There now is a minimally tested minimal version of a minimal shift
elliptic solver in cactus.

thorn_vectorlaplace computes the three component vector laplacian and
the source vector for 
   VectorLaplace(u)^i + N^i = 0.
So far minimal distortion and minimal strain are implemented.
The routines hooked up right now I produced using BAM's Mathematica
script. Ed Evans is working on Greg Daues' minimal distortion code. 
At the moment, most coefficients are computed on the fly, which is memory
efficient but quite inefficient in speed.

thorn_BAM_Elliptic is needed for the vector elliptic solve. Ed is also
looking at making a SOR solver available.

Reread the first paragraph :) The basic setup seems to be sound, but there
are lots of things to be done.

In particular, Carsten promised to add some functionality to thorn_exact
which allows testing with arbitrary slicings of analytic spacetimes.
Right now the only simple test I performed for the convergence of the
algorithm is based on the "flat Schwarzschild slice" of thorn_exact, 
see thorn_vectorlaplace/par/minshift.par.

Let me know if there is any feature you would like to put on the wish list
for coupled elliptic equations. Ed and I also plan to implement the
coupled conformal constraint equations for initial data problems.

Bernd


   

Created for the WUGRAV CoCoBoard Page Projects Page.

Created by The CoCoBoard.