In another thread there was a discussion about whether a higher recursion
level would make the image look better in the "illuminating hemisphere"
scene. It was also questioned, whether it is possible to speed up the
radiosity calculations using the aid of area lights without changing too
much the resulting image.

  Well, I decided to try.

  The first image uses radiosity recursion level 1.
  The second image is the exact same scene file, with the only difference
being the recursion level changed to 2.

  The second image took about 1 hour and 5 minutes to render. I tried to
help the radiosity with area lights, so I created a hemicube of light
using 5 area lights and then reduced the radiosity quality settings to
speed it up.
  The resulting image is the third one. It took about 40 minutes to render.

Post a reply to this message

Attachments:
Download 'rtest1.jpg' (27 KB) Download 'rtest2.jpg' (27 KB) Download 'rtest3.jpg' (26 KB)

Preview of image 'rtest1.jpg'

Preview of image 'rtest2.jpg'

Preview of image 'rtest3.jpg'

Seems to prove at least one thing to me, that light sources are necessary to
get a good render when using radiosity.  The white floor plane however has
lost most of the shadows but that must be from the full coverage of
lighting.
The contribution of illumination from the objects themselves in the 2nd
rendering is rather artificial looking but otherwise shows that a recursion
of 2 really ups the radiosity.  Which leads me to ask, did you do a
recursion of only 1 with the lighting?

Bob

Post a reply to this message

Bob Hughes <per### [at] aolcom?subject=pov-news:> wrote:
: The contribution of illumination from the objects themselves in the 2nd
: rendering is rather artificial looking but otherwise shows that a recursion
: of 2 really ups the radiosity.  Which leads me to ask, did you do a
: recursion of only 1 with the lighting?

  Of course not. The whole idea with the third image was to compare it to
the second one, so it uses recursion level 2 as well. However, I had to
reduce the diffuse value of the objects to get the same brightness, but
that's not a surprise.

  By the way, the shadows around the objects in the recursion level 1 image
may look cool, but I think they don't correspond to reality and are in fact
an artifact caused by the unrealistically low recursion level.
  Of course sometimes an unrealistic effect can look artistically better
than the physically accurate result.

-- 
main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
):_;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/

Post a reply to this message

Where does the red on the *inside* of the box come from?


-- 

* Doctor Jekyll had something * mailto:ber### [at] inamecom
* to Hyde...                  * http://www.enst.fr/~jberger
*******************************

Post a reply to this message

: 	Where does the red on the *inside* of the box come from?

  From the pigment.

-- 
main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
):_;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/

Post a reply to this message

Warp wrote:
> 

> :       Where does the red on the *inside* of the box come from?
> 
>   From the pigment.
> 
	Euh, what pigment? It seems to be a uniform light blue... or did you
give it a more complex pigment?


-- 

* Doctor Jekyll had something * mailto:ber### [at] inamecom
* to Hyde...                  * http://www.enst.fr/~jberger
*******************************

Post a reply to this message

>Warp wrote:
>> 

>> :       Where does the red on the *inside* of the box come from?
>> 
>>   From the pigment.
>> 
>	Euh, what pigment? It seems to be a uniform light blue... or did you
>give it a more complex pigment?

I also see that in the second and third images, the ceilling looks red. 
-- 
Cheers
Steve              email mailto:ste### [at] zeroppsuklinuxnet

%HAV-A-NICEDAY Error not enough coffee  0 pps. 

web http://www.zeropps.uklinux.net/

or  http://start.at/zero-pps

 11:06pm  up 11 days,  3:24,  2 users,  load average: 1.09, 1.09, 1.05

Post a reply to this message

Looks orange to me...

Post a reply to this message

Warp wrote:
> 
>   In another thread there was a discussion about whether a higher recursion
> level would make the image look better in the "illuminating hemisphere"
> scene. It was also questioned, whether it is possible to speed up the
> radiosity calculations using the aid of area lights without changing too
> much the resulting image.
> 
>   Well, I decided to try.
> 
>   The first image uses radiosity recursion level 1.
>   The second image is the exact same scene file, with the only difference
> being the recursion level changed to 2.
> 
>   The second image took about 1 hour and 5 minutes to render. I tried to
> help the radiosity with area lights, so I created a hemicube of light
> using 5 area lights and then reduced the radiosity quality settings to
> speed it up.
>   The resulting image is the third one. It took about 40 minutes to render.

How long did that first one take to render, Warp.. forty minutes, with
(probably) the default settings? Probably more like two or three. Gee,
seems like a hell of an optimization to me, especially given your
apparent attitude that image quality is barely an issue next to faster
output. How is it that you made such a grevious oversight?

And why didn't you bother to light it properly? Why did you even render
it at all? What are we supposed to glean from this image, other than the
fact that you're apparently using a nightlight to illuminate the scene?

-Xplo

Post a reply to this message

Steve <ste### [at] zeroppsuklinuxnet> wrote:
: I also see that in the second and third images, the ceilling looks red. 

  Because that's the pigment of the ceiling. Comprende?

-- 
main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
):_;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/

Post a reply to this message