[Bf-committers] Boolean op holes error handled

"Juan J. Peña M." jjulio.pena at verizon.net.do
Thu Dec 1 13:05:33 CET 2005


Arne Schmitz escribió:

>In other words: With a plane you would define the direction in which the 
>normal is facing "outside", and the other side "inside". You would then treat 
>the plane es being infinite. This way the object is "closed", it has no 
>boundary.
>
>Arne
>  
>
Yup, but carefull with the "infinite" thing. Imagine this scenerio:

You just modeled an human feet and it is an open mesh (let´s say you 
just modeled it untill the ankle).  You also have this beatiful CG beach 
scene modeled. Now you want to add footprints in the sand to have 
realistic water simulation later on. If every face on the feet would be 
infinitely extruded on it´s normal direction, the resulting new extruded 
faces will trespass the feet boundary and create some mutan shape.. like 
a sea urchin or the like.

Could it be coded so that the normals marks wich side of a face is 
"inside", but also to account with the collision with another faces?  
The feet/sea urchin situation would also happen on any "open mesh" that 
describes a U shape at some point.. "infinite extrudes" from each side 
of the "U" will eventually extend beyond the oposite "U" side.

Apollux.






More information about the Bf-committers mailing list