Hi Sun:
 
I'm going to need your assistance, as I suspect I am having trouble with a lighting bug.  I've tried to work around the problem, but I'm unable to.  Attached is simplified source code using only Sun classes that shows the problem.
 
Thanks,
Geoff.
 
 
 
PS. this email follows on from one I send to the interest group a couple of days ago (sorry that I couldn't easily continue the thread ... I deleted the message).
 
Original message (23/04/00):
 
I'm having a lighting problem when shapes butt up against each other.  Attached is an extreme example of the problem where there are 100 cylinder disks joined.
 
The black dots don't necessarily appear on a vertex, but can appear randomly at the intersection of two shapes.  It doesn't help to overlap the shapes.  I suspect that as the vertex normals at the joins are pointing in three different directions, this is my problem.  I've tried using Sun's NormalGenerator utility (instead of our own) but it's no help. 
We're using a directional light, and in this case these material properties are set:
    material.setEmissiveColor(0.0f, 0.0f, 0.0f);
    material.setSpecularColor(1.0f, 1.0f, 1.0f);
    material.setAmbientColor(1.0f, 0.0f, 0.0f);
    material.setDiffuseColor(1.0f, 0.0f, 0.0f);
    material.setShininess(64.0f);
 
Has anyone seen a similar problem, and were able to solve it, or perhaps can offer an explanation to help me work around it.
 
Thanks for any assistance.
 

Java3DTest.java

Reply via email to