Disappearing Skybox (again)

Hi people



I've just started out with JME, so it might well be that this is not a problem of JME, but of my code… :slight_smile:



I've noticed unusual behaviour of the bounding volumes in the SkyboxTest application. I've modified the code a bit; I've removed the torus and the cube and made the sphere smaller. The latter is exactly what's causing the problem.



When I start up the example, everything is rendering fine. But if I "fly" straight through the sphere, the skybox suddenly disappears. After this, the clipping of the skybox is linked with that of the sphere, which seems like the common problem of "forgetting" to set a bounding volume on an object. Now, as the skybox is a node and not a geometry, I can't set a bounding volume here.



Also, when displaying the bounding volumes (using the B-Key), one can see a change of the (presumably) root node's bounding volume when flying through the sphere.



I'm not sure if this is a bug, but how am I supposed to handle that? Why does


m_skybox.setCullMode(Spatial.CULL_NEVER);



have no effect on the clipping of the skybox?

Another effect I've noticed is that, when you remove

s.setRenderQueueMode(Renderer.QUEUE_OPAQUE);



the sphere isn't rendered when it's not very near to the camera. Seems like the skybox does write something into Z-Buffer?

Anyhow, thanks for your time and keep up the good work!

cheers,
Bernhard

Here's the source of the modifies SkyboxTest.java:

/*
 * Copyright (c) 2003-2005 jMonkeyEngine
 * All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are
 * met:
 *
 * * Redistributions of source code must retain the above copyright
 *

You indeed discovered a bug: When using a bounding sphere instead of the bounding box for the sphere the boundings were computed incorrectly.



Thanks for reporting!



Fixed  :slight_smile:

Is this fixed with Mojo's recent check-in?

Well, I think I fixed it with the changes to BoundingSphere "Minor, topic 2651: fixed bounding bug with replaced center vector (changing value now)". Not sure which checkin from Mojo you mean as he worked on shared mesh and tuts, only.

Ah, I saw a few from mojo and grouped it mentally with those when moving the emails to my jme mail folder. :)  So anyhow, can we confirm this issue is closed then?