Computer simulations have become an important way to study physics. The input to a simulation is of course human-made, and you might ask what news should come out if you know what you put in. The situation is the following: You might know how to calculate the square root of 31, and if you have enough time you can do the calculation by hand. It’s just very much faster to program a computer (or calculator) to do the job. And while you’re at it, you can program the computer to calculate any square root, or more complicated things like integrals, and so on. And when you do that, you also put in what you already know: the procedure to calculate the square root or integral. But you leave the tedious part, the actual calculation, to the computer.
So in a simulation, I put in rules on how to calculate the motion of a bunch of particles, but I leave the tedious part, the actual calculation, to the computer. And when it’s done, I can compare the results to experiment, and find out whether the rules I put in are correct. Or I can find out whether the rules I put in contradict other theoretical results. Or I can make predictions from rules I trust.
Two major simulation techniques: Molecular Dynamics and Monte Carlo.