Follow

Centroid N-faced Decomposition as a Preference Setting

We have recently found a problem in EnSight relating the use of "Centroid" N-faced Decomposition prior to the dataset being loaded. This would most often occur if you have your default Preferences for N-Faced decomposition set to "Centroid". The problem will also occur if you have set the N-faced decomposition method to "Centroid" prior to loading a dataset (perhaps a Case Add).

If users have specified "Centroid" as their default N-faced Decomposition prior to loading their dataset in, EnSight may experience a crash when users attempt to make a clip, isosurface, or subsequent operation on the fluid domain containing N-faced elements.

The problem only effects users of EnSight 10.0.1 and EnSight 10.0.2
The problem only effects users who have Centroid specified as their default N-faced decomposition before loading a dataset (typically in the Preferences File or changing to Centroid before loading in the dataset).

There are two potential fixes or workarounds for this problem:
1. If you'd like to stick with your current version of 10.0.1 or 10.0.2, then 
    a. Simply change the default preferences back to Concave, restart EnSight. (Edit -- > Preferences -- > Data N-faced decomposition method)
    b. Once in EnSight, if you need to change the N-faced decomposition to Centroid, you can change this interactively via Edit -- > Preferences -- > Data , and change the N-faced decomposition to Centroid after loading the dataset (changing after loading the dataset into EnSight bypasses the problem).
 nfaced-decomposition-preference.png

2. A proper fix to this issue has been implemented in EnSight 10.0.3(a). If you are able to update to this version, you can continue to use the Centroid Setting in the preferences file.


FYI the following is the directory and file where this preference is stored by EnSight.  

./ensight100/ensight_data_prefs.def

and the command in this file would be:
prefs: nfaced_decomp_method centroid


Was this article helpful?
0 out of 0 found this helpful
Have more questions? Submit a request

Comments