Lately I have been advancing Texas by leaps and bounds. It's true that the game is still a fair ways off, but it's coming together wonderfully. Here's part of what has happened to the overworld in the past two weeks:
Note: Left is before, right is after. As you can see, I also spent too much time refining the original map graphic.
I'm cutting, cutting, cutting. Things get moved around, and I mean entire buildings. I recently chopped off perhaps 1/3rd of the map, and that will probably be permanent. Another 1/3rd has been sectioned into several parts so that the player will uncover them. And all of this is not even part of the above area.
I've reduced the castle area from the first alpha to approximately 1/10th the size. This will open up as they play.
With each new alpha I can see people getting further and further into the game; the two people who tested the last version both played for a significant amount of time before giving up.
No, and for two reasons.
First reason: giving up is the most common way people finish games. Even major commercial titles are only some smallish percentage of the time actually beat by players. Think of how many games you've started but not finished-- and if you're reading this you're probably a hardcore gamer, the MOST likely to finish!
Second reason: seeing where people give up is the whole purpose of my design methodology (that methodology being user testing, user testing, user testing.)
I'm not pretending to know what problems a player will encounter before she encounters it. I'm just not that good.
In the end, it's my goal to create a game that most players will play all the way through, at least if they get into it at all. If I can accomplish this, it will only be through rigorous user testing and fearless cutting.
Overall I've made good decisions with the design Texas, though of course it's been another epic project. However, I am wishing I had avoided at least one mistake: Too much detail, too early.
ThiThe Cutting Room Floors isn't a deal-killer, but I spent a certain amount of time placing bushes, rocks, trees, grass, and so forth on the overworld. Now I am having to move these things around to fit gameplay, based on how I'm seeing people explore it, and I just end up deleting a lot of these objects.
The editor. Each green "pin" is an object I placed. Hopefully, I won't have to rearrange this particular screen, since if I do I'll have to delete every object.
Ideally I would have left things a bit more empty, and "sketched" the overworld. Then once I had the gameplay fully-in-place I could go back and fill in details like bushes and grass.
Or, I could have algorithms place these things.
I think what I will probably do for my next game is work even more "sketchy", outlining areas things in very simple block colors (so that I can still approximate the "feeling" of areas) before worrying about detail objects and so forth.
But for now it's sufficient to recognize my mistake and move on!