Subject:
|
Are my radiosity settings up to date?
|
Newsgroups:
|
lugnet.cad.ray
|
Date:
|
Tue, 6 Sep 2005 08:17:57 GMT
|
Viewed:
|
8260 times
|
| |
|
|
hi folks,
disapointed by the overall rendering performance of our radiosity settings (the
longest engaged 3 weeks to calculate) I stopped the developement of better
tweakings a long time ago. now I have to do a huge rendering for a show
(7000x5000) and would like to know if you have figured out better settings in
the meantime? what I'm currently using is a 2-pass:
file: radiosity.inc
// 2-pass rendering:
// first pass with PASS=1, half resolution, no antialiasing
// second pass with everything switched on, good antialiasing, etc.
#declare PASS = 1;
// Radiosity Settings
#declare AMB = 0;
#declare DIF = 1;
global_settings {
assumed_gamma 1
ambient_light 0
max_trace_level 25
radiosity {
pretrace_start 0.08
pretrace_end 0.01
count 150
error_bound 0.1
nearest_count 7
recursion_limit 1
low_error_factor 0.7
gray_threshold 0
minimum_reuse 0.01
brightness 1.0
adc_bailout 0.01/2
normal on
media off
#if (PASS=1) // first pass, save radiosity values
save_file "Filename.rad"
#else // second pass, load radiosity values
pretrace_start 1
pretrace_end 1
load_file "Filename.rad"
always_sample off
#end
}
}
// Create an infinite sphere around scene and allow any pigment on it
sky_sphere {
pigment {
gradient y
color_map { [0.0 color rgb <1, 1, 1>] }
}
}
#declare FIN = finish {
ambient 0
diffuse 1
brilliance 1.2
#if (PASS = 2) reflection { 0.025,.5 falloff 4 } #end
specular 1
}
#include "lugnetcolors.inc"
combined with Orion's includes
http://www.pobursky.com/modules.php?op=modload&name=News&file=article&sid=8&mode=thread&order=0&thold=0
and Todd's colors http://news.lugnet.com/cad/ray/?n=1805.
compared to Koyan's and Jeroen's settings I get a better performance as well as
less errors on trans-colors (check out the wallpaper "1979" at my site
http://www.holly-wood.it/wallpapers2-de.html which took approximately 2 hours to
render the 1280x1024 without any stud-logo).
nontheless I believe we haven't reached the cutting-edge tweakings yet. has any
of you done some test in the near past and may came up with something better
than the 2-pass?
w.
|
|
1 Message in This Thread:
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
