On Oct 18, 2013, at 6:50 PM, James Snyder <[email protected]> wrote:
> If you're still interested: > http://ediacaran.mech.northwestern.edu/~jsnyder/podisoplot_conductor_animation.mp4 > http://ediacaran.mech.northwestern.edu/~jsnyder/podisoplot_insulator_animation.mp4 > > These are done with the 2-D unstructured mesh I was describing above. Thanks very much. These are very nice visualizations. > In theory, you should be able to use the VTKCellViewer to export your data, > but this isn't documented at all except in the source and it doesn't support > anything but processor 0 as far as I know. > > So if I try and dump data out it's going to be segmented to individual > processes and I'll need to rely on globalValue for the cases where it does > work? No, I *think* the VTKCellViewer would take care of exporting the global values, but it would do so by transferring everything to proc 0. Honestly, I'm not sure and don't have time to test right now. The XDMF solution more naturally segments the data (although doesn't have to) and the ParaView and VisIt viewers are happy to deal with that; they're designed around enormous simulations where the data can't all fit on one process. _______________________________________________ fipy mailing list [email protected] http://www.ctcms.nist.gov/fipy [ NIST internal ONLY: https://email.nist.gov/mailman/listinfo/fipy ]
