A Battle For Good Open-Source Game Graphics?

[numasan]

I'd be willing to bet that most of the development on these games done by the linux community is pretty lopsided to linux port meanwhile assets are primarily done using closed sourced solutions on other OS's. Even projects such as Big Buck relied on many closed apps (especially when it comes to audio).

BBB relied mostly on open source apps. Only audio was done with closed source. There is no reason why art assets can't be done purely in OSS, except personal preference.

[L33F3R]

ahhhh yes thats key. The assets themselves created in closed source apps like 3ds max and even programming in visual studio.

[deanjo]

I think a major project COULD be done but would take a large collective of people and some strong leadership. However i might add that the linux community is good with both of these things and even has some talent to show it off.

I'd be willing to bet that most of the development on these games done by the linux community is pretty lopsided to linux port meanwhile assets are primarily done using closed sourced solutions on other OS's. Even projects such as Big Buck relied on many closed apps (especially when it comes to audio).

[L33F3R]

I agree with many that the visual representation of a game is very important today, and I would love to see a modern FLOSS game rival, say, HL2. I'm just afraid it is too much to expect, and I don't think the majority of Windows gamers will even glance at open source games on Linux. It is much more important to get studios like Blizzard and Valve on board, than to convince people to play our free, open games, made with great engines and the best art we can possibly create. Instead of competing with commercial games, we should create fun, quirky games with a unique art style, manageable by 1 or 2 artists in average. I think that is already what is happening.

Its very important to recognize the little guys who dedicate there time for making little things for us to enjoy. They deserve alot more credit then they get.

However when it comes to larger projects its important to remember that rome wasnt built in a day and it wasnt built by a man and his wife. When you beat 1 of these games look at the credits, they go on for a LOOOONG time. It took 50 people to port GTA 4 to windows and it was an awful port.

I think a major project COULD be done but would take a large collective of people and some strong leadership. However i might add that the linux community is good with both of these things and even has some talent to show it off.

[numasan]

@Dragonlord
And what type of game do you want with this ?ber engine? Will the engines features matter if it's YAFPS? Do you have an idea for a highly innovative game and the skills to make it? Can you encourage others to make it? (judging by your other replies, I'd say no...)

I think it is wonderful that so many FLOSS games exist, and that the engine/game developers have fun doing it. Creating a modern game is a huge task, so in that regard FLOSS games will always look/play like something a few years behind times, but as stated by others: Gameplay first.

I am not a developer/programmer, and have no idea how to make useful and beautiful GLSL shaders, but I am a graphic artist (currently only compositing with a grain of 3D). I want to contribute with graphics, but there are some points:
1) Creating content for a modern looking game, never mind a "next-gen" type game, is impossible to do with only a handful of artists working in their spare time (MOD artists reuse a lot of art assets from the main game). The content needs to be equal in quality, so it's better to have a game look bland overall, than having a few models of high quality among many others of poorer quality. This requires a whole lot of planning (artists sticking to a defined concept) and a lot of hard work (equal quality of all content). No use of a FLOSS AAA-level engine if this point is not in place.
2) If I'm going to dedicate a portion of my free time to a FLOSS game project, it MUST be a game that I enjoy immensely, that I want to see live up to it's potential, and that I can have FUN while creating content. Almost all FLOSS games could use better graphics, but not all captures me enough to help out. It also doesn't help that I don't game that much anymore as I used to...
3) The workflow must also be considered, getting content in the engine easily, create ingame effects without relying mostly on developers, maybe some sort of visual GLSL creation tool, etc.

I agree with many that the visual representation of a game is very important today, and I would love to see a modern FLOSS game rival, say, HL2. I'm just afraid it is too much to expect, and I don't think the majority of Windows gamers will even glance at open source games on Linux. It is much more important to get studios like Blizzard and Valve on board, than to convince people to play our free, open games, made with great engines and the best art we can possibly create. Instead of competing with commercial games, we should create fun, quirky games with a unique art style, manageable by 1 or 2 artists in average. I think that is already what is happening.

[L33F3R]

@Irritant:
platform independent, runtime modular ( so not compile time modular ), huge world support, physics simulation ( various types ), good graphics rendering ( full multilayer transparency, mirroring, refraction/reflection, dynamic lighting/shadowing ), extensible graphics ( adding new abilities without changing the engine ), holed terrains, equally capable of indoor/outdoor, unrestricted bone weighted mesh deformation, animation system capable of handling parametric/procedural animations, Blender3D support for modeling/terrain/animation, fully scalable ( from high-end down to low-end ), multi-camera rendering, multi-world rendering, dynamic textures ( runtime generated ), dynamic sound/music, streamed music, 3D audio, using a typed scripting language, persistency usable for large scale world, streamed loading of the world, model/animation/physics in separate files, no pre-compilation steps, accurate world editor ( ingame rendering ), scalable network support able of both p2p/client-server

serious list

I have always been a huge fan of irrlicht myself due to its opengl3 support and ease of use. But then again im primarily a csg level designer so i urge you to note my bias. I know physx is supported to an extent on linux but as for open source there is plenty of physics systems. And i know for a fact raknet is nearly the best and is used by sony. Several sound systems can handle 3d audio just fine and im sure python or lua support would need to be added. It honestly looks like alot of work though. Which is one of the reasons why we need to rely on open source fps games to base our creations.

From an artwork perspective adding content into open source engines like ogre and irrlicht (which keep in mind are only rendering engines by themselves) is just as easy as say quake 3.

But for now we are stuck with FPS titles based on ID engines and cube.

If said licenses on open source engines happened to support commercial development we would have more linux games. I know the art camp likes commercializing their work more then the programming camp. At the end of the day the GPL is a very social license that protects his/her work from being sold (which is a damn good reason in my opinion).

Amen

[deanjo]

Yes, BSD is second class when compared to GPL, because GPL forces people to bring upgrades to community. Thankfully GPL makes companies to deal with it harder or impossible.

So much for freedom when you have impassible barriers. It still has restrictions on what you can and cannot use with it. If GPL was more loose I have no doubt you would see great innovations like dtrace and zfs already in the kernel for example.

GPL is the best option to give code freedom.

In your opinion. It barriers true freedom of what one can contribute on what terms they want. Public domain release is required for that.

Yes and not always. In my opinion license to be truly Open Source should allow you to modify code and pass it further. Not only look at it :> I'm talking about overall not FreeSpace or Homeworld license mentioned before. Btw. why Homeworld is still in terrible state on Linux? Isn't this, because license? One of the best games in my opinion and no one's interested in polishing it?!

The freespace license does not prohibit you in any way of making your code contributions free as you see fit.

[deanjo]

@Irritant:
platform independent, runtime modular ( so not compile time modular ), huge world support, physics simulation ( various types ), good graphics rendering ( full multilayer transparency, mirroring, refraction/reflection, dynamic lighting/shadowing ), extensible graphics ( adding new abilities without changing the engine ), holed terrains, equally capable of indoor/outdoor, unrestricted bone weighted mesh deformation, animation system capable of handling parametric/procedural animations, Blender3D support for modeling/terrain/animation, fully scalable ( from high-end down to low-end ), multi-camera rendering, multi-world rendering, dynamic textures ( runtime generated ), dynamic sound/music, streamed music, 3D audio, using a typed scripting language, persistency usable for large scale world, streamed loading of the world, model/animation/physics in separate files, no pre-compilation steps, accurate world editor ( ingame rendering ), scalable network support able of both p2p/client-server

You forgot hilarious easter eggs.

[Dragonlord]

@Irritant:
platform independent, runtime modular ( so not compile time modular ), huge world support, physics simulation ( various types ), good graphics rendering ( full multilayer transparency, mirroring, refraction/reflection, dynamic lighting/shadowing ), extensible graphics ( adding new abilities without changing the engine ), holed terrains, equally capable of indoor/outdoor, unrestricted bone weighted mesh deformation, animation system capable of handling parametric/procedural animations, Blender3D support for modeling/terrain/animation, fully scalable ( from high-end down to low-end ), multi-camera rendering, multi-world rendering, dynamic textures ( runtime generated ), dynamic sound/music, streamed music, 3D audio, using a typed scripting language, persistency usable for large scale world, streamed loading of the world, model/animation/physics in separate files, no pre-compilation steps, accurate world editor ( ingame rendering ), scalable network support able of both p2p/client-server

and especially

no compile time mess engine so no need to dismantle the engine to get new features done.

I didn't write down all points there are as this is just straight out of my head but it contains some important points.

[Snickersnack]

I know, write me off as just a troll that isn't a game dev, because yeah, I couldn't code to save my life. But it doesn't make what I'm saying any less true that there is a distinct lack of creativity in linux gaming,

So, in order to be more creative, Linux game devs should clone tired old console games instead of YAFPS?

Are you offended by the proliferation of Rogue-likes?