landscape testing new radiosity patches [198k] (Message 1 to 7 of 7)
From: Christoph Hormann
Subject: landscape testing new radiosity patches [198k]
Date: 26 Aug 2005 16:05:36
Message: <densgm$d64$1@chho.imagico.de>
I already made this scene some time ago but i thought this might be a
good occasion to post it since we have just released MegaPOV 1.2.1 which
includes the radiosity patches used.
This image demonstrates the use of the adaptive pretrace and adaptive
error_bound patches in a large render of a detailed scene. One major
problem of such renders is that they tend to be very slow at large size
since a lot of additional samples are taken in the final pass which also
results in a raise of memory use. The adaptive error_bound gives better
control over this - in this case there were only 19 radiosity samples
taken in the final pass so you can predict the render time and results
much better from a small size test render.
The adaptive pretrace can reduce the pretrace time in a scene with large
areas only requiring very few samples (like the sky here). However you
have to use a low pretrace_start to avoid areas to be completely skipped
by the pretrace. The pretrace time here was 4 hours (final pass time
was 13 hours for 6400 x 3600).
The radiosity settings used are:
radiosity {
pretrace_start 0.006
pretrace_end 0.0015
adaptive 1
count 100
nearest_count 3
error_bound { 0.8 adaptive 1.5, 20 }
recursion_limit 1
low_error_factor 0.2
gray_threshold 0.0
minimum_reuse 0.015
brightness 0.85
always_sample off
}
Christoph
--
POV-Ray tutorials, include files, Landscape of the week:
http://www.tu-bs.de/~y0013390/ (Last updated 24 Jul. 2005)
MegaPOV with mechanics simulation: http://megapov.inetart.net/
From: Orchid XP v2
Subject: Re: landscape testing new radiosity patches [198k]
Date: 27 Aug 2005 03:33:38
Message: <43101752$1@news.povray.org>
1. That's an increadible image. Are you *sure* that's not a photo?
2. 4 hours FOR THE PRETRACE? :-| You must have A LOT of free time...
(How long did it take to DESIGN the scene??)
From: Christoph Hormann
Subject: Re: landscape testing new radiosity patches [198k]
Date: 27 Aug 2005 05:45:02
Message: <depcgs$55j$1@chho.imagico.de>
Orchid XP v2 wrote:
> 1. That's an increadible image. Are you *sure* that's not a photo?> > 2. 4 hours FOR THE PRETRACE? :-| You must have A LOT of free time...
Actually no - it's just the computer that needs some free time. ;-)
> (How long did it take to DESIGN the scene??)
Not very long - 2-3 days each a few hours.
Christoph
--
POV-Ray tutorials, include files, Landscape of the week:
http://www.tu-bs.de/~y0013390/ (Last updated 24 Jul. 2005)
MegaPOV with mechanics simulation: http://megapov.inetart.net/
Christoph Hormann <chr### [at] gmxde> wrote:
> I already made this scene some time ago but i thought this might be a> good occasion to post it since we have just released MegaPOV 1.2.1 which> includes the radiosity patches used.
FanTAStic image, as is all of your work I have seen. I am curious: do you
use "normal" lighting in these scenes in addition to radiosity?
From: Sebastian H
Subject: Re: landscape testing new radiosity patches [198k]
Date: 28 Aug 2005 07:45:50
Message: <4311a3ee$1@news.povray.org>
Christoph Hormann wrote:
> > I already made this scene some time ago but i thought this might be a > good occasion to post it since we have just released MegaPOV 1.2.1 which > includes the radiosity patches used.> > This image demonstrates the use of the adaptive pretrace and adaptive > error_bound patches in a large render of a detailed scene. One major > problem of such renders is that they tend to be very slow at large size > since a lot of additional samples are taken in the final pass which also > results in a raise of memory use. The adaptive error_bound gives better > control over this - in this case there were only 19 radiosity samples > taken in the final pass so you can predict the render time and results > much better from a small size test render.> > The adaptive pretrace can reduce the pretrace time in a scene with large > areas only requiring very few samples (like the sky here). However you > have to use a low pretrace_start to avoid areas to be completely skipped > by the pretrace. The pretrace time here was 4 hours (final pass time > was 13 hours for 6400 x 3600).> > The radiosity settings used are:> > radiosity {> pretrace_start 0.006> pretrace_end 0.0015> adaptive 1> count 100> nearest_count 3> error_bound { 0.8 adaptive 1.5, 20 }> recursion_limit 1> low_error_factor 0.2> gray_threshold 0.0> minimum_reuse 0.015> brightness 0.85> always_sample off> }> > Christoph> > > ------------------------------------------------------------------------>
It's just that most of your images are fantastic...
Sebastian
From: Stefan Viljoen
Subject: Re: landscape testing new radiosity patches [198k]
Date: 1 Sep 2005 02:35:33
Message: <4316a133@news.povray.org>
Christoph Hormann spake:
> > I already made this scene some time ago but i thought this might be a> good occasion to post it since we have just released MegaPOV 1.2.1 which> includes the radiosity patches used.> > This image demonstrates the use of the adaptive pretrace and adaptive> error_bound patches in a large render of a detailed scene. One major
Very impressive...
--
Stefan Viljoen
Software Support Technician / Programmer
Polar Design Solutions
Mobile: (+27) 83 4060854
Tel: (+27) 18 297 1753
Fax: (+27) 18 290 5556
http://polard.com
de> wrote: