I spent some time looking at the Torque Game Builder 1.51 on Tuesday night. It has been almost a year since I've had the time to play around with the tool. I was a little nervous because I proposed writing a "research layer" for an independent study I'm doing at Teacher's College, Columbia University. I was happy to find that not only should TGB support this approach, but it has a new feature that will make it insanely easy for developers and researchers to add these research components to their game objects.
I am referring the TGB's addition of 'behaviors.
A behavior is essentially custom code that can be attached to a game object via the designers interface. So, for instance, a game sprite could have a Thrust move behavior (asteroids style), a takes damage behavior and so on. I am hoping to write data collection behaviors so that academics who want to research what makes games good for learning can focus on making good games, rather than on getting data back from their tool.
Why do I think this is a good idea?
Well, research methods are pretty flexible, but once you have defined the scope of your study, you need to be sure that your game will be recording the data that you will need to analyze. Truth be told, we academics are more likely to be comfortable writing the code for data collection than we are for producing a game that anyone will find enjoyable to play. It's not that we can't write good games. (well, maybe it is... But with people like Jessica teaching at TC that won't be able to be said for long.)
My goal is to make data collection so simple that the effort of subsequent projects at TC using TGB can be to hone gameplay.
As such, I am porting a game concept that I had prototyped in director to TGB and writing custom behavior to report research subject data. Ideally, this will allow a researcher to view the results of an entire level being played by each subject, and will allow aggregate data to be mapped as an overlay on top of the game world.
Hmmmm.... where have I seen something like this before.....?
I wonder how much of the research needs of academia could be met in larger commercial engines from the kinds of metrics work that Darius, Jeff and Craig are doing up at OrbusGameWorks. At any rate, I think I may need to pick Jeff's brain on some of the finer points of metrics API development.
Comments