On Sat, May 30, 2015 at 2:37 AM, Justin Chang <[email protected]> wrote:
> Hello everyone, > > In SNES ex62, it seems to solve the Stokes equation using TH elements > (i.e., using P2-P1 elements). However, this formulation is not locally > conservative, and as a result many people use the P2-P1/P0 elements where > the pressure space is enriched with a piece-wise constant to ensure mass > conservation at the element level. > > Is such an element combination possible with the current implementation > within DMPlex? > I have implemented P1_disc: https://bitbucket.org/petsc/petsc/src/779682a0c231963f14b692ce9dcbb6520a47bf75/config/builder.py?at=master#cl-497 I believe its working correctly based on my tests for function interpolation, but I have not tried anything hard with it. Thanks, Matt > Thanks, > Justin > -- What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead. -- Norbert Wiener
