Analysis criteria

Continuation of #21 (closed)

It remains to add some analysis criteria, e.g.

• various error norms
• convergence rates based on error norms

The above are common criteria for any simulation. Later on we might want to add more specialized criteria, e.g. rest mass conservation in the GRHD case etc.

Regarding error estimation: In some cases we have analytical solutions available (Advection equation), in others we have to fall back to simulations computed with very high resolution and a robust method.

Questions:

• What method or which program to use for reference results? A: Ideally, if all methods are consistent, which means they converge to the same result, then it should not matter which method to use.
• If analytical solution is given, then pointwise comparison is not a problem, because we can evaluate the exact solution on the desired points. However, for a reference resolution, which is likely to be computed on a different grid, we have to do some sort of interpolation. What interpolation to choose? Would linear one suffice if the reference resolution is really, really, really high resolved? A: Yes, let's go with very high resolution reference results and linear interpolation.