 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I've replaced that awful 'garbage can' in my last image with an H.R.
Gigeresque alien foetus like thing that I modeled some time ago. I have also
spruced up the apartment a little with more objects, it no longer looks so
much like an institution. By introducing the bright outdoor light that comes
through the slits in the boarded up window I have greatly increased the
number of artefacts, so I will probably need to do another render with
higher settings. Here are the radiosity settings I'm currently using:
pretrace_start 0.01
pretrace_end 0.005
recursion_limit 1
error_bound 0.6
nearest_count 10
count 1000
I could try increasing error_bound but then I would loose the little detail
that I already have in the shadows. Or I could increase the count, what is
the max for count anyway?
Post a reply to this message
Attachments:
Download 'bedroom2 2.JPG' (62 KB)
Preview of image 'bedroom2 2.JPG'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Oh yeah this is pure radiosity in this image, no lights of any kind.
"Thomas Lake" <tla### [at] home com> wrote in message
news:3b4e8ceb@news.povray.org...
> I've replaced that awful 'garbage can' in my last image with an H.R.
> Gigeresque alien foetus like thing that I modeled some time ago. I have
also
> spruced up the apartment a little with more objects, it no longer looks so
> much like an institution. By introducing the bright outdoor light that
comes
> through the slits in the boarded up window I have greatly increased the
> number of artefacts, so I will probably need to do another render with
> higher settings. Here are the radiosity settings I'm currently using:
>
> pretrace_start 0.01
> pretrace_end 0.005
> recursion_limit 1
> error_bound 0.6
> nearest_count 10
> count 1000
>
> I could try increasing error_bound but then I would loose the little
detail
> that I already have in the shadows. Or I could increase the count, what is
> the max for count anyway?
>
>
>
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Thomas Lake wrote:
>
> Oh yeah this is pure radiosity in this image, no lights of any kind.
I can tell from the shadows.
Since you have no "real" lights, you'll get much more realistic lighting
results in this image by increasing recursion to 2. You may get more
artifacts than you're willing to put up with, though.
Max count is 1600 IIRC.
-Xplo
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Looks pretty nice. The lighting on most of the objects look very realistic. I'd
suggest that those radiosity artefacts are less visible on more complex forms,
but show up really bad on the perfectly flat walls, so maybe you could just put
more detail in the room to hide the artefacts? I confess I don't know much about
good radiosity settings :)
--
Tek
http://www.evilsuperbrain.com
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Thomas Lake wrote:
>
> [...]
> Here are the radiosity settings I'm currently using:
>
> pretrace_start 0.01
> pretrace_end 0.005
> recursion_limit 1
> error_bound 0.6
> nearest_count 10
> count 1000
>
You should probably decrease 'adc_bailout', the suggested value is
'0.01 / brightest ambient object'
Christoph
--
Christoph Hormann <chr### [at] gmxde>
IsoWood include, radiosity tutorial, TransSkin and other
things on: http://www.schunter.etc.tu-bs.de/~chris/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Xplo Eristotle wrote:
>
> Max count is 1600 IIRC.
>
Is there any reason for this limitation BTW?
Christoph
--
Christoph Hormann <chr### [at] gmxde>
IsoWood include, radiosity tutorial, TransSkin and other
things on: http://www.schunter.etc.tu-bs.de/~chris/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Christoph Hormann wrote:
>
> Xplo Eristotle wrote:
> >
> > Max count is 1600 IIRC.
> >
>
> Is there any reason for this limitation BTW?
Not that *I* know of. I remember being pretty pissed when I found it.. I
was working on my own "light a room with only radiosity" project and I
just couldn't get rid of the artifacts.
-Xplo
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Christoph Hormann wrote:
>
> Thomas Lake wrote:
> >
> > [...]
> > Here are the radiosity settings I'm currently using:
> >
> > pretrace_start 0.01
> > pretrace_end 0.005
> > recursion_limit 1
> > error_bound 0.6
> > nearest_count 10
> > count 1000
> >
>
> You should probably decrease 'adc_bailout', the suggested value is
> '0.01 / brightest ambient object'
Assuming he's using MegaPOV, I would actually suggest *increasing* it to
1. This is counterintuitive, but in my tests it seemed to give a small
speed boost without significantly affecting the image, whereas other
values had no useful effect.
-Xplo
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
High!
Thomas Lake schrieb:
> I've replaced that awful 'garbage can' in my last image with an H.R.
> Gigeresque alien foetus like thing that I modeled some time ago. I have also
> spruced up the apartment a little with more objects, it no longer looks so
> much like an institution. By introducing the bright outdoor light that comes
> through the slits in the boarded up window I have greatly increased the
> number of artefacts, so I will probably need to do another render with
> higher settings. Here are the radiosity settings I'm currently using:
As you replied to a comment on the first version (hmm, up to now I haven't
found out how to adress news articles as URLs...), you originally went for a
gloomy, oppressing Orwellian mood - so what about a televisor displaying the
latest news from the Malabar front, one of those famous "Big Brother is
watching you!" posters, a crumpled pack of Victory cigarettes (with the tobacco
crumbling out of the paper) and a newspeak dictionary on the table, not to
mention a more worn-out and shabby overall look of the scene?
Now, the room looks more to me like the dwelling of a daylight-dodging cranky
computer hermit...
See you in Khyberspace (under construction)!
http://www.geocities.com/electricafghan/index-e.html
Yadgar
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Xplo Eristotle wrote:
>
> Assuming he's using MegaPOV, I would actually suggest *increasing* it to
> 1. This is counterintuitive, but in my tests it seemed to give a small
> speed boost without significantly affecting the image, whereas other
> values had no useful effect.
I remember i once managed to reduce artefacts by decreasing it, not sure
if it works the same in this situation of course.
Christoph
--
Christoph Hormann <chr### [at] gmxde>
IsoWood include, radiosity tutorial, TransSkin and other
things on: http://www.schunter.etc.tu-bs.de/~chris/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
