 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Jack wrote:
>
> All the images were assumed_gamma 1.0
Should be no gamma at all.
> (a+b)^c != a^c+b^c ? The math I get, but I don't understand how this
> relates to compositing.
Just try it. Render a two pass scene with two light sources with
assumed_gamma and do a composite. Then render both sources on
with assumed_gamma. The resulting images are not identical.
_____________
Kari Kivisalo
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Jack wrote:
>
> pretrace_start 0.08
> pretrace_end 0.005
> count 100
> nearest_count 10
> error_bound 0.1
> recursion_limit 10
> low_error_factor .01
> gray_threshold 0
> minimum_reuse 0.001
> brightness 2
> adc_bailout 0.01/2
> max_sample 1.5
>
I wonder how you choose that values, starting with settings from
rad_def.inc (in Povray 3.5) should usually lead to agreeable results, both
in quality and speed.
In detail:
- recursion_limit 10 must be incredibly slow.
- brightness 2 is exaggerated although it might serve some artistic
purpose.
Christoph
--
Christoph Hormann <chr### [at] gmx de>
IsoWood include, radiosity tutorial, TransSkin and other
things on: http://www.schunter.etc.tu-bs.de/~chris/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Mick Hazelgrove wrote:
>
> I've not tried this with pov but it is rapidly becomming the industry
> standard method as it's faster and offers more control.
And how can radiance calculations be correct, if you don't have
lit/unlit areas ? That's the whole concept behind "global
illumination", taking account of all interactions...
Fabien.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Jack wrote:
>
> After reading a couple chapters of Jeremy Birn's Digital Lighting &
> Rendering, I decided to try rendering in passes (radiosity pass, shadow
> pass, etc). Saves a lot of time and I can make slight adjustments with more
> control, i.e. adjusting the darkness of the shadows.
BTW, is the kitchen already built, or is that a project ?
If it's a project, there's some ergonomic faults to correct...
Fabien.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Jack wrote:
>
> I like your purist attitude
It's not about purism, it's about getting right result.
For example, the light passing trough the window blinds should "burn"
on the wall, and (through radiance) lighten the surfaces surrounding
the light spot.
Fabien.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Kari Kivisalo wrote:
> Default radiosity brightness is the correct one so no need to
> touch that.
Some "exposure" filtering would be good, in the future, though.
Fabien.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Sun, 11 Nov 2001, Kari Kivisalo wrote:
> So, Gimp has additive layer mode which is missing in Photoshop.
I had thought that Photoshop's "Lighten" mode did addition. Am I wrong
about that?
I tried to duplicate Jack's image from the three layers (using the
"lighten" mode for the light source layer); without gamma
correction it's very dark but it becomes better after gamma correction.
> If the new version runs longer than 5 minutes without crashing
> I'm going to find out if it can do gamma preview :)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
It is a kitchen to be built, but like I said the rendering is a WIP. There
should be a window in the obvious empty space on the wall--nothing I can do
about placing cabinets over that. With that in mind, what do you suggest? I
would love to do something better with the kitchen, but unfortunately the 2
windows (one that is casting the sunlight shadow and the one that I haven't
placed in the scene yet) cut up my wall space.
"Fabien Mosen" <fab### [at] skynetbe> wrote in message
news:3BEE54A5.75A03C01@skynet.be...
> Jack wrote:
> >
> > After reading a couple chapters of Jeremy Birn's Digital Lighting &
> > Rendering, I decided to try rendering in passes (radiosity pass, shadow
> > pass, etc). Saves a lot of time and I can make slight adjustments with
more
> > control, i.e. adjusting the darkness of the shadows.
>
> BTW, is the kitchen already built, or is that a project ?
> If it's a project, there's some ergonomic faults to correct...
>
> Fabien.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
They are not correct, which is a downside of rendering in layers--parts of
the illumination have to be faked, requiring more artist time. The advantage
is that rendering time is MUCH less. I guess you have to find the right
balance between these two.
"Fabien Mosen" <fab### [at] skynetbe> wrote in message
news:3BEE5467.FEF6D860@skynet.be...
> Mick Hazelgrove wrote:
> >
> > I've not tried this with pov but it is rapidly becomming the industry
> > standard method as it's faster and offers more control.
>
> And how can radiance calculations be correct, if you don't have
> lit/unlit areas ? That's the whole concept behind "global
> illumination", taking account of all interactions...
>
> Fabien.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I chose recursion limit 10 because less seemed to not cover the corners (for
instance under the soffit). Part of the problem I suppose is that I chose to
only have reflected light in the radiosity pass. Perhaps lower the recursion
limit and add some direct lighting in the radiosity calculation?
Brightness of 2 came out of rendering a dark scene and needing it to be
lighter. Is the "limit" 1, (is 2 super-bright like rgb 2,2,2?).
"Christoph Hormann" <chr### [at] gmxde> wrote in message
news:3BEE4563.2C80E3ED@gmx.de...
>
>
> Jack wrote:
> >
> > pretrace_start 0.08
> > pretrace_end 0.005
> > count 100
> > nearest_count 10
> > error_bound 0.1
> > recursion_limit 10
> > low_error_factor .01
> > gray_threshold 0
> > minimum_reuse 0.001
> > brightness 2
> > adc_bailout 0.01/2
> > max_sample 1.5
> >
>
> I wonder how you choose that values, starting with settings from
> rad_def.inc (in Povray 3.5) should usually lead to agreeable results, both
> in quality and speed.
>
> In detail:
>
> - recursion_limit 10 must be incredibly slow.
> - brightness 2 is exaggerated although it might serve some artistic
> purpose.
>
> 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
|
|
| |
| |
|
|
|
|
| |
