Computational physics problems that involve nontrivial CS concepts?

**Ragnord** · Nov1-09, 10:01 PM

I'm a 3rd year physics major and my friend is a 3rd year computer science major, and for a bit of fun we want to do some sort of project that will utilize both of our skills. Some sort of computational physics problem seems appropriate, but with my limited knowledge, it seems that a lot of them are somewhat trivial from a CS point of view. That is, they seem to involve very simple algorithms for doing a bunch of repetitive calculations, and don't have much to do with data structures and the sorts of things that computer scientists like. So can I have some examples of computational physics problems, or of ways of approaching them that involve concepts which are "interesting" from both a physics and computer science point of view? It doesn't have to be anything groundbreaking or new, even just trying to recreate some well known result would be alright, we're just doing this for fun. Thanks.

Edit: Oops, could I get this moved to Programming and Comp Sci?

**lawtonfogle** · Nov5-09, 02:14 PM

Originally Posted by Ragnord

I'm a 3rd year physics major and my friend is a 3rd year computer science major, and for a bit of fun we want to do some sort of project that will utilize both of our skills. Some sort of computational physics problem seems appropriate, but with my limited knowledge, it seems that a lot of them are somewhat trivial from a CS point of view. That is, they seem to involve very simple algorithms for doing a bunch of repetitive calculations, and don't have much to do with data structures and the sorts of things that computer scientists like. So can I have some examples of computational physics problems, or of ways of approaching them that involve concepts which are "interesting" from both a physics and computer science point of view? It doesn't have to be anything groundbreaking or new, even just trying to recreate some well known result would be alright, we're just doing this for fun. Thanks.

Edit: Oops, could I get this moved to Programming and Comp Sci?

Try to reverse your steps then. Look up some non-simplistic programming issues, like triangulation of a polygon, and apply them to physics.

Or you could just try to implement the O(n) time polygon triangulation algorithm into actual code. I don't think anyone will notice (or care) there was no physics if you do that.

Also are you trying to break new ground, or just do something? You can always try to make a simulator for the n-body problem. Even just doing a lot of simple calculations is impressive programming when you optimize it. Actually, optimized simplistic calculations is the more preferred code when compared to complex arcane programming to solve the same problem.

**dreiter** · Nov8-09, 07:04 PM

Originally Posted by lawtonfogle

Even just doing a lot of simple calculations is impressive programming when you optimize it. Actually, optimized simplistic calculations is the more preferred code when compared to complex arcane programming to solve the same problem.

Agreed.

**harborsparrow** · Nov10-09, 10:58 PM

Some kind of machine learning combined with sensors. For example, people listened to the keys being typed on a keyboard and were able to learn the text begin typed after a while using frequencies of letters and the distinctive key sounds learned dynamically from each keyboard.