I’ve found here an attempt to make a GUI for BGE:
http://www.gameblender.org/viewtopic.php?f=14&t=540&st=0&sk=t&sd=a
This quote reminded me of this thread: http://blenderartists.org/forum/showthread.php?t=158289
I’m not much of a programmer but implementing this would most likely be a project in it’s self. However at lest we do have the source code to make next gen water.
What I would really like to see is someone willing to do some “really good” documentation on the blender game engine. I think this is about the most impotent thing there is to a game engine… maybe after Twilight22 is done with their game.
Letter by letter messages.:yes:
I’m learning C++ at the moment. Once I’ve finished all the consoly stuff I’m going to learn all the game stuff. I just got the idea of putting PhysX into the GE, sure it’d be Nvidia cards exclusive but real time liquid and smoke FTW! Plus it works with OpenGL already.
I know people like having blender mulit platform but also thinking in the future (once I’ve learn’t a lot more of course) maybe making a Windows exclusive BGE with DirectX11 (by the time I’m experienced enough, I’m sure DX11 will the norm)
As for now it’s time to go back to Chapter 6.7 of my C++ tutorial book: Overloading functions
Actually… Shotgun implementing PhysX and DirectX, I’ll probably start late this year/early next year if I remember. No one else it to try until I’ve given up.
As for now feast your eyes on this http://www.youtube.com/watch?v=JzQjchtWjvw
I think we wil not go anywhere like this. There are many engines (devmaster.net) with many useful features, but somehow only few engines are really suitable for making gamesn, not just pretty demos will skies or advace water or complicated physics demos. I think the only way not to waste human resources, trying to imlement diverse number of features, is to concentrate on a particular game project.
So there was Yo Frankie - casual kinda game. Maybe it is better to start new community project like FPS and develop the engine to its needs.
That’s not a bad idea, in fact its a fantastic idea. Probably won’t happen 'til after Durian but given the advancements made by the frankie team I’d say another game should definitely be on the cards. Personally I’d like to see an open world FPS because open worlds are very big across the different genres now days and fps enemy AI and shooting would be very transferable. (in terms of setting I’d go fantasy or mythical as oposed to yawn space marine yawn)
you beat me to it 
Also documentation would be a great idea. Perhaps a dedicated wiki to the game engine would be handy?
Anyway, some things I’d donate for inclusion are sound system improvements, more advanced shadowing (that deffered lighting thing maybe?), antialising and some shaders. I was a bit suprised that bloom, HDR, AOO and a few others weren’t included in the 2D filter logic brick. I’d love to see more of them included as standard.
Something I’d love to see is a real time materials bank. A .blend that includes a load of materials like water with reflections & fresnel, plastics, metals, stones, woods and so on that you just link in (it gives instructions how to set up ones like the water that require code snipits as well). It would be useful for people to have a whole bunch of optimised, high quality materials on hand to speed up development of games.
Hello,
Here come a lot new feature request that I collected form this forum after search im still waiting for more suggestion and the we start create some group for featurres.
-glsl shader without use python (for faster?)
-glsl shader use with blender gui (custom glsl script material node in node editor?)
-create uv unwarp on all active or selected objects and show/editing in on time the uv editor
-normalmap baking with AA
-baking lightmap for instanced mesh (dupligroup or dulicated linked mesh)
-sun light shadow + alpha + soft
implementing glsl haders:
-reflect refract, glow, parallax, cubemap,cubemap ligthing,
-IBL(image based lighting)
-amb at Map To panel works with glsl
-Refl at Map To panel with Cube Env map
-dynamic 2d imposters (this is a lod part?):
http://www.gamasutra.com/features/20060105/davis_01.shtml
-render areas (set a box where you can setup a special rendering process)
-text object box multi line (left, center, right, align)
-work tangent normalmap with mirrored uv (or with mirror modifier?)
-unique uv map for instanced object
-terminal feedback inside blender window
-timer sensor: count back, reset, start, pause
-collision detection without physics
-up donw button for all visual logic brick blocks without need closing
-getIpoCurrentFrame value with python
After reading that last post I must agree that a simpler way of putting text in game would be great
I’d like to see a particle system implemented, perhaps with options for 2D or 3D particles. It seems like a basic game development tool that’s missing. I’m not too familiar with Blenders particle effects, but I’d be happy with a more advanced version of the AddObject actuator, with fields for multiple particle objects, rotation, scaling, alpha, etc. It would be nice to not have to rely on IPO animation for everything.
Video texture to be assigned by having a video file as normal texture (not only with python) and an actuator to control it.
The only problem with PhysX is that it’s so new that it only works well on the latest Nvidia cards. Not to mention it doesn’t work at all on ATI cards. If you where able to implement PhysX people might be unwilling to use it because they would need to set up Bullet for everyone else. If you could make it so most PhysX functions will automatically use the same bullet functions when PhysX is disabled you might be on to something.
As for DX 11 implementation I’m against it. I think time would be better spent making Open GL better.
Dude, you don’t go on open source software websites and talk about tying something to Microsoft. I’m a Linux user and I use blender quite a lot. It’s the only games dev tool on Linux I like. Keeping blender multiplatform isn’t some silly little triviality that Windows users afford us Linux users, it’s integral to the core spirit and goals of blender.
Feel free to code your own DX engine that reads blend files or plugs into blender but myself and many others will fiercely oppose any push to make the game engine tied to a single platform.
Remember Microsoft are the same people that tied DX10 to windows Vista to force the gamers to upgrade to their buggy, slow operating system. Are these the people you want controling PC gaming as a whole?
lol, that’s a good joke
I say we only make blender releases only in Japanese too. Since we’re already cutting out a good sized chunk of blender users, we might as well kick it up a notch!
Feel free to code your own DX engine that reads blend files or plugs into blender but myself and many others will fiercely oppose any push to make the game engine tied to a single platform.
He could do that, just so long as he understands it’ll never be an official implementation.
Remember Microsoft are the same people that tied DX10 to windows Vista to force the gamers to upgrade to their buggy, slow operating system. Are these the people you want controling PC gaming as a whole?
We seriously need to make some games buddy. I was working on SuperTux 3D but it seems blendermaster doesn’t really have either the will or the internet connection to do it. I’m starting to flake out, and he’s just plain disorganized, and I think that’s one of the real reasons no one is jumping on. Thinking about Deadman’s Frag as building/city design. But I’m mostly into texturing right now. No other open teams seem to be as organized. Someone out there has to have a decent script and concept of a stupid Oblivion knock off?
-
2D Filter optmization - FBO, multipass (not necessarily Deferred Rendering)
-
full replacement of skinned meshes to gsls skinned meshes (for armatures only, not for shape keys)
-
CEGUI as GUI system (would that fit in people needs?)
-
-
- note that some of the items proposed in this thread have people already started looking at them. However time is usually short (at least mine), and not so fun to code features are usually left behind. For I think being able to dedicate some paid time on that can help for some of them.
-
one thing im wondering is what exactly twilight22 or whatever is working on with the game engine. it would be very silly to pay for something that is already being worked on somewhere else.
or are they not actually working on anything?
anyways, my votes are optimizations, and networking support
@planaria, dont worry, developers will say if some feature is planned already for other projects. (eg LOD and Sound system I mentioned this)
two others feature:
-real load progress bar (show loading data stuff in right percent)-or this is possible yet?
-Add Property button - need List type?
That would be possible with dynamically loaded objects.
Actually come to think of it, dynamically loading objects would also mean you don’t have to do the whole startup .exe to linked scenes thing that we usually have to do to protect IP and ownership
ideasman : im surprised twilight22 is not working on networking since that is a glaring feature hole. or maybe they are … anyways im mostly just happy that they are still actually working on it 