 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
1. The light gathered by radiosity seems to be clipped to the interval
[0,1]. This should be significant when using very bright lightsources or
ambient-only scenes. Objects in shadow should be able to get brighter than
their diffuse values, I think.
2.Pov-3.1 did uncorrectly apply color-clipping and gamma-correction *before*
antialiasing. I'm not sure whether this has been changed in 3.5 because this
is difficult to confirm without looking into the source (or measureing the
brightness of single pixels)
CB (pov-ray3.5b4/athlon-600/384 MB RAM/Win98)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"CB" <fis### [at] gmx net> wrote:
> 1. The light gathered by radiosity seems to be clipped to the interval
> [0,1]. This should be significant when using very bright lightsources or
> ambient-only scenes. Objects in shadow should be able to get brighter than
> their diffuse values, I think.
>
Did you try changing the max_sample setting in radiosity?
--
Margus Ramst
Personal e-mail: mar### [at] peakeduee
TAG (Team Assistance Group) e-mail: mar### [at] tagpovrayorg
Home page http://www.hot.ee/margusrt
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"CB" wrote:
> 2.Pov-3.1 did uncorrectly apply color-clipping and
> gamma-correction *before* antialiasing. I'm not sure
> whether this has been changed in 3.5
I don't know about gamma correction but color clipping is still done before
antialiasing.
However, doing color clipping after antialiasing have both advantages and
disadvantages.
Advantages:
An area with many small but very bright spots will be smoothened out to a
bright area.
(This can for example be used to do a blurred reflection using a reflective
texture with a normal that is scaled very small. Currently that doesn't work
well with bright objects because color clipping is done before
antialiasing.)
Disadvantages:
A very bright object will appear not to be antialiased even though
antialiasing is used because the bright area of every antialiased pixel will
dominate the result. I'm afraid this would be a pretty serious problem.
So it's a design question, not a bug.
Rune
--
3D images and anims, include files, tutorials and more:
Rune's World: http://rsj.mobilixnet.dk (updated June 26)
POV-Ray Users: http://rsj.mobilixnet.dk/povrayusers/
POV-Ray Webring: http://webring.povray.co.uk
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
CB wrote:
>
> 1. The light gathered by radiosity seems to be clipped to the interval
> [0,1]. This should be significant when using very bright lightsources or
> ambient-only scenes. Objects in shadow should be able to get brighter than
> their diffuse values, I think.
I tested this with cornell.pov and didn't notice any clipping. Put
100 and then 200 as light source magnitude and observe how the front
of the small box gets brighter even when the floor, which reflects
light to the box, is saturated. The ceiling also gets brighter than
0.75*255. I don't have time to do a definitive test but if you have a
scene which demontrates this feel free to post it.
> 2.Pov-3.1 did uncorrectly apply color-clipping and gamma-correction *before*
> antialiasing.
I think that aa is best done after gamma correction in peceptually uniform
color space because it's purpose is to reduce visual artefacts. Just like
jpeg compression is best done to gamma corrected image. It spreads the
"effect", smoothing or jpeg artefacts, evenly (visually) among dim and
bright samples.
_____________
Kari Kivisalo
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
