I’ve been working on figuring collisions. While I do know how to get collisions to work on already visible meshes, I would like to be able to make objects which while visually complex have simpler collision geometry which is not rendered. So more or less it’s something that moves along with the “beauty” model(or perhaps the other way round), but collisions are handled by the invisible collider in stead.
I’ve tried some searches, but I’m not sure if I’m looking for the right things, since the closest terminology I can think of is “hitbox” but I haven’t turned up much using that or any other terminology I can think of.
More or less I’m looking for a tutorial/help to get me going on creating a mesh with an invisible collision mesh outside of it(working in blender and using chicken exporter). So do I model out the beauty mesh as well as the collider in blender then export? If so, are there some tags I attach to the collision mesh which tell it to be invisible/have collisions? or do I simply apply an all alpha texture to it to make it invisible?
Or am I just barking up the wrong tree and are collision solids the only avenue for invisible collider meshes in Panda?
You can also use:
barrier
the manual says it has the same effect as { Polyset descend }, but I’ve never tried it myself. panda3d.org/manual/index.php/Egg_Syntax
thanks for the responses, I am curious though, when using collision solids, it seems that the only way to tell it where to go and what orientation to have is through coordinates, like on the tube which requires two sets of end coordinates. However what if I’m setting up collisions for a character and want to use tubes for the different parts of the limbs (one for forearm, one for bicep, etc.), how do I line up the tube with each part of the body without figuring out the end points?
Also, I’ve found out that I can create a model in blender, then create a separate simpler mesh in the same scene, set the simpler mesh to be collideable but not visible and export the two into a single egg file, and when attempting to check for collisions, panda will only check the simpler yet invisible geometry to see if there was a collision, is this correct?
If you want to use the collision solid for an avatar with animations then the only way I know is to expose the joints and attach the solids to the bones (check out “looking & gripping” sample). If you make a animated mesh and try to use it for collisions it won’t actually be animated.