Game ideas wanted for a J2EE 4th yr project

HaVoC

I'm going back to college in September for my last year and i plain to do a java project at enterprise level.

The basic ideas is to do a 3 tier design using ejb's and servlets. Having several presentation layers running of the same lower layers i.e. wap, web browser, and application.

Basically I want to learn as much enterprise java as i can in the process of the project (I have done standard java for 2 years up to jdbc and some rmi).
Where I got the idea was a case study in a java book i have “Advanced java 2 Platform How To Program" a deitel book ,it has a case study of a book store running in a web browser , wap browser, and an i-mode browser.

So what I want to do is do a game in the same mould. Some kind of persistent world ,where the application version would be the most powerful and other presentation Layers would be just for updates and to responding to major things. I have had a look at these notes on java games http://fivedots.coe.psu.ac.th/~ad/jg/ and I’m afraid that if I go too much into the game side I’ll never get it done,the main focus of this project would be to learn enterprise java not game programming and it’s a one man project.

So what I’m looking for is any suggestions of games to base this on or even a interesting application, that wont end up in me biting off more then I can chew, but not a book store :P

zoro

What college are you in?

yossarin

good project ! well chosen.

You're completly right about not getting caught up in the game - have as little game AI as possible in it as it doesn't seem to be your bag. In fact if you can, just build a game comms/state engine and build a simple game(s) that uses it as proof of concept.

The simplest would be turn based games like draughts or chess or backgammon or tictactoe - all you need to do is to save game progress and respond to requests from players for the current game state. Use XML for state transfer, as lecturers tend to love standards based stuff and it makes you look good.

Somthing with no hard real time requirements either, as they'd cause problems with message delivery and game state.

Hobbes

Make something like Risk, or planets(forget the name, its MMPOG). That way you will be using the database, transaction logging, etc.

gobby

yossarin wrote:

...Use XML for state transfer, as lecturers tend to love standards based stuff and it makes you look good...

XML is standard!?

zoro

the reason I asked about the college is because I'm doing something similar
It's gonna be a very very interesting project, and I'm really looking forward to it!

gobby

I will have to do a final year project also. I am interested that people seem to know what they want to do already, and in detail! I haven't got a clue whats gonna happen!

yossarin

XML is a standard, sure
The point i was making is that its generally better to use some well known process to send your server-client messages then some made up parsed strings format, unless you're using binary as effiiciency is more important, which doesn't seem to be the case here. Also, while using XML for messages brings in inefficeincy in marshalling messages, it makes it easier to build a few clients on different platforms later on.

do what Hobbes suggested - Risk, or planets - they're more fun than chess or tictactoe and make for a better project. don't commit to too much though - lecturers have treacherously (sp?) good memories if you shoot your mouth off about the amazing features you'll build in

Doodee

yossarin wrote:

good project ! well chosen.

*** or tictactoe ***

Have you not seen Wargames!

As reguards making the game.

I did games design through Java about 2 years ago, and as was suggested, we kept it very simple. We designed our own "Hi lo" card game aswell as Naughts and crosses/x's and o's/ticktacktoe. I think, without the proper know how, doing anything further will only result in hair loss.

If you wanted to, you could try a type of Virtual pet approach. where the more advanced functions were availible to the PC users, and other stuff like Feeding or playing with your creature could be done via WAP. The AI required wouldn't be very advanced (In theory), very basic functions like Eat,Walk,Happy,Sad,ry etc could be used.

This my opinion anyways.

Best of luck.

bambam

If you wanted to stay away from complex game logic, you could impress people with dynamically rendering to different device types. So your raw presentation format would be xml based, then on detection of display device - xslt into a format for that device. You could support say, PC's, phones and PDA's. Shouldn't be too much hassle as long as you define a subset of widgets / controls that you will support. Then template their implementation for each display type.

zoro

Joodee wrote:

I did games design through Java about 2 years ago, and as was suggested, we kept it very simple. We designed our own "Hi lo" card game aswell as Naughts and crosses/x's and o's/ticktacktoe. I think, without the proper know how, doing anything further will only result in hair loss.

I've already written one of those tictactoe's ... it's ridiculously easy The whole thing is a bit quirky until you realise that you can't write it as if you have an unlimited screen, it's actually being used on a tiny screen

If you wanted to, you could try a type of Virtual pet approach. .... The AI required wouldn't be very advanced (In theory)

In theory is right! If you think about it, it's going to be extremely complex for it to be of any interest to end users at all. I mean, think of all of the variables that need to be taken into account! If they're asleep, they can't eat or drink, but they get hungry. If they eat too much they'll get fat, but at what rate? And by excercising, how much fitter will they get? And how much will their thirst increase and tiredness increase? Once they're too tired to excercise, but they've eaten **** loads and drunk enough to fill a small lake, they'll need to go to the toilet. If they're too tired to go to the toilet they'll sleep, but their general health is already low as they're fat, and ......

I could go on for ever It does seem like an easy enough one, but in reality I'd imagine that it's alot harder than you'd think!

PhilH

If developing the game AI is not central to the project (and it sounds like it really isn't), then don't bother with any of it. Pick a two player game where the server is just acting as a mediator between the players. Like battleships.

PHiL

Hobbes

PhilH wrote:

If developing the game AI is not central to the project (and it sounds like it really isn't), then don't bother with any of it. Pick a two player game where the server is just acting as a mediator between the players. Like battleships.

PHiL

Doesn't really show off J2EE though.

Would be better to make a simple resources game (as I mentioned earlier) and then set it for say 10-20 players and then document how you plan to scale it.

bonkey

Hobbes wrote:

Doesn't really show off J2EE though.

Would be better to make a simple resources game (as I mentioned earlier) and then set it for say 10-20 players and then document how you plan to scale it.

A neat one would be Diplomacy with - as suggested in teh original post - a variety (well, 2 at least, more if time permits) front ends using the same common back-end.

With Diplomacy you will have everything you'll want : scheduled tasks, background processing, the possibility for Push / Pull information updates, yadda yadda yadda.

As "extras", you can also look at hooking some stuff to allow various "conversation" options between players, as well as everything else.

jc

HaVoC

Ya i was thinks of doing risk or something along that lines alright(modify the rules maybe), turn based and the world is like an mmorpg active 24/7 or maybe rounds.

A.I. would not be an objective of the project pure PvP game but obviously some A.I. where needed.
Objectives would be more desiging a 3-tier system and all that jazz .
Probably set as low aims as i can and document the scaling of the project and implement them if i can.

oh ya i'm going to Carlow IT in september

theone

fair play you seem to know what your at ,I'm in the same boat looking for something to do for a fourth year project, l better get my arse into gear. I haven't got the foggiest what i'm going to do.