 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
The documentation on saving and loading of radiosity doesn't give an awful
lot of information.
When I save radiosity data to a file, does it save the pre-render data only,
or does it also save samples taken during the render phase? It seems to be
the pre-render only, but I want to be sure of this.
Is the save file independent of resolution? Can I save radiosity data on a
160x120 image and then load it on a 1600x1200 image? My guess is that it is
not resolution-dependent, but again, I want to be sure.
In what situations do people normally use save/load of radiosity? It seems
to have fairly limited utility, but it also seems to be fairly commonly
used. Is there something that I'm missing? I can tell you right now that
the only use that I have found for it is when I am rendering with multiple
processes via SMPOV. For some reason, certain radiosity settings tend to
not render well in tiles. Each tile looks decent by itself, but when I put
the image together, there are obvious seams where the radiosity
calculations were different for the separate tiles. Saving the values in a
file and then loading them for each tile seems to almost completely
eliminate this problem, although I rarely think I still can see some very
faint seams. I may be imagining things, though.
Thanks.
- dan B hentschel
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
news:web.416881092fa1b46b31e6d870@news.povray.org...
> When I save radiosity data to a file, does it save the pre-render data
only,
> or does it also save samples taken during the render phase? It seems to be
> the pre-render only, but I want to be sure of this.
No, it save all radiosity samples, the ones taken during the pre-trace and
the ones during the final trace.
> Is the save file independent of resolution? Can I save radiosity data on a
> 160x120 image and then load it on a 1600x1200 image? My guess is that it
is
> not resolution-dependent, but again, I want to be sure.
Yes.
> In what situations do people normally use save/load of radiosity?
It is used for the 2 pass rendering technic. This technic consist of doing a
first pass at a relatively low resolution without anti-aliasing and with a
simplified geometry but with high quality radiosity parameters. The
radiosity samples are saved and reused in a second render with
anti-aliasing, high resolution, complete geometry. For this final render, no
radiosity samples should be computed, only the ones from the first render
should be use so there is no pre-trace (pretrace_start and pretrace_end set
to 1) and no radiosity computation during the final phase (always_sample
off)
See "Dummy radiosity" on this page http://www.ignorancia.org/t_tips.php
Post a reply to this message
|
|
| |
| |
|
|
From: Tim Nikias
Subject: Re: Questions on saving/loading of radiosity data
Date: 10 Oct 2004 07:20:32
Message: <41691b00$1@news.povray.org>
|
|
|
| |
| |
|
|
> It is used for the 2 pass rendering technic. This technic consist of doing
a
SNIP
The 2 pass technique may also include multiplying the error_bound by 4 or
16, as this smoothenes the samples in the 2nd pass. You lose the small
details for a more softer appearance. In my tests I've found that this can
smoothen some artifacts away. See here:
http://www.nolights.de/past_pros/radiosity/radiosity.html
Regards,
Tim
--
"Tim Nikias v2.0"
Homepage: <http://www.nolights.de>
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Thank you very much! Both responses (and links) were very helpful! I now
have a much better understanding. Off to the experiments...
- dan B hentschel
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|
