Is Universe Sandbox 2 realistic?
Universe Sandbox 2 is the second installment of Universe Sandbox that is available on Steam. It is an educational simulation game based on astronomy and anything space, it ranges from supernova animations to climate and atmospheric simulations (and of course collisions) but do any of these features hold water? Do they truly resemble the dynamics of space?
Not sure if these kinds of questions are allowed here, but I may as well answer it.
Universe Sandbox 2 is mostly realistic. Its accuracy degrades as you increase the flow of time. The slower the time, the more accurate the calculations. This
means that if you want to simulate the Solar System over a long period of time, it won't be very accurate. If you want to simulate day-long events, like planets crashing, it might be better.
However, Universe Sandbox 2 uses 300-year old Newtonian physics, rather than general relativity. Newtonian physics just requires n-body mechanics, so it is much easier to implement. General relativity requires simulating the spacetime itself. That is, taking your simulation space, discretizing it to a hi-res 3-D grid and checking the effect that each and every point in that grid has on all neighboring points at every timestep.
Instead of simulating N number of bodies, they would be simulating a huge number of points. They would start with some initial data of the shape of spacetime and then see how it evolves according to the Einstein equations, which are 10 highly non-linear partial differential equations. Accurate general relativity simulations require supercomputers.
Still, Newtonian physics can get the job done for the most part. The developers seem interested in incorporating some GR concepts, like gravity propagating at the speed of light, spinning black holes and ways to better approximate GR.
I wonder... have you ever played the game? You seem to be very knowledgeable about it. Unlike me, although I have it. +1 for relatability and detail, like always.
Modern computers are extremely powerful -- my desktop's CPU alone would have made the Supercomputer Top 500 any year prior to 2001; factor in a high-end video card, and you could probably remain on the list until the mid-2000s. Accurate GR simulations aren't too far out of reach.
@Mark Considering Newtonian physics run slowly with Universe Sandbox, GR would be nuts.
Can you say what N-body and hydrodynamic codes it uses? Given that the best possible astrophysical theory models are approximations to reality, I doubt this could be said to be realistic. But I haven't looked at it. Not using GRis rarely an issue unless you are doing large scale cosmological simulations or studying black hilesand neutron stars. Newtonian physics is hard enough once you get to say a cluster of 10,000 stars.
@RobJeffries I honestly wasn't sure, so I looked it up and found on their FAQ: "Universe Sandbox strives for realism...Thomas, our physics programmer, discusses at length various aspects of the physics implementation on his YouTube channel." They seem pretty intricate.
Approximating GR effects in orbital mechanics in the solar system can be done quite well just by adding a term or two to the Newtonian forces. See answers to How to calculate the planets and moons beyond Newtons's gravitational force? It's an approximation yes, but still gets you most of the way there, and can just be dropped into the integrator as an additional acceleration term. I wonder if something like this might actually be found in Universe Sandbox?