I was looking online for some good weathering references, and stumbled 
across something even more exciting - subsurface scattering.

Subsurface scattering takes into account that many materials are slightly 
translucent. Marble allows some light to pass a very short distance, as 
does milk. Light hitting skin is not immediately reflected, but will 
penetrate a few layers, diffuse, and then travel towards the viewer. The 
examples I've seen are incredible. Such a small effect makes images 
massively more realistic.

I haven't read the paper yet, but a copy is available at:
http://www-graphics.stanford.edu/~henrik/papers/bssrdf/

and some more example images are at:
http://www-graphics.stanford.edu/~henrik/images/subsurf.html


Please, take a look! I am not familiar enough with the inner workings of 
POV to judge if this is implementable. If it is, it would be an incredible 
step forward!


-- Simon

Post a reply to this message

This has been advertised on this server many times before. Fascinating stuff
that I hope will someday make it into 4.0... BTW, I have created a wish list
of stuff I'd like in 4.0 - I'll append an editted-down version of it here.

BSSRDF:-
http://graphics.stanford.edu/papers/bssrdf/
http://graphics.stanford.edu/papers/scatteringeqns/
http://graphics.lcs.mit.edu/~dorsey/papers/stone/
Blinn Shading:-  see MegaPOV
Support for RenderMan Shaders:-  see POV-Man
Blurred reflections and transparency:-  see MegaPOV and Chris Huff's work
Weathered Textures:-
http://www.sciam.com/2000/0200issue/0200dorsey.html
http://graphics.lcs.mit.edu/~dorsey/weathering/
http://graphics.lcs.mit.edu/~dorsey/papers/patina
http://graphics.lcs.mit.edu/~dorsey/papers/wet_materials
http://graphics.lcs.mit.edu/~dorsey/papers/flow/
Smoke Simulation:-  http://graphics.stanford.edu/papers/smoke/
Liquids Simulation:-  http://graphics.stanford.edu/papers/water/water.pdf
Cloth Simulation:-  a generalized solution should be developed from the work
done so far
Tesselation:- see Warp's patch
Subdivision Surfaces:-  see VanSickle's work, and how to hard-code it
Non-linear transformations
Mesh-->Patch conversion:-
http://graphics.stanford.edu/papers/surfacefitting/
displacement-mapping:-  http://graphics.stanford.edu/papers/displace/
NURBS
Motion-blur:- see MegaPOV
Post-processing:-  teach POV to output extra data to be fed to a helper
utility
New focal blur solutions:-  http://www.flarg.com/bokeh.html
Better (?) radiosity:-
http://graphics.stanford.edu/papers/veach_thesis/
http://graphics.stanford.edu/papers/metro/
Efficient Rendering of Scenes with Many Objects (Like Plants):-
http://graphics.stanford.edu/papers/ecosys/
http://graphics.stanford.edu/papers/coherentrt/
Simple Bones / Inverse-Kinematics System

Post a reply to this message

I got copies of lots of papers with stuff that would be on our wish
lists.  One that interested me was the paper on cellular textures.  They
are made of little cells with various properties (scalar values and
vectors?) that allow you to make things like thorns and fur and to
simulate chemical properties.
Brendan

Post a reply to this message

[Andrea Ryan <ary### [at] global2000net>]
| One that interested me was the paper on cellular textures. 

This one?
http://citeseer.nj.nec.com/fleischer95cellular.html 

-- 
This message has not been scanned for viruses.

Post a reply to this message

Tony[B] <ben### [at] catholicorg> wrote:
: Blurred reflections and transparency:-  see MegaPOV and Chris Huff's work

  If I have understood it correctly, I think that the idea behind the decision
that they were not included in 3.5 is that they are too specific. By this
I mean that the goal for pov4 is to create some generic way of defining
surface properties which allows, among other things, to create blurred
reflection/refraction effects defined by the user. That is, you don't tell
povray to use blurred reflection with a keyword, but instead there's a more
generic way of doing it so that the end result is the same (but you can also
make much more with it).

: Non-linear transformations
: displacement-mapping:-  http://graphics.stanford.edu/papers/displace/

  I suppose these would require tesselation.

: Efficient Rendering of Scenes with Many Objects (Like Plants):-

  What would be more efficient than using meshes?

-- 
#macro N(D,I)#if(I<6)cylinder{M()#local D[I]=div(D[I],104);M().5,2pigment{
rgb M()}}N(D,(D[I]>99?I:I+1))#end#end#macro M()<mod(D[I],13)-6,mod(div(D[I
],13),8)-3,10>#end blob{N(array[6]{11117333955,
7382340,3358,3900569407,970,4254934330},0)}//                     - Warp -

Post a reply to this message

Warp wrote:

> : Efficient Rendering of Scenes with Many Objects (Like Plants):-
> 
>   What would be more efficient than using meshes?

Meshes do have thier limitations. If I create something in POV-Ray
using primitve objects and CSG, then want to make several copies
of them, I would like them to be resource friendly. Meshes have
the problem that to get high detail they also require high triangle
counts. Even with the mesh optimizations you are still working
with extremely large file sizes. A three triangle blade of grass
is effiecent but a 3 million triangle mesh is boarders on
ridiculous.

-- 
Ken Tyler

Post a reply to this message

Ken <tyl### [at] pacbellnet> wrote:
: A three triangle blade of grass
: is effiecent but a 3 million triangle mesh is boarders on
: ridiculous.

  Then you should use bicubic patches in order to create the triangles
on the fly... :)

-- 
#macro N(D,I)#if(I<6)cylinder{M()#local D[I]=div(D[I],104);M().5,2pigment{
rgb M()}}N(D,(D[I]>99?I:I+1))#end#end#macro M()<mod(D[I],13)-6,mod(div(D[I
],13),8)-3,10>#end blob{N(array[6]{11117333955,
7382340,3358,3900569407,970,4254934330},0)}//                     - Warp -

Post a reply to this message

>   If I have understood it correctly, I think that the idea behind the
decision
> that they were not included in 3.5 is that they are too specific. By this
> I mean that the goal for pov4 is to create some generic way of defining
> surface properties which allows, among other things, to create blurred
> reflection/refraction effects defined by the user.

Interesting. I have no idea how that can be, but if they do, so be it.

>   I suppose these would require tesselation.

Hence the reason I include it after tesselation in the list. :)

>   What would be more efficient than using meshes?

The paper discusses optimization algorithms to reduce triangle counts /
memory consumption for large meshes, I think. At least that's what skimming
it very quickly revealed to me. :)

Post a reply to this message

I found out about cellular textures in a pdf file I found in a different
place.  I forget where I got it.

Cellular Texture Generation
Kurt W. Fleischer
David H. Laidlaw
Bena L. Currin
Alan H. Barr
California Institute of Technology
Pasadena, CA 91125

Brendan (no, I didn't write any of the paper :-)

Post a reply to this message

Here's links to stuff that could be on our wish lists.  I found the
paper on cellular texturing by following links on these pages.
Brendan

http://graphics.Stanford.EDU/~henrik/
http://www.cs.cmu.edu/~ph/

http://astronomy.swin.edu.au/pbourke/
(no wish list stuff, but lots of povray stuff such as renderings of
Mars)

Brendan

Post a reply to this message