#2381 closed defect (fixed)
lighting/shading for vector objects backward in nviz
Reported by: | cmbarton | Owned by: | |
---|---|---|---|
Priority: | normal | Milestone: | 7.2.0 |
Component: | wxGUI | Version: | svn-trunk |
Keywords: | wxnviz | Cc: | |
CPU: | Unspecified | Platform: | MacOSX |
Description
I just tried to display a map with an overlay of vector points and they all looked black, regardless of the color I chose. Turns out that when the view is from the SE and the light is also from the SE (both defaults), the vector spheres are in shadow. If I switch the light to the NW, the face of the spheres are lit--but of course the shadows on the surface are going the other way. This reversal also applies to light height. When the light is low, a sphere is lit from the top; when the light is high, there is very oblique lighting--again reversed from the surface. So there is a flipped variable somewhere.
I've attached a screenshot of a surface and points with a view from the SE and lighting from the SE. The shadows on the surface are to the NW (correct), but the spheres in completely in shadow (incorrect).
Attachments (1)
Change History (11)
by , 10 years ago
Attachment: | nviz_backward_vector_lighting.jpg added |
---|
comment:1 by , 10 years ago
Component: | Tcl/Tk NVIZ → wxGUI |
---|---|
Keywords: | wxnviz added |
I don't have such problem. Are you looking at the surface from top? No idea what could cause this...
comment:2 by , 10 years ago
Did you look at the screenshot? I tried this again with the nc_08 data, to make sure it was not something in my files, and have the same results (displaying elevation and firestations). The telling thing is to make the sphere's fairly large (e.g., 500) and watch the light model as you move the puck in the appearance pane.
For example, set the view to the west (observer puck in the east). Switch to the appearance pane and move the lighting puck from east to west. The "light model" sphere will go from light to dark. But the vector point spheres will go from dark to light. Height is harder to figure out because it is also affected by the direction of the light. Height may in fact be OK once the horizontal direction is reversed.
comment:3 by , 10 years ago
I can see it now but only on Mac. I feel like I've seen it before but I don't remember. Do you have any idea when was the last time it worked?
follow-up: 5 comment:4 by , 10 years ago
It works fine with GRASS 7.0 beta 1, compiled 3 April 2014.
comment:5 by , 10 years ago
Replying to cmbarton:
It works fine with GRASS 7.0 beta 1, compiled 3 April 2014.
It is reversed on GRASS 7.0 beta 2 svn, compiled 14 July.
follow-up: 8 comment:6 by , 10 years ago
Platform: | Unspecified → MacOSX |
---|
That's helpful to know. I assume the wxPython version is the same. There were a few changes in nviz gui done by hcho 4 weeks ago so we could start investigating there. BTW, there are no problems on Windows.
comment:7 by , 10 years ago
I made no change in wxPython versions (that I remember) over that short time frame.
comment:8 by , 10 years ago
Resolution: | → fixed |
---|---|
Status: | new → closed |
screenshot of surface and points in nviz showing lighting