Time to for the FOSS community get to work on a capable, easy-to-use 3D engine.
Valve has just announced that they're going to release the Source engine SDK for free. Is this cool news? Sure, I'll probably check it out myself. But, the big downside here is that as high quality, closed-source engines are made free as in beer, the FOSS community loses out because we don't have anything that can compete with Unity3D and Source in terms of both features and ease of use (one or the other, perhaps, but not both).
We're losing mindshare here, folks. If you're one of those people (you know who you are) who thinks it's okay to force everyone to constantly switch applications and muck around in text editors in order to do anything useful with a game engine, then how do you explain the popularity of engines that are easier to use?
It's one thing if programming time is an issue -- we all have limited time. On the other hand, it's just plain lazy to justify a lack of features and ease of use by claiming that these sorts of things are completely unnecessary.
If your software is unintuitive and hard to use, and requires people to read forums and poke around through (often incomplete) documentation, it's not finished yet. Time to get to work, folks. :)
Peace out,
Bart
- bart's blog
- Log in or register to post comments
Comments
Is there actually a good game that has been created with the Unity Engine?
Also I have to quote this:
It seems they made a common mistake - creating the tool for building the product before they get the product itself :).
It doesn't works this way - the only possible way to build a tool - first You get the WORKING product later You extract the tools from it. And never in another direction.
Well, there are plenty of good, working FOSS 3D engines. BGE, Irrlicht, XReal, etc. Rather than making a new one, it would be nice to have a solid, easy-to-use, integrated editor for those. BGE is probably the closest, but it need to improve by leaps and bounds to become intuitive.
I agree about the game thing, though. In my own experience with building my JRPG engine, I realized a lot of things when I started building my tech demo. I imagine that once I get a chance to build a real game with it, I'll learn a lot more. Dogfooding is hella important.
I personally think it's impossible for a hobby project/group to create an engine that is easy to use and covers most/all game types. In my opinion, it would be way more useful if there was a really good one per game type. At least one very flexible game with easily editable and scriptable maps and a good editor. In some areas we already start having such games, we're not quite there yet, but it is already much better than 5 years ago.
Also, ogre looks quite good right now. I saw someone working on an editor for ogre scenes which looked good too.
I have argued this point for a while now. There isnt anything in the FOSS world (yet) to compete with 3d game engine platforms like Source/Unity/Unreal. Some people will argue that IOQuake3 + Radiant + Blender can compete, but TBH radiant is a very complex tool to learn, and you still have to do a lot of engine changes on ioq3 to get out of the FPS range (it's an FPS engine after all).
And before someone suggests Ogre3d or Irrlitch, remember that those are general purpose engines, not game-specific. A lot of work has to be done to have something (remotely)near to Source.
To my knowledge the only FOSS project actually seeking this: nice 3d rendering, hi-res assets, easy to build games, cross-platform (Linux, Windows, MacOSX) is OctaForge http://octaforge.org/
OctaForge is based on the Cube2 engine which has been time-tested with a few games/mods and it consists of a very simple core (almost all Opengl). They want to allow users to build games using Lua (for game logic) and the existing Cube2 realtime level editor. The editor allows multiple artists to work on the same level in realtime. This project is decendant of the Syntensity project which now has ended.
OctaForge is a rather new project but under heavy development. IMO this is the project to watch.
Games using Cube2 engine:
Sauerbraten - http://sauerbraten.org/
Red Eclipse - http://redeclipse.net
I didn't know about OctaForge, thanks for the info, it looks promising
thanks for free advertising ghostshell :)
Yep .. OctaForge is under heavy development and goes forward pretty fast. It's almost reaching beta release currently. Freel free to try it out and go bitching about it on our IRC :P I'll be glad for all that :)
Does anyone else think the industry wants to kill FOSS engines? o_O I know a project that switched from OGRE to the Unreal 3 engine when they released it for free, and I guess there were probably more. I heard even the new CryEngine will be free of charge for non-commercial projects, or is it already, or was it one of the old engines? I'm well informed. :) At least id Software's old engines are really free.
I honestly think they begin to consider the FOSS stuff a small but real concurrent. And that's the way they deal with. It looks very similar to what microsoft did with their "open source" licenses and then with the Express versions of the VS products. I'm not really sure what to think of it tho.
My vote is on BGE, it's what I plan to use sometime in the future. Last year blender had 3 GSoC projects regarding BGE:
-Recast&Detour integration (pathfinding and navigation)
-Python API improvement for BGE
-Improved shader handling via scripts for BGE
This year there are also 3 GSoC projects for BGE in progress:
-Animation improvements
-General bug-fixing and improvements
-Node-based logic editor (THE thing if it gets done)
In addition, there are efforts being done on BGE particle engine http://enja.org/
and a library to handle GUIs in BGE https://mogurijin.wordpress.com/category/bgui/
One smaller, but recent addition is also anisotropic texture filtering included in 2.58
Not to mention that with the game engine you get an integrated editor to create the content. There are of course still things missing, like better post-process effects (with more control over them), anti-aliasing and other things, but I have my hopes high for BGE.
TBH i think its a non-issue.
The idea that Source engine/Unity/UE3 engines are some types of dev panaceas is just not true. All engines have their quirks, their positives and negatives. I have worked professionally with UE3, its great for some things, but would be total overkill for others. I have only shipped 1 product with Unity, but my experience was that its pretty far from the technology when you use it... and that is not always what you want, depending on your project. Ogre is very close to the tech, but lacks certain advanced tools.... and that may be an appropriate trade-off depending on the project.
If you are an indie dev, take these words seriously: enough overkill in an indie project will destroy it.
having said that i want to reply to bart's comment:
Well, there are plenty of good, working FOSS 3D engines. BGE, Irrlicht, XReal, etc. Rather than making a new one, it would be nice to have a solid, easy-to-use, integrated editor for those.
The idea that there will ever be a generic engine/editor for games is an idealistic goal that is ultimately is a fantasy. Tools are one of the hardest areas of game dev. Unity advertises itself to be the perfect catch-all tool - but it isn't, and nothing can ever be. I see Unity as the 'Flash', for 3D game development: perfect for some things, terrible for others. Even with Unity you will always need some level of specialized tools for a given project, and the more innovative the project the more the tools must follow. So I'll end with a small advice for indie devs who choose engine's without editors that suit their needs: as much as possible avoid building editors. Instead, use your content creation software as editors: use max/maya with maxscript and MEL to export xml, if your gameplay is 2D you can even just use photoshop... paint bmps with your collisions and use colors to position your objects. Don't be sold on shiny editors, simple is better and thats will always be true if you want to get anything done as an indie.
read my comment. i think you missed it.
read my comment. i think you missed it.
Hello Anonymous. :)
I feel like maybe you meant to be logged in when you posted that.
im guessing that anonymous was probably pompei, my opinion is in the same vein as his.
Unfortunately, while BGE is promising, it comes up short in terms of ease of use.
The node editor, for instance, is incredibly wonky. What's particuarly frustrating is that other parts of Blender make very good use of its node editing functionality, but the arrangement of the BGE node editor makes for a horrific and disorganized mess.
Furthermore, there's no simple, code-free way of doing mouse look. For something as simple and common as mouse look, the solution shouldn't be "use this python script". It should be built in and ready to go.
BGE also suffers from being generally a bit unintuitive.
Anyway, I'm not putting it down. BGE is close, and it's come a long way. I just think it has a ways to go before it will draw attention from the likes of Unreal and Source.
Just to note that I wasn't that anon dude :)
While I'm at it, one more opinion I have is that it will come naturally the way I described it, because nowadays the games have to rely on user-made content to have a big lifetime, and I see more and more of the open source game makers realize that. All those things just take sooooo much time in the open source world... (I do understand why that is.)
The BGE node stuff get currently a major overhaul in the GSoC project... in fact the logic blocks as they are called in the BGE are a old remain of stuff that was done way before the node editor was introduced and now they are trying to integrate a similar/better fuctionality in the node editor to get rid of that mess.
They also plan to make node groups much more reusable to that you can have templates for different stuff...
but ultimatly I feel that for a large part of the games made, making everything from scratch as required by the BGE or other general purpose engines is really a waste of time. If you want to make a RTS, base it on a already working RTS engine for Christ sake... same for 3rd person or FPS games (can be the same engine for those).
Only if you are doing something really out of the ordinary it makes sense to revert to these general purpose engines.
Yet somehow in OS game dev everyone wants to reinvent the wheel :(
That said... some nice LUA etc scripting in those working engines for easier modability would definity help a lot.