flipcode - Building A Game On Your Own

Building A Game On Your Own
by (20 June 2002)

Introduction

Over the course of the past three months I have been working on a little game called Lunar Lander 3D. My main reason for embarking on this project was to gain some knowledge about what it takes to create a game. Other than that, I just wanted something to distract me from school. Although my game is simple and small in scope, the process of creating it has given me a great deal of useful knowledge. In this article, I am going to go over some of the important things that I learned while creating my game.

Before I get into the meat of the article, I want to point out that I am by no means a professional. I am just another goon with a keyboard. This article is meant for hobbyist programmers who would like to create a game on their own time. The main emphasis throughout this article will be on finishing the game you are working on, not making a flashy technology-bloated half-game. And when I say finish, I don't mean working on a game for a long period of time only to ditch it because a shiny piece of metal caught your attention.

Your ideas are not sacred

If you ever want to finish a game, you need to realize this. No matter how much you think of yourself and the ideas that you have come up with, you will need to toss most of them out. When I started working on my game I had a mountain of ideas that I wanted to use. I wanted to have moving platforms and hazards in my worlds, a complex camera system and a number of other things. But I knew that if I ever wanted to finish my game, I would have to ditch most of the ideas I had. It is important to remember that when you are working alone, you can't come up with a project that would take a professional team a year or more to make. While a year or so may be OK for a professional team, it is a bad idea for a hobbyist programmer. Try to tailor your ideas so that you can work on your game for a period of, for example, six months or less and you will be much better off.

Don't start from scratch

If you don't have a base of code to work off of, you shouldn't be writing a game in the first place. It is that simple. This may sound harsh, but let's take a closer look at the situation. If you don't have a set of code that you use consistently, you probably haven't been programming for all that long. If you haven't been programming that long, jumping into a game is the last thing you want to do. When I started working on my game I had a large base of code to work off of. I had code for general window creation, input management, particles, math and various pieces of utility code. With all of this code ready to be used, I was able to immediately begin working on things that were related to my game. Additionally, I didn't have to test this code since I had tested it when I originally wrote it. You don't have to write every piece of code you will ever need before you start your game. Just take some time to lay a groundwork that you can build off of. If you feel that you need to build up a code base, write small applications to test things out before you start your game. I did this with my particle, input management and command execution code. It may not be as exciting as a game, but it will definitely help you in the long run.

Borrow ideas

I can't even begin to stress how important this section is, so let me say it again. This section is important. When you are writing a game, don't be afraid to use other people's ideas, libraries or utilities. Borrow as much as you can, it will do nothing but make your life easier. In my game, I remembered Conor Stokes' soft hands and used his article on axis aligned bounding box trees, I used Paul Nettle's collision detection document and I used id Software's Q3Radiant level editor. Why did I use other people's ideas and tools to solve problems that I could have solved on my own? The answer to this question is simple: I used all of these different resources because it saved me time. By using an established method for collision detection, I didn't have to waste my time coming up with and testing my own collision method. By adapting id's map editor to my own devious purposes, I avoided having to come up with my own. And, as an added bonus, id's map editor lit levels for me, which is yet another thing I didn't have to deal with. Don't get caught up in the belief that you need to have custom tools and solutions for your game. If you are smart, you will know when to come up with new ideas and when to use well-established ones.

Avoid the tar pits

What in the world are the tar pits? The tar pits are the so-called 3D and terrain engines that everyone loves to work on. If you are working on an engine in order to create a game, you will be like a mammoth stuck in the tar. No matter how much you try, you aren't going to be going anywhere. The experience of working on a game has made it clear to me that an engine is not needed when writing a game on your own. Remember, you are a single person writing a game, not a company writing a family of games. You don't need an engine to reuse over and over. While a good reason to write an engine is code re-use, this can easily be obtained without an engine. With a reasonable design you will be able to reuse parts from your game in other projects. For example, of all the new code I wrote for my game that is not game specific, I can reuse 100% of it in other projects. This is pretty good considering I never had this in mind as a goal. Writing an engine to gain knowledge from the engine itself and the design process behind it is fine, but you should avoid it if you are trying to create a game.

Don't get cute with graphics

When you are working on your game, don't bother adding graphic features that don't make your game better. You know that cool extension on the GeForce20 that you want to use? Don't use it. You know how you want per-pixel lit, bump mapped, texture animated, per-pixel depth shaded magic snails in your game? You don't need them. If you define your game by the extensions and eye candy in it, you probably don't have much of a game. What you have is a technology demo. The most complex graphics feature that I use in my game is multitexture, but you don't even need it to run the game. I know that I'm not id Software, so I have no business requiring insane hardware to play my simple game. If you keep things simple, you will not only broaden your audience, but you will also be able to focus on more important things in your game, like gameplay.

Conclusion

Well, I hope some of what I have said makes sense. It makes sense to me, but who knows what other people will think of it. If you have any questions or comments, feel free to e-mail me at nathanm@uci.edu and I will try to get back to you as soon as I can.

If you are interested in my game, you can visit the Lunar Lander 3D page and download it.

Nate Miller