Hi everybody,

I have the problem that photons are nearly invisible on surfaces with 
normals. It may be a bug, maybe it is realistic, but it is very annoying, 
because surfaces without normals usually don't look good (unless they are 
Isos with much "true" detail, but I didn't test yet if those work better 
with photons). 

Does anybody know a way around this problem? (an bright, small spotlight  
aimed at the reflecting/refracting object might do the trick in some 
cases, but it has the disadvantage that every additional lightsource 
slows down glass-like objects significantly)

Lutz-Peter

Post a reply to this message

Lutz-Peter Hooge wrote:
> 
> Hi everybody,
> 
> I have the problem that photons are nearly invisible on surfaces with
> normals. It may be a bug, maybe it is realistic, but it is very annoying,
> because surfaces without normals usually don't look good (unless they are
> Isos with much "true" detail, but I didn't test yet if those work better
> with photons).
> 

I experienced the same, see for example the balcony demo scene in Povray
3.5.  If you remove the normal of the cloth, the caustics on the table are
much stronger.  

I'm not sure if this should be considered as a bug, a test with normals
and heightfield/isosurface alternative would clarify things.

Christoph

-- 
Christoph Hormann <chr### [at] gmxde>
IsoWood include, radiosity tutorial, TransSkin and other 
things on: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

Christoph Hormann wrote:
> 
> Lutz-Peter Hooge wrote:
> >
> > Hi everybody,
> >
> > I have the problem that photons are nearly invisible on surfaces with
> > normals. It may be a bug, maybe it is realistic, but it is very annoying,
> > because surfaces without normals usually don't look good (unless they are
> > Isos with much "true" detail, but I didn't test yet if those work better
> > with photons).
> >
> 
> I experienced the same, see for example the balcony demo scene in Povray
> 3.5.  If you remove the normal of the cloth, the caustics on the table are
> much stronger.
> 
> I'm not sure if this should be considered as a bug, a test with normals
> and heightfield/isosurface alternative would clarify things.

I wouldn't necessarily classify this as a bug. It is more a limitation
with the implementation of the normal perturbation method that POV-Ray
uses.

-- 
Ken Tyler

Post a reply to this message

In article <3BDAF7DB.20FED1F5@pacbell.net>, tyl### [at] pacbellnet says...

> I wouldn't necessarily classify this as a bug. It is more a limitation
> with the implementation of the normal perturbation method that POV-Ray
> uses.

Are you sure?
I just testet it now with an Isosurface and a heightfield (without a 
normal map of course).
The results are the same as with a normal.

However, this is not realistic. I tried it with a glass of water and 
placed different objects in the caustics. Leather, paper, and cloth.
In all cases the caustics remained very visible, they relative brightness 
to the rest of the object didn't chage significantly.

Lutz-Peter

Post a reply to this message

In article <MPG.1645210baa7a8ee99896a6@news.povray.org>, lpv### [at] gmxde 
says...

> I just testet it now with an Isosurface and a heightfield (without a 
> normal map of course).

I postet the test to the images.binaries group. subject "photons & rough 
surfaces"

Post a reply to this message

Lutz-Peter Hooge wrote:
> 
> However, this is not realistic. I tried it with a glass of water and
> placed different objects in the caustics. Leather, paper, and cloth.
> In all cases the caustics remained very visible, they relative brightness
> to the rest of the object didn't chage significantly.
> 

I agree and although i don't know what the reason is, there really seems
to be something going quite wrong.  Seem my post in p.b.i.

Christoph

-- 
Christoph Hormann <chr### [at] gmxde>
IsoWood include, radiosity tutorial, TransSkin and other 
things on: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

Christoph Hormann wrote:
>
> I agree and although i don't know what the reason is, there really seems
> to be something going quite wrong.  Seem my post in p.b.i.

The edge problems disappear when more photons are used but
the shading on the sphere is not right. Problem in normal
scaling?

 
_____________
Kari Kivisalo

Post a reply to this message

Kari Kivisalo wrote:
> 
> The edge problems disappear when more photons are used but
> the shading on the sphere is not right. Problem in normal
> scaling?
> 

To me it seems to have something to do with the surface curvature, even
with more photons, the edges of the box stay dark (although the dark parts
are thinner)  Of course this could be the side effect of something totally
different.

Christoph

-- 
Christoph Hormann <chr### [at] gmxde>
IsoWood include, radiosity tutorial, TransSkin and other 
things on: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message