Dafang, Ayla, hi,

Dafang if this is like your hybrid meshes in the past, as I recall it seemed like you had rather large (relatively speaking) contiguous regions of a single element type with well defined interfaces between them.

If that is the case you could perhaps treat each single-element-type "sub-mesh" as a single mesh object in SCIRun and then just visualize them jointly in the same viewer window? would that work?

Otherwise, Ayla, I'm guesing that it would be too messy to create a new derived hybrid mesh type where, say, each element had an extra field with an index to determine which element type that mesh was? I can imagine this might break a lot of stuff that tried to use it :-)

best,

Dana

On 11/1/13 2:35 PM, Ayla Khan wrote:

6EB1F29A-3C18-4283-8962-BB93907C654D@sci.utah.edu" type="cite">Hi Dafang,

The SCIRun field datatype only supports one mesh type at a time. Either ASCII or Matlab formats should work. If you want to use the Matlab file format, the documentation page for the InterfaceWithMatlab module explains how your data should be formatted for import into SCIRun.

Ayla

On Oct 31, 2013, at 10:04 PM, Dafang Wang wrote:

Hi,

I am trying to visualize a 3D finite element mesh that contains a mixture of hex/prism/tetrahedral elements. I am wondering whether SCIRun can visualize such data and if so, what kind of file format is the most convenient. My mesh is in both the ascii and matlab formats.

Thanks very much for any comments or advice.

Cheers,
Dafang
--
Dafang Wang, Ph.D.
Postdoctoral Fellow
Institute of Computational Medicine
Hackerman Hall, Room 218
Johns Hopkins University, Baltimore, 21218
http://lagniappe.icm.jhu.edu/~dwang/