On Thu, 15 May 2003 17:13:37 +0200, "Hugo Asm" <hua### [at] post3teledk> wrote:
> But my
> isosurface mostly does surface displacement on a rounded box, so maybe I
> should try another approch: To import a dense mesh of a box, and apply my
> functions to it, then save the result.

(considering that rounded box macro has "linear" behaviour like in rounded
version of isocsg library)

You have notice that it will not be the same. When you have function of rounded
box and then you apply function of deformation then you receive new surface made
from points <x,y,z> where:

f_rbox(x,y,z) - f_def(x,y,z) = threshold

But if you would first create mesh for box and then displace about value of
f_def at box surface then result may be at different place.

I would do it going along normal of rounded box direction until it could reach
condition listed above. You can use optimization similiar to intersecting
isosurface surface but it would need knowledge written in isosurf.cpp :-)

ABX

Post a reply to this message