The averaging method is equivalent to a "blur" or "filter", and reduces changes from cell to cell. Scientific decisions based on the resulting isosurface may be incorrect. For example, data values above a critical threshold may not be found.
An alternative is to generate a new grid with nodes at the cell centers (CC) of the original grid. Then, the original cell data is assigned to the nodes of the new grid. Isosurfaces will be based on the original data without averaging.
There are at least two reasons the CC method may not be suitable. First, it assumes that the data changes linearly between the cell centers instead of staying the same across the cell. Second, the new grid stops half a cell from the boundaries of the original.
-----The attached network starts with a field where cells have a scalar value of 0 except that the center cell is1. Then, each conversion method is used to produce an isosurface and an image. The original grid, drawn with lines, is shown in both images.
In the image using the CC method, an isosurface at a value of 0.13 is shown along with the new grid, drawn with tubes.
In the other image, using the 'Post" method, there is no isosurface at all because the averaging has resulted in a maximum of around 0.125.
Run the isosurface value up and down to see the difference between the two methods. If there is enough interest, perhaps someone could create a macro called, say, PostCC..
Kent - - - Kent Eschenberg [EMAIL PROTECTED] (412)268-6829 Scientific Visualization Specialist, Pittsburgh Supercomputing Center There are 10 types of people in the world: those who understand binary, and those who don't.
ccNodeData.cfg
Description: Binary data
ccNodeData.net
Description: Binary data
