> The "current" one is FE0 = FE_Q<dim>(2); > The "other" one is FE1 = FESystem<dim>(FE_Q<dim>(1),2); > > Now I would like to have some thing as bellow: > > FEValues<dim> FEview(FE0, ....); > > FEview[0].get_function_values( from a vector solution in FE1, to a DOF > array in FE0);
You misunderstand the role of FEValues. FEValues gives you the values of the shape functions *of one particular finite element space* at quadrature points, or by extension the values of a finite element function in this space. If you need the values of two functions from two different finite element spaces at quadrature points, you will want to use two different FEValues objects that correspond to the two finite elements you use, but you would use the same quadrature object for the two FEValues objects to make sure you evaluate them at the same points and in the same order. Does that explain things? If not, you need to explain what you want to achieve. Best Wolfgang > PS.: [email protected] suggested that I can use the > FEFieldFunction. Is this expensive, right? Yes, but only if you evaluate the field at many points. If you do it at a few points, or infrequently, it may not matter. -- ------------------------------------------------------------------------- Wolfgang Bangerth email: [email protected] www: http://www.math.tamu.edu/~bangerth/ _______________________________________________ dealii mailing list http://poisson.dealii.org/mailman/listinfo/dealii
