in discontinuous galerkin method,a node varible may be shared by different cell,thus it has multivalues.because i want to plot a pressure isoline,and how do i treat the multivalues in the node in TECPLOT? thanks a lot

If your DG code is based on modal approximation, you can choose the piecewise-constant value at the cell center.

but i choose the nodal basis,the modal basis DG is underbuiding

Can I ask you something? Why did you begin with nodal based one? I wonder if there are differences between one another.

nodal basis belong to H adpapive,while modal basis can build P apdapive

Here is a suggestion.

Plot contours within each element. You could divide each element into smaller elements thus creating new nodes. Then set the solution at these nodes using the DG solution. Use these to plot the contours. The solution is discontinuous across elements. So you cannot any standard software. You can generate a contour data file which can be used with gnuplot. This way you will be able to visualize the discontinuous solution. See NS2KE for a subroutine which generates data for gnuplot contour visualization.

Nodal bases produce natural sparse preconditioning matrices.

So, can you insist that one is superior than another?

but where to get NS2KE?

http://pauillac.inria.fr/cdrom/www/nsc2ke/eng.htm

See the files result.f and isovat.f in nsc2ke.

thank you very much