I'm doing a CS course and am having trouble (but enjoying) learning about how to make games. I used to want to pursue a career in games development but I'm no longer interested in it - I'd rather have complete control over game development (then I only have myself to blame!) What I'm wondering is how advanced can a game be if you're making something in your own time? Suppose you want to make a game like Crash Bandicoot (graphics, level design, gameplay) over a few months: - You have several hours each day - You've 1 or 2 years experience in coding (C++, Java, Haskell) - You have an open source game engine - You do your own artwork - You're good at sketching but are clueless on how to use 3DS Max, Maya, Blender etc. - You've already implemented certain feature in a Space Invaders game such as collision detection). Do you think it's realistic to make something as good as Crash Bandicoot or better? Even if it's just 1 level. I'm interested in making games as a hobby for my own satisfaction as well as making me look smart on my CV. It's disheartening when people say that you're not going to get very far working on your own.

Crash Bandicoot is a bit ambitious. Realistically there is no way you could do even one level in the way you describe. There is just too much code, artwork, animation and 3dness.

But there are lots of games that can quite easily be done by one person, and still make money: Casual games. You should start with something like that.

Building a game involves several different things - game design, writing, programming, artwork, sound, music, etc. And its tough to be good and efficient at all of those. I have been coding up a turn-based strategy game in the vein of the old SSI/Ubisoft Panzer general series (I am actually planning on being able to emulate any of the games from theat series). After about a month I have the basics down, including map and scenario editors, and most of the game engine framework (though no AI), but the artwork is still very primitive (no images- its all geometry and text) and there is no sound. I know its not something thats going to be sellable but it may impress niche wargamers who liked that series(there still a small core who still play it) and have been trying to get their hands on the source code to fix the bugs and add some new features (which, if I finish this right, won't be necessary!)

I understand what you guys are saying but with a game engine wouldn't it be easier? I'm using one called SAGE (Simple Academic Game Engine), the tutorials are shorter than I thought they'd be but would take a while to sink in.

Try it and see [grin]

maybe if you had more coding experience / game-making experience. Plus if you can't stick to your computer programming class then can you really stick to a long-term project like a game?

spaceJockey123,

I admire you for being interested in pursuing development on your own. I'm only a hobbyist developer myself but enjoy working on games by myself--it gives me ultimate control over how the game will appear. My latest (finished) game is called Marble Tactics, and I completed it in about 3 months in my spare time. I consider myself lucky enough that my game is hosted on GarageGames's website:

http://www.garagegames.com/products/torque/tgb/demos

A Crash Bandicoot style game may be a little ambitious, but if that's what you really want to do then I say go for it! However, just remember that the more effects, levels, and pretty graphics you want in your game, the longer it's going to take. So, for example, if you only want to spend a year making a game, I would try to make a game that would take a team of 2 or 3 people maybe 6 months. Since you are creating everything yourself, you'll need to take into consideration the extra time it will take.

Best of luck!

@Funkymunky:
I'm still doing my computer programming class, I never quit it. I just have different ideas as to what I want to do after my degree.

@OrangyTang:
Thanks, I thought I was the only one that thought the AI and physics in Crash Bandicoot were quite simple. I think I'll give the 3d modelling thing a try as it's something I've always wanted to learn to do, though I'll probably make it as simple as possible and then take it from there.

@morx:
Once I have a better understanding of the kind of code that needs to be written and get some stuff on the screen I might convince people to work with me! Maybe one day my game will be on www.garagegames.com ...

I've been making games as a career for 2 years (and as a hobby for about 6), and I wouldn't be able to build Crash Bandicoot in a few months... Well... maybe. But only if I was able to steal all my art/animations/audio clips from someone.

I am by no means trying to persuade you from trying. All I'm saying is that you should try to have fun with it and not worry about how long it takes you. If you already have some engine, why not just get a character or some scenery (preexisting) loading, rendering, and then maybe animating. After that you could add code to move your character around (even if the world is a big flat coloured cube). If your engine has tutorials then use them as a sandbox and try all sorts of things. Then you can port these things to your game.