Revisiting an old project

2 minutes read

Well, I've decided to revisit an old project that I haven't worked on for a while. The project was a game called "Tanks!" that was going to be the basis for a set of articles on game development with Managed DirectX. I'm still doubtful that the articles will make a come back but I definitely want to revive the project, with a twist that is.

Instead of a flat out Tanks game, I'm going to make it more of a programming game. Think of Terrarium meets Robot Battle (or Mind Rover (choose one)) meets Combat. You will be able to create your own Tanks in .NET that will be run in the game itself. I've been wanting to explore this kind of idea for a while and I think this might server as an idea vehicle (no pun intended) for doing so.

The few of you that attended the talk I gave way back at the Portland Code Camp might remember me talking about the architecture of the engine. Well, that will _radically_ change now. I don't know if I've just matured (doubtful) or just have a fresh pair of eyes to look at the problem (more likely), but I've realized that the messaging-based architecture I was using was WAY overkill for the problem. When it comes to the saying "using the right tool for the right job" well, I was using a chainsaw to slice up cucumbers for a salad (and in my opinion it shows).

For the first version I will be ripping out the physics engine entirely (as I don't think it really helps with the gameplay itself). I will be ripping out the messaging infrastructure. This should not only make it less overkill, but it should make it more approachable for the less experienced programmer as well. It may even be that the first version has just manual controls to control the Tanks as well. So be it :).

Maybe this time I can bite off a significant less chunk and see it through to at least the first public release (which would be a drastic step up for me (those of you who know me, feel free to shake your head in enthusiastic agreement at this point)). This time around, more unit tests, more integration, more re-use/adoption rather than roll-your-own.

I think it should be a fun thing. Who knows, maybe I can follow it through to completion with at least programmer art (and I'm sure some of you wouldn't mind since a finished set of articles).

Updated:

Leave a Comment