Let's make a game....

Started by
6 comments, last by Daishim 18 years, 7 months ago
Hello everyone, i want to start making a REAL game (i mean, not like my crappy demos), where should i start first? Thank you for your help!
Advertisement
ermh.. have you considered consulting Google for your question? What are you expecting that people will guide you to the process of Game development just like that? ...
Well, you'll first have to start deciding what programming language you'll use. Then , if you master this language, you need to choose an Operating System, or Console, and a graphics API- a way of communicating with your graphics card. Then , you need to think, deeply, what your game is all about and how it is fun and appealing to play for other users. (repeat this step several times because this is not easy). Then, be very disciplined, and reserve all your free time, to make your game come true.
The line between "real game" and "crappy demo" is blurred. The more crappy demos you make, the more your demos improve. Eventually you'll find you're not making crappy demos anymore, you're making "real games". Keep up the good work! [grin]

Of course, if by "real game" you mean a published, professional game you should find a job in a game studio or maybe start doing some Indie work with online developers. If you want to have any real hope of a game reaching paying customers, you'll probably need a dev team. It's a tall order to make a "real game" all by your lonesome.
Learn Programming Language > Learn API > Make a game.
-----------------------------Language: C++API: Win32, DirectXCompiler: VC++ 2003
Learn C/C++. Choose either to use. Most use C++. Learn Win32. Learn OpenGL/DirectX. Learn SDL. Learn software design. Learn data sturctures and algorithyms. Learn game type data structures and algorithyms. That's about it...
We should do this the Microsoft way: "WAHOOOO!!! IT COMPILES! SHIP IT!"
I find that making a real game is best done from adding onto crappy demos. Say you have a demo with a user-controlled car that drives around on a plane, meant to learn how to use ODE for physics. You slowly add onto it by making the plane rolling hills (using a sine wave and a grid), then you divide the grid up so you can use multiple textures and octree culling (a nice smooth grid would lower your FPS by a lot, even on newer cards). You might also use hardware tesselation for patches, increasing your FPS even further. Anyway, you could then model a circular road with fences on the sides for your car to drive on, change the terrain to heightmaps, make a better road and a menu that lets you load different "tracks" (a heightmap, road and starting position).

Add checkpoints to the track file, basically just loops (or in my game, torus knots). Make each checkpoint disappear when the car runs into it, add a counter for checkpoints that have been hit, and display a flashy "You Win!!!" messagebox when the counter hits 0. Now, you have a real game, though without a challenge. So you add a simple "perfect" ai that does pathfinding between the points and moves another car model towards them (at slightly slower speeds then the user, of course, since it would be impossible for the AI to lose if it went just as fast as the user AND was perfect). Add a counter for the AI and a "You Lose!!!" messagebox when the AI counter hits 0.

You can add to the game by creating a best times board (possibly saving it in the track file), music and sound, a spedometer, dynamic scenary, an inperfect AI with multiple cars, tunable cars, different terrain, locked courses... Anything you want. And PRESTO!!! you have a fully-featured racing game that originated from a little ODE demo.

The hard part is keeping the project expandable. OOP helps a lot with this, you can just add to your car class or track without worryting much about how this will affecct other classes (though you'll probably have use a lot of "hacks" anyway). The most important point is to keep your demos organized from the beginning, and not to do everything at one, no matter how tempting. If you try to make a complete game from scratch, it'll be a while before you get any visible progress and you'll get discouraged.
Lack of info again, sorry:
I already know how to programm very well in C/C++, and i'm using OpenGL as my graphics API.
What i mean when i say from where should i start is, what are the steps for making a game (e.g. first, create the game core, etc...).
Thanks for the replyes,
PEACE!
Quote:Original post by nuno_silva_pt
Lack of info again, sorry:
I already know how to programm very well in C/C++, and i'm using OpenGL as my graphics API.
What i mean when i say from where should i start is, what are the steps for making a game (e.g. first, create the game core, etc...).
Thanks for the replyes,
PEACE!


There is no definitive answer to your question. Though, a good place to start would be to draw up on paper what you would like your engine to be able to do. Then start mapping out the engine based upon what it needs to do. Then start coding it.

I know only that which I know, but I do not know what I know.

This topic is closed to new replies.

Advertisement