artwork is definitely the problem - details in terms of higher-poly models, better diffuse/normal/specular maps all contribute to the scene. People gape when they see massively reflective scenes, and think its cool - xreal supports that, but having a map do that looks pretty silly :P.

The engine is in development still, but is nearing feature-completion. However, its main problem is that the XreaL project currently has no artists. It has been an almost solo effort - so this is quite an achievement.

P.S. Hint, If you are an artist, and need something to do, head over to #xreal, and Tr3B will have his mile-long scroll art TODO list.

EDIT: Noticed the article missed that XreaL supports skeletal animation, something that none of the popular opensource FPS games implement (there's MD4 support in ioquake3, but I do not believe its been utilised much). This would be crucial if one wanted to implement ragdoll physics. However, realistic physics hasn't been a goal of the engine, as XreaL has a tendency towards Quake3 style gameplay. However, I'm sure modders would be welcome to add support for it.

EDIT2: Binaries are included in the SVN only for Windows. Linux binaries are not included and the chief development environment is on Linux. The sole reason is that there are a lot of windows gamers/modders/mappers who want to get started quickly, and setting up a build environment on Windows is darn difficult for non-programmers (and can be quite annoying for programmers too!). On linux is pretty much runnning scons builds the project, so its no problem.

EDIT3:

I wasn't really complaining, merely stating the fact, and asking why an OSS game would use features you can't use on OSS drivers.
To me that defeats the point of being an OSS game.

That is a very strange question. First of all, XreaL is meant to use the latest capabilities of graphics cards. Some open-source drivers do not support the latest features, and in your case, features that were present 3-4 years ago. XreaL is OpenGL 3.x compatible, which means it is NOT OpenGL 2.x, 1.x backwards compatible. I repeat XreaL is designed for new architectures. Its bit like saying, "Why can't I install Ubuntu x86-64 on my 386 computer?" Granted, these are drivers, but the drivers have features pertaining to an older generation of graphics card.

You can't fault an opensource game that faithfully implements the newest OpenGL spec because opensource drivers do not support the spec fully, or only provide partial support. In fact, you should be encouraged that an opensource game has managed it and before commercial games!

[superppl]

I think we're missing something important here. These efforts are fine and dandy, but... Another shooter? Can't we have something else?
I mean I know at this point it's just an engine, but it seems to me that that's where it's going.

[russofris]

I wasn't really complaining, merely stating the fact, and asking why an OSS game would use features you can't use on OSS drivers.
To me that defeats the point of being an OSS game.

You are correct, you cannot play a game that does not exist with a driver that does not exist. However, teams tend to work on these things in parallel, so that when the driver is finished, games will be available to take advantage of them. It's not like there is one single team writing all OSS software in a serial fashion.

F

[Raedwulf]

I think we're missing something important here. These efforts are fine and dandy, but... Another shooter? Can't we have something else?
I mean I know at this point it's just an engine, but it seems to me that that's where it's going.

The game was only the partial focus of the project. If you want RTS, try Glest. If you want a generic engine, there's Irrlicht,Crystalspace,Ogre,Horde3D.
XreaL is based off ioQuake3, which is very FPS oriented, so its hard to make it do something else.

[Dragonlord]

What I talked about are technical things not artwork things. A HDR renderer works also with mediocre normal maps and specularity maps. It's about lighting and material system. Neither does look here too convincing to be honest.

[Eragon]

The game was only the partial focus of the project. If you want RTS, try Glest. If you want a generic engine, there's Irrlicht,Crystalspace,Ogre,Horde3D.
XreaL is based off ioQuake3, which is very FPS oriented, so its hard to make it do something else.

Ufo: Alien Invasion is a turn based strategy game and is based off ioQuake3 as well, right?

[hax0r]

eh still looks like crap, I prefer my sega dreamcast.

[elanthis]

I think we're missing something important here. These efforts are fine and dandy, but... Another shooter? Can't we have something else?
I mean I know at this point it's just an engine, but it seems to me that that's where it's going.

That's been my thoughts for the last few Open Source games announced.

I don't mind the game involving shooting. I'm just bored to freakin' tears of every Open Source quake-based game being a Quake 3 Arena "shoot the other players and/or bots and see who gets the most kills" (or one of the popular variations, e.g. CTF).

I love games. I'm a game design major. Games aren't (just) about getting pretty graphics.

Games are the ultimate form of art. They include and use traditional 2D art (in a great many forms, depending on the specific scenes, settings, and artistic design the game offers), cinemagraphic art, literary art, music, voice acting, and not least the arts of immersion and puzzle craft. Every other artform created by mankind is encompassed by games in a way that no other medium can possibly hope to achieve.

And all Open Source gives us is a ton of Quake 3 upgrades.

Where is the original story telling? The original art (not just Yet Another Fucking Space Marine In a Suite, but something new, something beautiful, something that truly impacts the viewer)? Original music scores besides just Techno Remix #4558632? Puzzles? Innovative gameplay that makes the player think in new ways?

All the truly creative gameplay that hobbyist produce end up being tiny puzzle games. These ARE truly great things, and I appreciate them. But Open Source has never produced anything on a magnificient scale. Open Source has never (and I'm starting to believe never will) produce an entirely original work on the scale of Fallout 3, even using old technology (the original Resident Evil looks like crap compared to XreaL, yet it is 1000x more important and meaningful to the art of gamecraft than XreaL could conceivably ever be).

OK, so maybe Open Source hackers are just the kind of people who really like shooters. Why is every freaking shooter they make just a ripoff of the Quake gameplay, though? Why is it that with all the creative juices the Open Source community has, all of the actual gameplay and gamecraft innovation in the shooter genre still comes from the commercial world? Why is Open Source still upgrading Quake instead of making the next of Gears of War, Call of Duty, Left 4 Dead, Dead Space, Resistance, or Metroid Prime?

It's not just shooters, although those are the most popular Open Source games by far. Tux Cart is just a Mario Kart clone. Glest is not meaningfully different than Myth 2, Warcraft 3, or Age of Empires. I can name of dozens of clones for other games, or engine reimplementations, or so on... there are certainly Open Source projects that offer new artwork, but they aren't new expressions; they're just remakes of the same kind of art and the same kind of gameplay that the commercial guys have been putting out for years and years.

Look at a game like Zelda: Wind Waker. The technology needed for an engine like that is nothing compared to what XreaL can do. What made that game impressive was not the technical power of its engine, but the creative vision that produced it. It's not just "cartoony art," it's art with a very specific and unique stylization that sets it apart from any other game made before it. It has a narrative. It has original elements to its gameplay (though not as original as its predecessor, Ocarina of Time, of coures). It has unique puzzles, powerful music, immersing game play. It didn't feel like a clone in the least, because it wasn't one. Open Source hasn't produced a game on that level in 30 years.

Is it because Open Source is only appealing on a large scale to the purely technical types -- the people who are more interested in code and optimizations and graphics pipelines than they are in artistry and immersion -- leaving these projects lacking a creative direction or a focused art production team? Does Open Source just not attract the kind of people who can create such a work?

I don't believe that's true. I know highly creative people who are devoted to Open Source. I know a great many Open Source projects that have fantastic artists working as part of their team. I've seen Open Source novels. I've seen Open Source music.

What, then, is lacking? What is the community missing to make it possible for Open Source to deliver the next Game of the Year? I honestly want to know, because both games and Open Source are important to me. I just cannot figure out what the missing puzzle piece is here.

[superppl]

The bottom line is not graphics. It's content. And content takes hard work, time, specific skill, direction and lots of people (and with that comes management). I think it would be an accurate statement that we lack all of them.
You see a lot of puzzles because they only need one person with those specific skills to get done.
But something at the Fallout scale will need probably a dozen people to get it done. Those people would have to work hard, long, and fast to get it out in a reasonable time frame.

Now I think are a few problems here, I kinda stated one: managment. We need a visionary, someone who says this is what I want, and this is how we are going to get it done.
I think a lot of projects lack this, but even there is another problem: motivation. One would think that there is a difference if one is getting paid for hard work with cash or comments.

[deanjo]

That's been my thoughts for the last few Open Source games announced.

I don't mind the game involving shooting. I'm just bored to freakin' tears of every Open Source quake-based game being a Quake 3 Arena "shoot the other players and/or bots and see who gets the most kills" (or one of the popular variations, e.g. CTF).

I love games. I'm a game design major. Games aren't (just) about getting pretty graphics.

Games are the ultimate form of art. They include and use traditional 2D art (in a great many forms, depending on the specific scenes, settings, and artistic design the game offers), cinemagraphic art, literary art, music, voice acting, and not least the arts of immersion and puzzle craft. Every other artform created by mankind is encompassed by games in a way that no other medium can possibly hope to achieve.

And all Open Source gives us is a ton of Quake 3 upgrades.

Where is the original story telling? The original art (not just Yet Another Fucking Space Marine In a Suite, but something new, something beautiful, something that truly impacts the viewer)? Original music scores besides just Techno Remix #4558632? Puzzles? Innovative gameplay that makes the player think in new ways?

All the truly creative gameplay that hobbyist produce end up being tiny puzzle games. These ARE truly great things, and I appreciate them. But Open Source has never produced anything on a magnificient scale. Open Source has never (and I'm starting to believe never will) produce an entirely original work on the scale of Fallout 3, even using old technology (the original Resident Evil looks like crap compared to XreaL, yet it is 1000x more important and meaningful to the art of gamecraft than XreaL could conceivably ever be).

OK, so maybe Open Source hackers are just the kind of people who really like shooters. Why is every freaking shooter they make just a ripoff of the Quake gameplay, though? Why is it that with all the creative juices the Open Source community has, all of the actual gameplay and gamecraft innovation in the shooter genre still comes from the commercial world? Why is Open Source still upgrading Quake instead of making the next of Gears of War, Call of Duty, Left 4 Dead, Dead Space, Resistance, or Metroid Prime?

It's not just shooters, although those are the most popular Open Source games by far. Tux Cart is just a Mario Kart clone. Glest is not meaningfully different than Myth 2, Warcraft 3, or Age of Empires. I can name of dozens of clones for other games, or engine reimplementations, or so on... there are certainly Open Source projects that offer new artwork, but they aren't new expressions; they're just remakes of the same kind of art and the same kind of gameplay that the commercial guys have been putting out for years and years.

Look at a game like Zelda: Wind Waker. The technology needed for an engine like that is nothing compared to what XreaL can do. What made that game impressive was not the technical power of its engine, but the creative vision that produced it. It's not just "cartoony art," it's art with a very specific and unique stylization that sets it apart from any other game made before it. It has a narrative. It has original elements to its gameplay (though not as original as its predecessor, Ocarina of Time, of coures). It has unique puzzles, powerful music, immersing game play. It didn't feel like a clone in the least, because it wasn't one. Open Source hasn't produced a game on that level in 30 years.

Is it because Open Source is only appealing on a large scale to the purely technical types -- the people who are more interested in code and optimizations and graphics pipelines than they are in artistry and immersion -- leaving these projects lacking a creative direction or a focused art production team? Does Open Source just not attract the kind of people who can create such a work?

I don't believe that's true. I know highly creative people who are devoted to Open Source. I know a great many Open Source projects that have fantastic artists working as part of their team. I've seen Open Source novels. I've seen Open Source music.

What, then, is lacking? What is the community missing to make it possible for Open Source to deliver the next Game of the Year? I honestly want to know, because both games and Open Source are important to me. I just cannot figure out what the missing puzzle piece is here.

I think one thing that is missing in a lot of these projects is a good level editor and the likes. I say that from the amazing mods that are found out there in the commercial domain made by the contributing community. The talent is most definitely out there. Beyong the Red Line is a good example of how refined something can look and feel when you have a talented set of core contributors. Unfortunately is also exemplifies one of the biggest weaknesses of a development model. People disagree and then split ways and attempt to redo the same basic project with slight differences and you wind up getting 2 projects that progress so far and then are forgotten about.