Fast magnetosonic waves under the default test conditions (v = 0, Pgas = rho = 1, B = [.5 .3 0]) exhibit a resolution independent error, falling behind by .17 radians per cycle from nominal.
The error is not constant but "wobbles", indicating the simulation is not running a pure MS wave. There is also harmonic distortion at all harmonics, where sonic waves show exactly zero HD at (one of even/odd, don't remember which) harmonics, down to machine precision.
This is a "code not acceptable, FULL STOP" error. Absolutely no production MHD work can be done until it is hunted down, wounded, cornered and killed with an elephant gun.
Fast magnetosonic waves under the default test conditions (v = 0, Pgas = rho = 1, B = [.5 .3 0]) exhibit a resolution independent error, falling behind by .17 radians per cycle from nominal.
The error is not constant but "wobbles", indicating the simulation is not running a pure MS wave. There is also harmonic distortion at all harmonics, where sonic waves show exactly zero HD at (one of even/odd, don't remember which) harmonics, down to machine precision.
This is a "code not acceptable, FULL STOP" error. Absolutely no production MHD work can be done until it is hunted down, wounded, cornered and killed with an elephant gun.