[Nek5000-users] Visit reader for double precision f files

[nek5000-users at lists.mcs.anl.gov]

Dear Neks,

Reference to previous discussion (Sep.13th) about the support for the reading 64 bit files of nek5000 in VisIt, we have put up together a 3D cylinder as test case. For a double-precision format nek5000 dump file the 3D view and especially iso-contours are working fine whereas any 2D projection slice, or cut of the data is not working properly. In general operations considering the whole grid seem to work fine, but those related to a section of the grid such as box don't.
This anomaly is happening regardless of choosing various pressure solver scheme (Pn_Pn/Pn_Pn-2) or including the grid data to be dumped in the file or not. I was wondering if you can help me with this issue.

With many thanks
Reza

P.S. The 3D cylinder test-case located at the origin (0,0,0) of the coordinate system and extruded along the z-axis (perpendicular to the xy plane). The box extends from -15 to 25 in x direction, -9 to 9 in y direction and -2 to 2 in z directions. The mesh is constructed from the 2D cylinder in nek5000 repository. The number of elements is reduced, then extended in the z direction. An example of the dumped file can be found in the ftn:
ftp://ftp.mech.kth.se/pub/pschlatt/nek5000




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/nek5000-users/attachments/20121116/925b6737/attachment.html>