Talk:N-body problem

Cool Story, Bro!

What we supposed to be doing here? What's the task itself? –Donal Fellows (talk) 14:43, 20 April 2014 (UTC)

I've added a clear task here; simulate a 3 body gravitation system for 20 time-steps. It might not be perfect, but it's a definite thing for programmers to program, rather than just wallowing in mathematics. –Donal Fellows (talk) 07:14, 26 May 2014 (UTC)

Caution

"Beware that because of the inherently chaotic nature of the n-body problem, two different but correct implementations may diverge after a sufficiently long sequence of $\delta T$ steps." -- [1] --Rdm (talk) 00:33, 16 April 2014 (UTC)

tcl implementation

I animated [projections of the data from] the tcl implementation, and I noticed that the second and third bodies practically bounce near the end. You can see this by comparing the values in the last two rows with the previous row. What's odd is that the first body reacts similarly at roughly the same point in space, but two time increments earlier than the other two.

Anyways, my concern is that delta-T is probably way too big if things are bouncing so sharply.

But this raises another, more important issue: how can we be sure our results are correct (or even "close to plausible") for a problem like this? --Rdm (talk) 07:48, 26 May 2014 (UTC)