Need Suggestion please

Start by deciding what kind of game you want to make. Have you tried the flag rush tutorials yet? Trying these would definitely give you a good idea on how to make a game using jme.

Most important thing you need to do is come up with a really good concept.



It will have to be VERY simple if you're going to do it in 2 months starting from no base in 3D.

If it isn't then you will likely end up with something that looks rough and incomplete.

But you should also make it some concept that works well in 3D - or the question will be asked, why not just do it in 2D.



Also I'd suggest some idea that won't need too many 3D models, and those you do use should be very simple (so you'll need a game style that fits), or things you can easily get hold of. You don't have a lot of time and sounds like you have a lot of ground to cover on the programming side, which I guess is where you'll get marked, so minimize the time you have to spend on other areas.



Once you've got your idea then dive in to the tutorials and jmetest, get a feel for the engine and maybe build a prototype of your environment based on some of that code. Should be enough to demonstrate progress.



I'd also consider a contingency of maybe doing a none-game 3D app if time gets too tight. Something like a visualisation/walkthrough, or perhaps previewing some product that you can put together on a website. It would keep some of the wow factor of Java 3D while letting you use more of your enterprise skills, and requiring less learning of the complex 3D stuff.

I'd suggest a game of chess, or dama perhaps… It would only need number of non-moving models… the familiarization with jMonkey's coordinate system (Vector3f positioning)… and some basic knowledge on 3d modeling… For a begginner, I'd suggest Google SketchUp… it's free, easy to learn and has ColladaExport which jMonkey supports… then a discrete number of lines of code to import these models you've made with sketch-up… position them in jMonkey's scenegraph… and now your only problem is how you make an AI to compete with in-game… or you might just want it to be a network game so you would not have any problems about implementing AIs… what do you think?

Chess or any other board game is actually a goodidea, since you don#t have to care about rotations(I guess chess figures will always be standing, you need no physic, you need no animations, you have fixed amount of models and don't need to load some later ingame (and also won't have memory leak problems then))



Then to impress your professor add when that works the HDR game state, and whater below the game board (refelctions yay) + shadows to have some eyecatchers.  (Most of them are pretty ready to use without more work needed than a few minutes/hours to add them)

thanx guys… it's really help me to give an alternative ideas.



for my friends momoko_fan, alric, mhelz0001, Empire Phoenix sory im forgot to said my game concept.



My proffesor want me to build a serious game, it must have 3 good component. good 3D graphics, good AI and good Interface. The Game Genre is Real Time Strategy if it's online then i will get A plus.



I realize, consider what i didn't know and how much time i have i can't do that in 2 month  :frowning:

so B minus is my target now :D, i will use your advises as my new plan and then i'll make my proffesor sign on it, even i must cry on him  :wink:



thanx guys, if it's complete i will make showcase in here…

Making an RTS in two months isn't difficult. jME has a solid terrain system and a solid texture splatting system which should make for a good start. There's the strategic mouse handler on the forum. Additionally RTSes mostly use static models so you don't have to worry much about art pipeline. Establishing pathfinding should be easy with a grid-based node system and A* search… I dunno what other issues you may run into, this pretty much takes care of all the difficult stuff.

  :-o wow thanx momoko that is absolutely open my mind…



thanx… :smiley: