Hello! I want to get started developing games. I know a little C++. What would you recommend me to do first?

I think the most important thing I can recommend to you is that you decide what you like to do. Do you want to focus on learning to program, does your heart beat faster when you make a game work? Something else?
What kind of games do you like? Simulation games, 3rd person shooters, platform games, strategy games, puzzle games, adventure games? (I am sure my list is not complete, but hopefully you get the idea).

Set yourself a goal, and try to reach it.
It is important to think small. Smaller is easier to understand, and easier program, it gives results quickly, and the project doesn't take too long.
Last but not least, if you get stuck, it is easier to explain your problem to others, and get a bit of help.

To give an idea of small, think Freecell or that mine-hunting game.
They are not going to be the next blockbuster game, but games like these do contain a lot of the problems and techniques of larger games, so knowing how to solve these is a step in the right direction. Their size makes that you're not stuck with it for several years, like I am with my game :)

_________________
My project: Messing about in FreeRCT, dev blog, and IRC #freerct at oftc.net

Hi and thank you for the reply. I will start brainstorming for a little game tomorrow!

I have something in mind but it's complicated so I don't think I will succeed soon.

In the mean time I will watch tutorials and try and write another kind of pacman game.

That's ok as long as you are aware of that.

For a first game, that sounds quite complicated already, the ghosts move independently from player actions, which needs to be controlled.

A simpler game is that falling blocks thing (sorry, I am terrible with names), where differently shaped blocks fall down and you have to place them at the right spot so you get filled lines.
A breakout clone is a similar game.

These games 'just' need user-input, display of the field, and movement in it. The only movement here is time-based. It would give you experience of outputting everything in graphics, instead of by using text as with 'normal' programs.

Even simpler would be the fifteen-puzzle, or hangman, where the only actions are done by the user (ie you can always wait until the user does something).

Both games have additional stuff around it though, introduction screens, manual, high scores, main menu selection.


@general:
There does not seem to be a Wiki page about this, is there?

_________________
My project: Messing about in FreeRCT, dev blog, and IRC #freerct at oftc.net

Don't be discouraging man, gamedev needs to be challenging, and pacman is an excellent excercise. It got very bare physics, ai, scores, levels, everything that marks the basics of a game.

Speaking of which, is there pages in the wiki with game templates for mentioned titles?

_________________
Long pork is people!

wzl's burrow

Start small. That's the best advice I can give. Make something simple, small, get itn working. Once it's solid - add to it, expand on it.

A pretty generic excerise for an aspiring programmer is to create a basic breakout clone. Do that, and do it well - and you can probably go on to create something special.

It's a simple excersise to challenge your basic programming knowledge, it also covers things like a game loop, colission detection, dynamic spawning/despawning of items.. score tracking.. basic graphics implementation.. sound design,etc.

Think breakout clones are boring? Check out a game called Shatter. I'm still playing that one.

start small, and expand from there.

_________________
"None are more hopelessly enslaved than those who falsely believe they are free."
"It is no measure of health to be well adjusted to a profoundly sick society."
"Hope is the first step on the road to dissapointment." -Jonah Orion
http://tankzgame.blogspot.com

By that standard, my FreeRCT project seems like a feasible first game :D


I'd rather err on the too easy side with advice for new game programmers, tbh.

Making the switch to a game loop, and continuously output the right graphics at the right time, is something that needs getting used to.
Remember that most programming books/tutorials assume single run, from input to output, programs.

If that switch is simple for him/her, the game is created quickly, giving a good first experience ("that was easy, I can do this"). Experience that can be used in the next project.

When you aim too high, it just feels like climbing an endless vertical wall. It sucks, everything is against you and you're not making any progress. "Game programming is too difficult for me."
(no offence to the OP, but new game programmers don't know how to aim, and it is VERY easy to aim too high)

_________________
My project: Messing about in FreeRCT, dev blog, and IRC #freerct at oftc.net

I went ahead and wrote a few pro's and con's for the classic titles, since you can not really say 'meh it's to hard for you' it all depends on the knowledge and desires of the programmer what he wants to learn, and pointing in the right direction is probably a better thing than just saying 'you should do this because its easy'. Easy doesn't help if it will not teach him the right stuff

http://content.gpwiki.org/index.php/The_First_Game

Feel free to extend.

_________________
Long pork is people!

wzl's burrow

Thanks for all the attention! I am watching a tutorial now, a simple one, and when I went through that I'm gonig to start with my pacman game. Btw I have the basics of a complicated game in mind already, for the future!