Here is the Zeldovich Pancake compared against Enzo again, now I show different Reconstruction and Solver combinations :

Only Advected Internal Energy

The Parabolic reconstruction methods look much more similar to the Enzo evolution

Dual Energy Off: Now I compare when Dual Energy is off on both codes

Surprisingly the evolutions are not that different.

If I set \(\eta=0.001\) ( supposedly ) equal to the value used in Enzo I get that the gas that is using the Total Internal Energy follow the same evolution as the noDual_Energy enzo, this indicates that we are not using the same prescription for selecting which internal_energy to use that Enzo is using:

One more thing: When Using Advected Internal Energy the heating comes from the \(p \nabla \cdot \mathbf{v}\) term, by changing the implementation to:

\[p_{j}^{n}\left(\frac{\overline{v}_{j+1 / 2}-\overline{v}_{j-1 / 2}}{\Delta x_{j}}\right)\]

now the results are slightly different, in the animation below newPDivV is for the evolution using the updated \(p \nabla \cdot \mathbf{v}\) term.