The same could be said for a GPL 3D program, which is in the same boat as the BGE in this case.
FumeFX, Realflow, Thinking Particles, can’t have those if you’re a Blender user. Fast and fully integrated workflows with the industry standard render engines, can’t have that either unless you go the more unsustainable route of having multiple custom builds that have to be updated to keep up with the Blender feature-set. Blender on the go with Apple and Win8 mobile devices, nope, can’t have that either unless you expect everyone in the community to only use Android. Easy and reliable export tools that see rapid development with the ability to use the industry standards (so you can at least use Blender assets in non-GPL software), it’s either with great difficulty and frustration on the developer’s part or, yes, you can’t even have that.
I agree that this is a realistic plan, but it’s also a really boring one. I don’t want to make interactions; I want to make games. Who the hell wants to play an “interaction”?
And therein lies your primary problems (outside the ever present and already discussed licensing issue). The engine is very quirky, the key features the BGE is lacking are present in almost all the (serious) alternatives, and given Campbell’s comment - not even the BF takes it seriously enough to pay for it’s continued development. Think of how someone not already enamoured by the idea of using the “Blender’s own game engine” sees that. It’s not a solid basis for starting a project you actually want to see through to the end.
Quite honestly, I (& others) see the BGE as a toy. It’s good enough for artists & designers to prototype in for sure… but as soon as you push the limits of what it can do, you find yourself in a relatively small box. I can understand those limitations when you’re working on, say, a game meant to carry across mobile devices as well, but the BGE doesn’t do that (nor do I really expect it to).
The way I see the BGE is the way I see the Sauerbraten & Tesseract projects. Interesting and fun for putting together something quick and nasty to demonstrate a feature, map, idea… but ultimately their limitations outweigh the prototyping speed. BGE also has the licensing issue on top.
Actually people have said that about new features when they were deployed in a less than useful state. Thing is the developers fixed the problems, improved the solution, and got buy in from the Blender Foundation to check in & maintain their code. My understanding is that the BGE lacks someone doing that and as long as that persists, it is going to stagnate.
Off the top of my head - people using Blender for architectural visualisation, people putting together throw-away game prototypes and animation/movie previz, people wanting to quickly navigate a complex scene, machinima-style animators, etc.
Looking at the alternative though, considering some of Ton’s recent statements about the state of .fbx support, it’s possible that without the BGE, the best answer a Blender user will be able to give someone who wants to make something with Unity is “Buy a Maya LT license” (in other words, Blender being completely and utterly dead when it comes to game creation, even if just for asset creation). Yes, there’s someone working on improving the support, but the same mail also said that the work could only be done in earnest if he had access to a Maya/Motionbuilder license.
Also, I’ve been resuming work on one of my game projects and I don’t feel trapped in a tiny box at all (and yes, it’s a bit more complex than “move the box to the sphere”).
Um, I think you’re reading more into what Ton is saying in that thread than what is there. Unlike the line in the Blender Meeting Notes, which was ambiguous but has been clarified, that email (and that email thread) is basically stating that the official FBX SDK is a no-go, not FBX as a supported file format.
FWIW, I’m happy that you are getting some use out of the BGE and I never claimed it to be “move box to the sphere” simplistic. However, any objective analysis of what it can do compared to other offerings (both commercial and open-source) does leave BGE looking sub-par. Let us know when you are releasing you game and we’ll see what limits you’ve stretched. Happy to be proven wrong on the BGE, it’s just no-one has released anything that has done so yet.
The only reason I haven’t used the BGE much recently really is because I couldn’t really do that and render with Cycles at the same time. Otherwise I might have a bit more stuff to show (I’ve waited around 8 years to render with GI when Cycles debuted so I had a ton of ideas lined up).
I do have a Miniature Golf game in the Finished Games forum, 3 modes, custom code for bouncing off of walls, custom GLSL water shading for some holes, surfaces with different amounts of friction, separate highscores for each course, ect… It’s a full game, not a prototype. I will also note the fact that the reason why the BGE doesn’t have some built-in features like multiplayer and in-game file saving is because of its ability to access the general Python API as seen at Python.org (people will just use modules like sockets and OS for that type of thing).
With all due respect, that’s not a game pushing any limits. It might be somewhat prettier than some of those done in the 24/48 hour game dev competitions, but it really isn’t something that couldn’t have the fundamentals done in that time period or even have been accomplished ten years ago with a number of different engines.
I’m not dissing your game, I’m sure it is a fun mini golf sim. I’m simply stating that it doesn’t demonstrate that the potential of BGE is worth investing in over the alternatives. You want people to take BGE seriously, show it’s capable of serious projects. Mini golf sims utilising a physics engine that is integrated in nigh on every other engine isn’t going to to that.
Well, you attempted to make a point saying that the BGE user finds himself in a ‘very small’ box. This implies that the engine simply can’t do much in terms of making any sort of game at all. We know it currently can’t do something like a modern AAA game with hundreds of animated characters on screen at once, but if you’re looking at the indie level it can already do quite a bit. Your choice of words make it seem like my projects for instance just couldn’t possibly be doable in the BGE. One thing to note, I actually did make my own custom deflection code for bouncing off of walls (same with the friction as Bullet has trouble applying this along with restitution to rolling objects with sphere bounds).
However, recent improvements have made it not nearly as restrictive with animation as it used to, what with multithreading in animation as well as animation culling, so one of the biggest things that keep people from making serious projects have been addressed to an extent.
A final note, when judging an engine, I really would look at projects that aren’t made as part of game jams or 2 day competitions. That’s one of the less flattering things I find about the indie scene, all of these ‘contests’ that celebrate quantity over quality (it’d be like if I judged Unity by the thousands of small throw-away projects I see all over the Google Play Store and sites like Kongregate).
No, that would be your knee-jerk reaction to anything less than positive being said about the BGE. I said (and I quote) “as soon as you push the limits of what it can do, you find yourself in a relatively small box”. A mini-golf game doesn’t push the limits of any widely used engine. For the most part, mini-golf is a basic physics simulation with graphics ranging from mediocre to uniquely intriguing… but at it’s core, it is a simple concept that can be handled by most engines released over the past ten years as open-source physics engines are pretty old hat.
That’s great and, again, I’m not trying to pick on your game at all. Thing is, that isn’t really the kind of thing that shows off the potential of the BGE that you believe/claim is worth the effort you’re suggesting be dedicated to it. I’m sorry if this sounds crass, condescending, or even rude but deflection off a plane and tweaking the friction mechanics of a physics sim are pretty run-of-the-mill coding achievements in modern games, even indie ones.
You must be misunderstanding me here. I wasn’t using the product of the competitions as the examples of what the BGE is capable of. I was stating that a basic mini-golf game could be produced as the product of one of these competitions. I agree with you - if you want to see what a game engine can do, look beyond what is capable in a 24/48hr competition. As the core mechanics & setup of a mini-golf is at the level of complexity I would expect could be accomplished from them - it’s not what I would consider a fine example of an engine’s potential… or, if it is, that’s an indictment not a plus.
I’m not trying to argue that my game is the penultimate example of BGE work, it’s just to show that the BGE can indeed be be used to make fully playable games (and fully complete ones that also have a decent length in terms of total gameplay).
People keep saying that the BGE is for nothing more than simple prototypes or throw-away learning projects, if this was the case then some of the projects I’ve mentioned before shouldn’t even exist (that or they just don’t want to acknowledge that such projects do exist, maybe I’m just imagining all of the activity in the GE forums and in reality are just blank pages that haven’t seen a single post in years).
It’s really that simple to make a chart/spreadsheet where UE4, Unity, CE, Source, and a few open source engines can be compared to Blender, feature by feature.
Well, where are those glorious games that people release for free, or commercially, that have visible and persistent presence on the gaming landscape?
The only game that was made with BGE was Dead Cyborg (a game that got a couple of episodes released and had some following). Which despite having community and being Greenlit, is not pushing anything, and for whatever dumb reason is not being published on Steam (why the hell did endi go for Greenlight?)
So can QuickBasic. I’m not going to say it’s worth investing huge amount of development effort into it though.
One can draw incredible works of art in Microsoft Paint. You’ll still find most artists prefer software that allow them to make their art quicker, easier, and without having to workaround large limitations in the chosen tool. Just because a tool can be used for a given task doesn’t mean it’s the best or even preferred tool for the task.
No-one is denying that there are people who will use the BGE. There are people that use Microsoft Paint too. The question is whether or not it is worth the development focus you’re asking from the Blender Foundation.
I personally don’t see anything coming out of all that activity that cannot be just as easily done (if not more so) in an alternate engine (be it commercial or open source).
i don’t think blender needs to compete with unity, ue4,…
but what i would like to see is an official way to export to webgl. with logic-bricks/nodes/scripts and everything. that would be awesome!
but the interaction mode plan (with tighter integration into blender’s internals) sounds to me like it wouldn’t be possible easily? or maybe the whole blender could just be compiled with emscripten? the result probably would become too huge though.
There is a lot of focus on areas of Blender being better/worse then other applications.
But you can look at it a bit different…
Is the existing user-base happy with the tools.
Are bugs fixed (are users supported).
Are improvements being made each release to address user needs.
Even if the software is limited compared to alternatives (worse in some areas), if its being improved to support a user-base who are satisfied to work with within these limits, then its still valuable to have.
The problem I see with the BGE is no developers really want to support users and push the BGE forward.
Probably no developers is too harsh, but not enough development at least.
Everybody’s making a game engine these days (me included :p). Maybe one day Adobe will create a game engine?
Game engines enivornments are becoming more and more diversified with different requirements that they favor specialization. Thus you see lots of people rolling their own engines; not to mention, assembly is not a strict requirement anymore!
You got PC, mobile, console, WebGL, etc.; coming soon cars and TVs…Then you have various OS’s, APIs, favored languages/development styles. It’s tricky to build an engine that performs well on all platforms without tweaking from the developer. Of course, this is one reason you’d pay for a commercial engine as the company is supposed to provide solutions to that problem.
The irony of the BGE is that it is well-suited for the indie crowd, but the indie crowd focuses a lot on mobile, which the BGE can only partially support. The little guys don’t have the resources/developers to deal with all the shortcomings of an open-source engine, and the bigger guys have the resources to simply use a commercial engine with cutting-edge features and support. As we have witnessed, this leaves the BGE for hobbyists and enthusiasts.
I’m not sure how much of a factor the GPL is for the BGE, seeing that other open-source engines with a permissive license aren’t much better off than the BGE. Purely anecdotal, but I haven’t see any open-source game engine seriously compared to a commercial solution.
IMHO open-source engines/gaming in general are just not there yet. There is no Linux of game engines so far, at least any with significant commercial backing. Actually, the BGE might be the closest thing to it with the BF.
