On 02/24/2012 03:00 AM, Roland Haas wrote:
Hello Christian,
I am wondering why MoL does not have support for arrays anymore (the provided functions are all commented out). I guess there is a particular reason for this?
As far as I understand it, the problem is with mesh refinement and the step-ahead in the Berger-Oliger scheme. Before a fine step is taken, the coarser grids step ahead so that there is data for prolongation. For an array this would require extra copies of the array for the stepping ahead. If you never feed the array content back into the grid then you might be able to get away updating in global mode (or on the finest level) and essentially do a unigrid evolution.
The discussion is here: http://cactuscode.org/pipermail/developers/2007-September/005428.html (Tanja Bode had the problem already while at PSU)
Yours, Roland
When was this changed? The RIT code uses MoL to evolve grid arrays (we use the Maxwell version of ET). It requires some nasty code to get the arrays evolved correctly and you need to make sure that, for a given a level, there is only 1 component per processor.
Users mailing list Users@einsteintoolkit.org http://lists.einsteintoolkit.org/mailman/listinfo/users