Hugo Asm <hua### [at] post3teledk> wrote:
> radiosity. This is the same term as global illumination.

  It's better to use term "global illumination" (and if you want to be
more specific, add "stochastic" before it) than "radiosity" when writing
to not-only-povray groups to not to cause confusion.
  "Radiosity" is the name of a specific algorithm (which has nothing to do
with the algorithm used by POV-Ray) and the term is usually used in this
meaning.

-- 
plane{-x+y,-1pigment{bozo color_map{[0rgb x][1rgb x+y]}turbulence 1}}
sphere{0,2pigment{rgbt 1}interior{media{emission 1density{spherical
density_map{[0rgb 0][.5rgb<1,.5>][1rgb 1]}turbulence.9}}}scale
<1,1,3>hollow}text{ttf"timrom""Warp".1,0translate<-1,-.1,2>}//  - Warp -

Post a reply to this message

"Warp" <war### [at] tagpovrayorg> wrote:
>   It's better to use term "global illumination" (and if you want to be
> more specific, add "stochastic" before it) than "radiosity" when writing
> to not-only-povray groups to not to cause confusion.
>   "Radiosity" is the name of a specific algorithm (which has nothing to do
> with the algorithm used by POV-Ray) and the term is usually used in this
> meaning.

So, to sum it up, radiosity is a subset of global illumination, i.e., one of
the method or algorithm used in global illumination.  The context of
radiosity in POV-Ray means achieving the same effect but not implemented
using that particular "Radiosity" algorithm.  If that is the case, any
reason for the difference in implementation?

Daniel.

Post a reply to this message

Warp wrote:
> It's better to use term "global illumination" (and if you want to be
> more specific, add "stochastic" before it) than "radiosity" when
> writing to not-only-povray groups to not to cause confusion.

Thanks Warp. I rarely go outside the POV-Ray forums to discuss these things,
so I didn't know. When you say, the algoritm used in POV-Ray isn't typical
radiosity, I assume it was modified. What happened to the "backwards monte
carlo" ?

Daniel wrote:
> any reason for the difference in implementation?

I think this has to do with how POV-Ray handle objects. Most raytracers work
on triangles, POV-Ray does not only do this.

Regards,
Hugo

Post a reply to this message

On Sat, 24 May 2003 21:15:56 +1200, "Daniel Ngu" <ngu### [at] xtraconz> wrote:


>What modeler would you recommend for POVRay?  I strived to do everything
>using a text editor as I reckon I will have better control over the scene
>but is it really feasible to come up with higly complex scene without the
>help of a modeler.
Try Moray, you would think that it had been made for Povray :-} 
One advantage is that it shares the same newsgroup and everyone is friendly. I
don't think I've seen a RTFM comment. It is not free but IMO cheap at the price.
http://www.stmuc.com/moray/index.html


Regards
        Stephen

Post a reply to this message

Daniel Ngu <ngu### [at] xtraconz> wrote:
> If that is the case, any reason for the difference in implementation?

  The radiosity algorithm works only with polygons. Also, it can't take
into account some things that raytracing supports, such as refraction
and reflection (unless some kind of "fake" reflection is used, as is
typical in scanline rendering).

-- 
#macro M(A,N,D,L)plane{-z,-9pigment{mandel L*9translate N color_map{[0rgb x]
[1rgb 9]}scale<D,D*3D>*1e3}rotate y*A*8}#end M(-3<1.206434.28623>70,7)M(
-1<.7438.1795>1,20)M(1<.77595.13699>30,20)M(3<.75923.07145>80,99)// - Warp -

Post a reply to this message

Hugo Asm <hua### [at] post3teledk> wrote:
> Thanks Warp. I rarely go outside the POV-Ray forums to discuss these things,
> so I didn't know. When you say, the algoritm used in POV-Ray isn't typical
> radiosity, I assume it was modified. What happened to the "backwards monte
> carlo" ?

  Radiosity and monte-carlo sampling are just two completely different
algorithms for implementing global illumination. POV-Ray uses the latter
(because it's more suitable to raytracing of non-polygonal surfaces).

-- 
plane{-x+y,-1pigment{bozo color_map{[0rgb x][1rgb x+y]}turbulence 1}}
sphere{0,2pigment{rgbt 1}interior{media{emission 1density{spherical
density_map{[0rgb 0][.5rgb<1,.5>][1rgb 1]}turbulence.9}}}scale
<1,1,3>hollow}text{ttf"timrom""Warp".1,0translate<-1,-.1,2>}//  - Warp -

Post a reply to this message

Daniel Ngu <ngu### [at] xtraconz> wrote in message
news:3ecf37b3@news.povray.org...

> What modeler would you recommend for POVRay?  I strived to do everything
> using a text editor as I reckon I will have better control over the scene
> but is it really feasible to come up with higly complex scene without the
> help of a modeler.
>
> Daniel.

There are many different methods of modelling, some are better for certain
things.
You can try Hamapatch, ( http://www.geocities.com/hamapatch/program/ ) a
good little spline modeller for a lot of things, although some kinds of
highly detailed modeling can get very tedious in it.
Wings3d ( http://www.wings3d.com/ ) is a good subdivision modeller which is
capable of very detailed objects.
Also Blender has a lot of excellent tools as Andreas mentioned, although be
forewarned its interface is not very intuitive...  (It seems to be one of
those interfaces certain types love altogether, while everyone else just
shakes their head. Unfortunately, I'm the head shaking type... I find the
program completely unusable :-(  But it does have a lot of good things in
one package.)

But never discount POV's SDL, I only started using POV last August (after a
couple years of using Bryce & Hamapatch) and already find I'm using SDL most
of the time and only using external modellers for a few specific things
(People mostly, for which I use Poser although I'm trying to learn to model
characters in Wings3d.)

Enjoy! RG

Post a reply to this message

On Sun, 25 May 2003 06:32:28 +1200, "Daniel Ngu" <ngu### [at] xtraconz>
wrote:

>So, to sum it up, radiosity is a subset of global illumination, i.e., one of
>the method or algorithm used in global illumination.

Precisely.

>The context of radiosity in POV-Ray means achieving the same effect but
>not implemented using that particular "Radiosity" algorithm.

Correct again.

>If that is the case, any reason for the difference in implementation?

Theory of operation.

In most 3D packages, objects are represented by meshes of triangles,
polygons or NURBS (which are further subdivided into such meshes or
rendered on-the-fly). In POV-Ray and most other pure raytracers,
objects are internally represented by their mathematical description,
for example a spere is not just a million triangles but rather "the
surface defined as all points equidistand from the center".

Radiosity is a means of simulating global illumination in
polygon-based renderers. The principle is very basic. Normally, the
color of a pixel is determined by several things:

1) The texture of the triangle at that point
2) The amount of shadow at that point as determined by the shadow map,
if any
3) The amount of transparency, if any
4) The environment map, if any
5) The bump map, if any

etc.

Radiosity works by creating an additional map which is "painted" over
the triangles, the so-called "light map". It's similar to any other
map being closest to the shadow map.

The way it works is the following. First, every triangle is, in
layman's terms, turned into a camera, and what this triangle "sees" in
terms of brightness is stored onto it as its light map. When all
triangles have been traversed, the complete scene's light map has been
built. If further accuracy is desired, the process is repeated, but
this time the already accumulated light map is taken into account,
i.e. you can get surfaces lit indirectly from already indirectly lit
surfaces (this is called a recursion level and this is level 2). Etc.
etc. until you reach the desired level of accuracy.

Obviously, this method can not work with rendereds where objects are
simply not polygons. Therefore POV implements the so-called Stochastic
Global Illumination, which you can look up in the docs under
"Radiosity". 

I hope this is making the picture a bit more clear.


Peter Popov ICQ : 15002700
Personal e-mail : pet### [at] vipbg
TAG      e-mail : pet### [at] tagpovrayorg

Post a reply to this message

In article <3ecfba2a@news.povray.org>, "Daniel Ngu" <ngu### [at] xtraconz> 
wrote:

> So, to sum it up, radiosity is a subset of global illumination, i.e., one of
> the method or algorithm used in global illumination.  The context of
> radiosity in POV-Ray means achieving the same effect but not implemented
> using that particular "Radiosity" algorithm.  If that is the case, any
> reason for the difference in implementation?

More precisely, radiosity refers to global illumination. However, over 
time it has been associated with a specific algorithm based on using 
polygonal patches, so just saying "radiosity" can result in confusion. 
POV radiosity is radiosity, but is not what most people will think of 
when you use the word.

-- 
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

In article <3ecfc513@news.povray.org>, "Hugo Asm" <hua### [at] post3teledk> 
wrote:

> > It's better to use term "global illumination" (and if you want to be
> > more specific, add "stochastic" before it) than "radiosity" when
> > writing to not-only-povray groups to not to cause confusion.
> 
> Thanks Warp. I rarely go outside the POV-Ray forums to discuss these things,
> so I didn't know. When you say, the algoritm used in POV-Ray isn't typical
> radiosity, I assume it was modified. What happened to the "backwards monte
> carlo" ?

"Monte carlo" in this sense refers using multiple randomized samples to 
obtain a result. As for the "backwards", POV radiosity still traces 
light through the scene back to its source as in ordinary backwards 
raytracing, it just follows the indirect reflection routes. Forward 
raytracing would follow the light in the actual direction it travels, 
photon mapping is an example of this. Photon mapping can also be used 
for global illumination, though this isn't implemented in POV.

-- 
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