| Re: Matching volumes and surfaces [message #47086] |
Fri, 20 January 2006 07:04 |
K. Bowman
Messages: 330
Registered: May 2000
|
Senior Member |
|
|
In article <k-bowman-4168BE.20523319012006@news.tamu.edu>,
"Kenneth P. Bowman" <k-bowman@removethis.tamu.edu> wrote:
I follow up my previous posting with a little more info on my isosurface
rendering problem.
I realized that you can show the mesh used to create the isosurface. This gives
some hints as to the possible problem. Turning on the mesh and rotating the
view gives this
http://csrp.tamu.edu/hiaper/archive/render/rendering2.jpg
The yellow square is the output of iSurface. The gray square and mesh is the
output of iVolume with isosurface. Because of the way the isosurface slices
diagonally through the volume, the mesh has long narrow triangles at the grid
boundaries of the original 3-D grid. I think this is correct.
This view, however, shows that the location of the mesh boundaries drifts with
respect to the grid coordinates (~0.9, 1.8, ...) I think the mesh lines should
coincide with the grid coordinates (~1, 2, ...).
This strongly suggests to me that there is an error in the rendering or mesh
calculation, with the right and top edges of the gray isosurface ending at 9.0
rather than 10.0 (dividing the data range by n instead of n-1 somewhere).
If I re-run the program with VOLUME_DIMENSIONS = [11, 11, 11], I see this
http://csrp.tamu.edu/hiaper/archive/render/rendering3.jpg
Now the isosurface and the mesh appear to be aligned properly with the grid,
although it now thinks the volume extends to 11 in x and y (indicated by the
light gray shading of the volume walls.
So, I think there is something wrong with the rendering or mesh coordinate
calculation. Can anyone show me the error of my ways?
Ken
P.S. At least this may provide a workaround, but I still need to be certain
that my data is displayed in the correct place in the data coordinates.
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
