Below is the density field for an isentropic vortex centered on (0,0) with the element rotated 140 degrees about the centroid. The element is an RT7.
I replaced the prior delaunay triangulation with a golang API wrapped on the C "triangle" library, which fixed a bug with order 4 element triangulation in this project. That was causing the inability to support runs with P=3 and some other orders, which won't happen anymore - yay!
I'm planning to simulate an attenuated shock wave that would look similar to a shock captured by the method using Persson dissipation. With this in place, I'll test methods to damp the interpolation such that a TVD principle is held - no new minima or maxima will be created. The concern with this is that in smooth regions, new minima and maxima should be allowed, such as the case of the isentropic vortex, so I'll be testing the damping mechanism and shock finder to ensure that we don't smear smooth fields with the damped interpolation.