Well of course you gotta know what you actually want to do before you start doing just something.
"Start with" was more or less ment for the point where you know what you want and you are in front of the decission of creating levels first and look how you can bring them to life or you start with simpel test levels to create your functionallity and then just replace the environment when it works.
I think the second one is better because the environment can be replaced more easily.
Of course I had a quite specific vision in my mind when I started creating Dragonrise.
I had a basic storyline and my main characters and the important points of the story where those characters meet and what results from those situations.
On the other hand I was quite sure what kind of game system I want to have (well this slightly changed after 2 years, for better without having too much to change).
So based on this (storyline, characters and game system) I started making those technical thoughts I described above.
I had a certain idea which suddenly came to my mind... over night so to say.
After thining more about it, I made a list of levels and features, without technical details.
Later I figured out the battle system in fact does not work out as great as expected. I had to change the controls, the speed of the game, the camera.
It worked better now, but still not good enough. Changed everything again, added new features I didn't think of before hoping making the gampelay better.
The whole menu system was redone at least 3 times graphic wise. The problem was I had soemthign cool in mind and it was great, but regardless what I tried, I was not able to recreate it completely.
My luck is I that I did the final level details after everything else was finsihed. I had a set of 4 levels with different features I used for testing. The longer I worked on it, the more I realized that I shoudl wait with graphics, as it will all look old if I need so long for coding
At the end I got more as I expected, but still - there are better ways to do.
Probably the best thing was experiencing how to NOT do things. Now I know better.
aybe someone can move this discussion to a separate thread? I think it's very interesting.
