Gilles Tran wrote:
> 
> Well, it happens that I use brightness 2 in scenes where I have direct
> photographic references (I just cut and pasted the settings here), so the
> choice is only dictated by the similarity between the "real" scene and its
> rendered version (I'm not talking about the Cornell box here, but of complex,
> real-life pictures). These images also require extremely bright lights.
> But you're right,  brightness level depends on the scene. My "Contrast" image
> used 1.2 for instance (but it was not based upon a photography). I don't
> remember using values lower than 1. It's more an aesthetic choice than a
> technical one IMHO.
> 

It could be that Indoor scenes often work quite well with increased
brightness, it probably is also mainly a questions what settings the
objects' colors and finishes are tweaked for.  NTL, could you point me to
some pictures using brightness 2?

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

In article <3AC### [at] inapginrafr>, tra### [at] inapginrafr says...

> - have only one light source
Yes, I switched the lamps off.
Only the sunlight remains (arealight, 7x7)

> - all objects have ambient = 0
Done that. 
> - have few reflections
Wel... hmmm... the floor, the sculptures, the window-glass and the water 
are reflective.

> - have few (or none !) objects with ior
The windows glass...

> - colors are normal (below rgb 1)
All the white objects have colors of 1.5-2. Does that slow down 
rendering? With brightness 0.3 the scene look Ok.

Post a reply to this message

In article <3AC4C65C.5F516FD9@gmx.de>, chr### [at] gmxde says...
> As Gilles said, all finishes need to be changed, if you want to avoid
> adapting the colors, you can try to play with the diffuse finish
> component. 
Does the brightness have an influence on rendering speed?

> Be sure to turn off all additional light sources (just leave the sun
> light).
Of course i've done that. But still 0 PPS

Lutz-Peter

Post a reply to this message

Christoph Hormann wrote:

> It could be that Indoor scenes often work quite well with increased
> brightness, it probably is also mainly a questions what settings the
> objects' colors and finishes are tweaked for.

Probably. There's always a lot of tweaking involved in radiosity anyway.

> NTL, could you point me to some pictures using brightness 2?

One at least, uploaded there on a temporary basis (others will come... someday
;-)).
http://www.oyonale.com/temp/classroom.jpg

G.

--

**********************
http://www.oyonale.com
**********************
Graphic experiments
Pov-ray gallery

Post a reply to this message

Lutz-Peter Hooge wrote:

> > - have few (or none !) objects with ior
> The windows glass...

This could be the main culprit. I have found that a few (and even small)
objects with ior used with radiosity could make a computer weep. I don't
know about a whole pane...

Lots of reflective objects don't help either, but to a lesser extent. The
other problem with using reflective objects with radiosity is the apparition
of ugly black spots. The spot remover is a higher max_trace_level (not the
MTL of 15), which in turns causer slower rendering.

> All the white objects have colors of 1.5-2. Does that slow down
> rendering?

Not that I know.
Personally, I prefer harsh lights with "normal" colors than the other way
round. More a habit than something rational though and I use higher than 1
colors when needed.

G.

--

**********************
http://www.oyonale.com
**********************
Graphic experiments
Pov-ray gallery

Post a reply to this message

Gilles Tran wrote:

> The spot remover is a higher max_trace_level (not the
> MTL of 15), which in turns causer slower rendering.

"*** note *** the MTL of 15", silly me.

G.
--

**********************
http://www.oyonale.com
**********************
Graphic experiments
Pov-ray gallery

Post a reply to this message

In article <3AC4CFC9.C901DE8F@inapg.inra.fr>, tra### [at] inapginrafr says...

> This could be the main culprit. I have found that a few (and even small)
> objects with ior used with radiosity could make a computer weep. I don't
> know about a whole pane...
If I just outcomment the "interior" part of the glass texture I get 1 
PPS. But still with low quality radiosity...
If I disable the whole glass stuff -> ~12 PPS
with better quality its all 0 pps.
I fear there are simply too much slow things in this scene for radiosity.

Lutz-Peter

Post a reply to this message

Gilles Tran wrote:
> 
> One at least, uploaded there on a temporary basis (others will come... someday
> ;-)).
> http://www.oyonale.com/temp/classroom.jpg
> 

Thanks, it looks quite realistic, the only thing i find a bit strange is
the missing bluish touch near the window, that would result from a blue
sky outside.  If you add a blue sky for getting that, this could strongly
influence the whole lighting situation, even more because of the increased
brightness.  

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

Lutz-Peter Hooge wrote:
> 
> If I just outcomment the "interior" part of the glass texture I get 1
> PPS. But still with low quality radiosity...
> If I disable the whole glass stuff -> ~12 PPS
> with better quality its all 0 pps.
> I fear there are simply too much slow things in this scene for radiosity.
> 

That comes near what i would expect, maybe try default radiosity settings
except recursion_limit 1, that would be faster than the values Gilles
suggested, but of course also lower quality.  

'brightness' should not have any influence on speed BTW.

Furthermore you could try to optimize things, if the background mountains
are an isosurface, try a heightfield instead, if the water is an
isosurface, use only a normal statement.  

And of course patience :-), 1 digit PPS rates or even 0 PPS are nothing
seldom when working with radiosity.  

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

Yet another wonderful image, but does her left arm intersect with her chest?

Kev
--
http://perso.libertysurf.co.uk/kevin.ellis

Post a reply to this message