> You might even be able to tie them together by using the same nodes (so that > the solutions are continuous and the sparsity pattern is correct). > > You might not be able to read this mesh from a file (although I think Exodus > supports it). But if you just manually add a 2D element to a mesh and then > add a 1D elementŠ why wouldn't it work? If you give them different subdomains > then you should even be able to restrict variables to the different pieces.
I don't think there is anything fundamental that will keep it from working, but you'll have to be careful in user code. Right now if you just iterate over all the elements in the mesh and call fe->reinit(elem) I think the underlying FE and Quadrature rules are expecting all the same dimension. Of course you could get around this by manually creating multiple FE objects and calling the right one, but that is a hassle that would be nice to avoid. And I'm not sure how broken this is currently (if at all?) - perhaps just overzealous asserts... -Ben ------------------------------------------------------------------------------ Virtualization & Cloud Management Using Capacity Planning Cloud computing makes use of virtualization - but cloud computing also focuses on allowing computing to be delivered as a service. http://www.accelacomm.com/jaw/sfnl/114/51521223/ _______________________________________________ Libmesh-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/libmesh-users
