View Full Version : Project idea / discussion

05-06-2003, 10:11 PM
Howdy gang,

I'm new here and have had an idea for a game for a while now. I am a Delphi 5 programmer with experience writing business like applications, but very limited experience in the game arena.

I'd like to toss some ideas out there and see what your collective thoughts are on the best way for me to go about making this happen.

Myself and my friends, play several table top games.. Games such as Warhammer/Warhammer 40k.. etc.. They deal with a lot of little miniatures and gobs of rules governing their actions on a big 4x8 table.

I would like to try to make a 3D engine that will allow players to play a similar type game on the computer.
Each model has stats and point costs etc., players take turns making their "army" do their bidding.. The games are very strategic.

This isn't a first person shooter style game at all, but rather a turn based strategy style game in which everything is decided on dice (or random computer) rolls. (Thus 3D collision detection is not needed..)

The thing is, I'm not really sure where to begin.. I understand 3D graphics as I am a Lightwave user (and having been for years) so I can make the 3D models. But the main crux of the game engine eludes me..

I have played with GLScene a bit in the past and understand it's basics but really not how to build an engine around the GLScene engine.

Where should I start?

Luckily, I have all of the games rules already printed out in a nice big fat manual already edited and everything, so I don't have to come up with that..



06-06-2003, 08:37 AM
My first question would be, are you familiar with programming games? I mean, have you already created something before?
Because if that is not that case then I would not recommend starting with opengl. Why not try 2d first. It's easier and it will give you a good understanding of what it takes to make a game.

The "nice big fat manual" suggests a really large project, perhaps a difficult one as well. One that could easily take a year or two to develop. So why make a difficult project even harder with an engine that is hard to build as well?

Perhaps an 2d isometric engine would fit your current plans as well. There are a lot of resources on the Net about those and probably even more people who can help you with coding problems.

Anyway, those are just my thoughts

PS. I like your site. The only thing that bothers me is the menu. It's very slow :(

06-06-2003, 09:40 AM
Hmmm, nice idea, Warhammer like thingie on the PC.

Ofcourse, yo do realise that you WOULD need a collision detection, because what would happen if a player (accidentally) told a unit to move inside another unit? That shouldn't be possible....

Anyway, I'd also go with plain old 2d, because I am getting sick and tired of games putting 3d in there, just because 3d exists. In most cases, unless it serves a gameplay aspect, 3d is unnecessary and will only unnecessarily slow PCs and performance. A good example(IMHO) of this is Neverwinter Nights, they should've just stuck with the good old 2d INfinity engine, but they made it 3d, which makes it run very badly on my PC, and I've seen it on other PCs, and the graphics have NOT become better since BG 2.

((Nevermind that rant about NWN, though. I do believe however, that you should seriously do it in 2d. It simplifies things, and it can give cooler graphics...))

06-06-2003, 05:51 PM
Traveler; I guess the best possible answer in the shortest and most simple terms is.. "No." I'd love to lie to you or fool myself and say that I am familiar with programming games, but the simple fact of the matter is, I'm really not.

I have written fully functioning games before, but that was back in the day of my Commodore 64 or my Amiga 500 on which I wrote an Artillery Duel / Scortched Earth simulator. Nothing fancy, but fun all the same.

Back in those days though, it was not an objecto oriented / event driven language, but rather very simple basic.. 10 Do something. 20 Goto 10.. etc..

I understood the main "loop" if you will. I guess with a game engine there still must be a main loop, but just don't fully grok the way it happens these days..

A friend of mine had suggested a 2D Isometric engine as well, and forwhatever reason that hadn't occured to me.. I have accepted that as my new engine and am most likely going to go that route.. but graphics engine asside, I still have the game engine to consider.

Once the game engine is written, I should be able to plug in any graphics engine right?

And yes the game can be quite complex, but I can easily start small and still use the book because there are many small areas that *should* be easy.

And thank you for your thoughts on my site.. I have only had a "slow menu" complaint once before... For all seven of my computers, they are near instant. Can you tell me what browser version and computer system you're running so I can try to fix it?

Shadowcore; Actually now that I am going to go 2d, collision detection is way moot but I think I could get away with not using it simply by using a rules system for not allowing two models to occupy the same space. Because this game would be played mostly on a flat "board" instead of in 3 space, collision detection can be avoided.

And I do agree with your NWN rant on some level. Though personally I thought the game would have been way cooler had htey gone even MORE 3D.. Either NO 3D at all, or 100% 3D and make it first person.. But that's for another thread.. heh

07-06-2003, 08:28 AM
I have been experimenting with different games I have in my library and it occurs to me that the game Cossacks which isn't a bad looking engine, would server nicely for what I am considering.. just zoomed in a tiny bit..

How do I go about making a Isometric engine similar to that?

Also are these boards always this dead?

07-06-2003, 10:40 AM
I don't think you can just plug in any graphics engine. 2D or 3D are very different from eachother, a square based tilemap offers less options than an hexagonal tilemap. With a strategy game like yours I would go for the grahics engine first. That way you know what it's limits are, what variables you have for calculating sprite movements for example.

For my own games I usually follow this path

Create a tilemap
create characters (just the image and/or animations)
create movement rules for character (how fast, limits, forbidden areas) (see why I needed the tilemap first ?)
create character stats (hit points, lives, walk points etc)
create combat rules
create ai

I'm sure you could add more to this list, but the idea remains you'll need the basis first.

Then the matter of collision detection, I believe Palamedes is correct when he says that no collision detection is needed. Afterall we are talking about a turn based, 2d, game. (like x-com, civilisation) This means you'll need somekind of tilemap. But the only thing you need to check is whether a tile is already occupied by another sprite.

Cossacks is indeed a great game, but I'm not quite sure how they made it. Have a look at the editor or the fog of war, the hills, the water all very nice... really wish I could help you here...

I have checked your website on a different pc, and it worked fine on that one. I think it has something to do with the graphics card in the other. It's a diamond viper 5000 if I'm not mistaken. Not exactly the fastest card on the market :(

07-06-2003, 06:24 PM
Well, about the collision detection, it all depends on how far you want to take the engine itself. If you just want to limit movements and occupations to tiles itself(FYI(in case you didn't know yet) tiles are the parts the entire map is made up of), real collision detection won't be needed(because you can just check occupation), BUT if you are going to go with all of the possibilities of for instance warhammer, I think colision detection should be needed, because units and such can occupy parts of tiles, and can be angled etc.
But the collision detection is only of later concern, so I wouldn't worry about it too much, also, in 2d, it shouldn't be too hard.....

I would indeed start with the graphics engine, since you'll run into some problems if you don't, things like AI and a lot of rules have to do with a movement map, so it is best if you start with that part first, and then to be able to see it, you'll need a graphics engine.

Also, if you're looking for a strategy game which is like warhammer, you should take a look at the warhammer strategy game which was made a couple of years back(can't remeber it's name, though). It might give you some ideas.

11-06-2003, 12:25 AM

The previous game was called chaos gate.. and it was baaad.. heheh

I have already started on the rules engine that will make the game happen.. weeeeeee

12-06-2003, 10:15 AM
Have a look at this code (http://www.planetsourcecode.com/vb/scripts/ShowCode.asp?txtCodeId=69&lngWId=7) (a game of Asteriods). It's the most object-oriented game example you'll find and you'll probably pick up a few ideas on how to structure your code. There are a couple of things to fix before it compiles, however. First, you need to make the files writeable (some are read-only). Also, it may complain in a couple of places about "TInput" -- change the variables to be of type "Input.TInput" (i.e., prefix the unit name first).

The only thing I don't like about that code is the indentation/hungarian notation. Just say no to hungarian notation.

Also, welcome to the forums. Let us know how your game progresses.

13-06-2003, 05:24 PM
Thanks Ali,

I have downloaded it but haven't had time to look at it yet.. They actually expect me to do work at work.. sheesh!

I have been working in GLScene to make a 3D isometric style game.. it's going very well really. Taking it one step at a time..

16-06-2003, 06:48 AM
wow, sounds cool,... anything we can view already ?

16-06-2003, 04:51 PM
Eventually.. Not just yet..

At some point I am going to post code snippets to see if you guys know of a better way to do things..

This is a long way out.. I have to do all of the 3d modeling, art and textures, programming..etc.. all myself.. so it's not somethign that will be finished any time soon..

16-06-2003, 06:48 PM
Heh :) We all have to walk that path. And I know, it's long and full of bumps 'n holes, but in the end it's all worth it. :D