| Commit message (Collapse) | Author | Age |
|
|
|
|
|
|
|
|
|
|
| |
The BSSN_MoL tests either used LegoExcision (which we will remove) or have
equivalent ML tests.
The ML tests now use fdOrder and still pass with the old data (only
modification to hdf5 file was renaming of datasets and regeneration of the
Parameters group)
git-svn-id: http://svn.einsteintoolkit.org/cactus/EinsteinAnalysis/AHFinderDirect/trunk@1563 f88db872-0e4f-0410-b76b-b9085cfa78c5
|
|
|
|
|
|
|
|
|
|
| |
* use ReflectionSymmetry instead of the deprecated bitant symmetry
* set number of timelevels to minimum required by the prolongation order.
Otherwise recovery fails inside CarpetLib.
* use CoordBase for domain description
git-svn-id: http://svn.einsteintoolkit.org/cactus/EinsteinAnalysis/AHFinderDirect/trunk@1559 f88db872-0e4f-0410-b76b-b9085cfa78c5
|
|
|
|
|
|
|
|
|
|
| |
3rd order interpolation in order to avoid an issue with different
interpolation stencils being used for certain points due to roundoff
differences in the coordinates when running on a single or multiple
processors.
git-svn-id: http://svn.einsteintoolkit.org/cactus/EinsteinAnalysis/AHFinderDirect/trunk@1555 f88db872-0e4f-0410-b76b-b9085cfa78c5
|
|
copies of AHFinderDirect's recover and checkpoint testsuites modified to use
McLachlan instead of MoL_BSSN. Centroid location does not aggree with results
in old files, but AH area etc do. Even re-running the old test suites with
BSSN_MoL shows differences to the saved data.
git-svn-id: http://svn.einsteintoolkit.org/cactus/EinsteinAnalysis/AHFinderDirect/trunk@1543 f88db872-0e4f-0410-b76b-b9085cfa78c5
|