In article <3ef6bed8$1@news.povray.org>,
 "Matthias Hullin" <new### [at] hullinnet> wrote:

> The upper series consists of exponential blobs, which I generated using the
> isosurface object. To me, it looks much smoother, kind of "more organic"
> than the linear blobs shown below.

The blobs in POV-Ray are not linear. Linear functions do not "blob", I'm 
not really sure what you meant. The shading problems are due to 
discontinuities in the second derivative of the function. Also, it's not 
clear what function you are using...but if the function is not bounded, 
each new component will affect the entire blob, making computation 
slower and the entire blob less controllable.


> What about adding an "exponential" keyword to the blob object? All it would
> have to do is toconvert the objects in a suitable function and then
> calculate the isosurface.

The root solver used by the blob object is very special-purpose. Using a 
more general solver similar to that used by isosurfaces makes many more 
primitives possible, but it would be better to make an entirely new 
object to take advantage of that flexibility: a "superblob" or "blob2" 
primitive.

http://news.povray.org/povray.advanced-users/28049/
http://news.povray.org/povray.binaries.images/28102/
http://news.povray.org/povray.binaries.images/28197/

I now have a working version of the shape patch I described in that last 
one. (though the pattern is currently non-functional) It currently 
supports spheres, boxes, and toruses, but I'm planning to eventually 
implement cylinders, cones, superellipsoids, shells, tubes, and 
function-based primitives, as well as the ability to define the falloff 
curve with a function. The solver is a custom one I wrote from scratch, 
nothing based on the isosurface primitive code. If anyone wants more 
details, I'll start a thread in *.programming.

-- 
Christopher James Huff <cja### [at] earthlinknet>
http://home.earthlink.net/~cjameshuff/
POV-Ray TAG: chr### [at] tagpovrayorg
http://tag.povray.org/

Post a reply to this message