POV-Ray : Newsgroups : povray.binaries.images : GI test. No radiosity - just light dome. Server Time
17 Aug 2024 18:21:35 EDT (-0400)
  GI test. No radiosity - just light dome. (Message 1 to 5 of 5)  
From: Andy Cocker
Subject: GI test. No radiosity - just light dome.
Date: 20 Aug 2001 23:58:20
Message: <3b81dc5c@news.povray.org>
Hi,

I was just experimenting with using a dome of low powered point lights
instead of radiosity, and here is the resulting image.

The small image shows the placement of the light grid.

Unfortunately, I placed the lights using a couple of #while loops, and this
leads to more lights being grouped together directly above the scene. Anyone
know how to code a geodesic dome kind of thing, so the lights are evenly
spaced? The code should easily allow for using a small number of lights for
test renders, and then upping the number of iterations (?) for final
renders.

I quite like the effect anyway, although even 100% jpeg is unkind to the
smooth areas.

Andy Cocker


Post a reply to this message


Attachments:
Download 'GI-test-show-grid.jpg' (6 KB) Download 'GI-test.jpg' (54 KB)

Preview of image 'GI-test-show-grid.jpg'
GI-test-show-grid.jpg

Preview of image 'GI-test.jpg'
GI-test.jpg


 

From: Norbert Kern
Subject: Re: GI test. No radiosity - just light dome.
Date: 21 Aug 2001 00:43:44
Message: <3b81e700$1@news.povray.org>
Hi,

a buckminsterfulleren has a theoretically ideal distribution on a sphere.
But a half sphere is limited to only 30 points.
Here I have the coordinates (from my warm_up entry last round)

light_source{< 14580,    8540,  -29570> color rgb 0.1}
light_source{<  1320,   10620,  -32310> color rgb 0.1}
light_source{< 21160,   16820,  -20690> color rgb 0.1}
light_source{< 14450,   27140,  -14650> color rgb 0.1}
light_source{<  1100,   29230,  -17430> color rgb 0.1}
light_source{< -5450,   21000,  -26210> color rgb 0.1}
light_source{< 29880,   11510,  -11560> color rgb 0.1}
light_source{<-18180,   19080,  -21550> color rgb 0.1}
light_source{<-24260,    6770,  -22920> color rgb 0.1}
light_source{<-31590,    1520,  -12620> color rgb 0.1}
light_source{<-19570,   26180,   -9870> color rgb 0.1}
light_source{< -7620,   32460,   -7320> color rgb 0.1}
light_source{< 19050,   28180,   -1790> color rgb 0.1}
light_source{< 28580,   18540,     150> color rgb 0.1}
light_source{<-33040,    8630,    -820> color rgb 0.1}
light_source{<-27170,   20980,     480> color rgb 0.1}
light_source{< -2940,   33720,    5650> color rgb 0.1}
light_source{< 10330,   31450,    8450> color rgb 0.1}
light_source{< 29440,   11970,   12250> color rgb 0.1}
light_source{<-32360,    2060,   11240> color rgb 0.1}
light_source{<-22700,   22150,   13090> color rgb 0.1}
light_source{<-10250,   28770,   15770> color rgb 0.1}
light_source{< 11310,   25060,   20570> color rgb 0.1}
light_source{< 20810,   15150,   22440> color rgb 0.1}
light_source{< 17610,    3530,   29050> color rgb 0.1}
light_source{<-26080,   10060,   20000> color rgb 0.1}
light_source{< -1110,   23170,   25300> color rgb 0.1}
light_source{< -4450,   11060,   32240> color rgb 0.1}
light_source{<  4820,    1500,   33960> color rgb 0.1}
light_source{<-16900,    4490,   29540> color rgb 0.1}

But it looks not very homogenous with primitives, you can see distict
shadows on planes.

Norbert



"Andy Cocker" <big### [at] mariner9fsnetcouk> schrieb im Newsbeitrag
news:3b81dc5c@news.povray.org...
> Hi,
>
> I was just experimenting with using a dome of low powered point lights
> instead of radiosity, and here is the resulting image.
>
> The small image shows the placement of the light grid.
>
> Unfortunately, I placed the lights using a couple of #while loops, and
this
> leads to more lights being grouped together directly above the scene.
Anyone
> know how to code a geodesic dome kind of thing, so the lights are evenly
> spaced? The code should easily allow for using a small number of lights
for
> test renders, and then upping the number of iterations (?) for final
> renders.
>
> I quite like the effect anyway, although even 100% jpeg is unkind to the
> smooth areas.
>
> Andy Cocker
>
>
>


Post a reply to this message

From: Kari Kivisalo
Subject: Re: GI test. No radiosity - just light dome.
Date: 21 Aug 2001 01:05:22
Message: <3B81ECB3.BA81DA7D@engineer.com>
Andy Cocker wrote:
> 
> I was just experimenting with using a dome of low powered point lights
> instead of radiosity

To be specific about this you are using point lights instead of a dome
and 1 bounce radiosity. Radiosity can do much more than that :)

BTW, where does this desire to use a hemisphere of lights to replace
radiosity (http://roy.gmxhome.de/) come from ? It's a rather limited
arrangement and 1 bounce radiosity should be as fast.


_____________
Kari Kivisalo


Post a reply to this message

From: Steve
Subject: Re: GI test. No radiosity - just light dome.
Date: 21 Aug 2001 06:36:48
Message: <slrn9o4e5l.7u3.steve@zero-pps.localdomain>
On Tue, 21 Aug 2001 04:53:38 +0100, Andy Cocker wrote:
>Hi,
>
>I was just experimenting with using a dome of low powered point lights
>instead of radiosity, and here is the resulting image.

Interesting stuff. 

--
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

 11:31am  up 38 days, 13:35,  2 users,  load average: 1.01, 1.02, 1.00


Post a reply to this message

From: Equiprawn
Subject: Re: GI test. No radiosity - just light dome.
Date: 21 Aug 2001 14:11:51
Message: <3b82a467@news.povray.org>
Hey,

Looks cool, though the shadows at the base of the pillars look odd (you can
see the smaller column, which is quite in shade, thought the ground just
below it isn't).

As for the whole geodesic dome thing, take a look at
http://www.cris.com/~rjbono/html/domes.html. There is C++ source code for
generating geodesic dome coordinates, as well as a Windows program that
outputs Povra scene files. Just found it through a Google search.

All the best,

Element


Post a reply to this message

Copyright 2003-2023 Persistence of Vision Raytracer Pty. Ltd.