Yet, he didn't write a scientific paper about what he considered "just playing around". Same as i wouldn't write a paper about writing "hello world" in visual basic. Yet, i could become a world famous programmer, scientist and then write a paper about whatever started with learning how to program. There is a big difference here.
Feynman worked on the plates problem just for his own, he didn't intent to publish a paper on "how does a sign on a plate look when the plate wobbles" and certainly he didn't expect people to tell him how important his scientific research on plates could be for the world :P
In fact, it's not even a realistic simulation. I think you may agree when you have ever been at a heavy metal concert.
Don't get me wrong, I partly agree with your stand. My problem is that is hard to dismiss something just on the basis of "it is not science since is not useful", because is almost impossible to draw a dividing line. Many useful discoveries were the product of people just "fooling" around and I believe that this attitude should be encouraged in science. Yeah, you will always have those guys who go and analyze how people slam into each other in rock concerts, but they are just outliers, most people out there are producing research that has some kind of usefulness.