Re: [rendering] Bent Normals

Date : Thu, 7 Feb 2008 21:01:36 -0800
To : XSI(at)Softimage.COM
From : "Alan Fregtman" <alan.fregtman(at)gmail.com>
Subject : Re: [rendering] Bent Normals
I use the bent normals option sometimes to render a "normals pass" to
do fake lighting enhancements in some renders in post.

What other uses do you guys give to it? I'd love to know,

Cheers,

   -- Alan


On Feb 7, 2008 1:37 PM, guillaume laforge
<guillaume.laforge.3d(at)gmail.com> wrote:
> Hi Amaan
>
>
> >I think the AO node is adding 0.5 to all three XYZ components before
> >sending the output to a Vector2Colour node. The question is, why is 0.5
> > being added?
>
> I guess it is the way to store negative values of the bent normal vector in
> the rgb space. If you render a bent normal "as is" you will clamp the -x and
> -z values to zero.
>
> Cheers
>
> Guillaume Laforge
>
>
>
>
> On Feb 7, 2008 4:44 PM, Amaan Akram <xsilist(at)warpedspace.org> wrote:
> > Hello everyone,
> >
> > I am just trying to get my head around bent normals in Mental Ray --
> > following is my (mis?)understanding of what they are (please correct me
> > if I am wrong)
> >
> > -a bent normal is a direction vector that is calculated by taking an
> > average of all the un-occluded samples fired from around the surface
> > normal of an intersection point
> >
> > -for completely un-occluded areas, a bent normal is the same as the
> > surface normal
> >
> > Based on the above understanding, I run two tests
> >
> > Test case 1
> > Take a default primitive grid, plug a Vector State node straight in to
> > the Surface port of the Material node and set the Vector State to
> > "Normal Vector"
> >
> > Test case 2
> > Take a default primitive grid, plug an Ambient occlusion node straight
> > in to the Surface port of the Material node and set the ambient
> > occlusion output mode to "Return Bent Normals (World Space)"
> >
> >
> > According to my understanding of bent normals, both the above test cases
> > should produce the exact same result--but they don't. The RGB output for
> > the shader in both test cases should be:
> > R: 0
> > G: 255
> > B: 0
> >
> > The above mentioned result is the output of Test Case 1 (with Vector
> > State node), but...
> >
> > in Test Case 2, the RGB output is:
> > R: 127
> > G: 255
> > B: 127
> >
> >
> > I think the AO node is adding 0.5 to all three XYZ components before
> > sending the output to a Vector2Colour node. The question is, why is 0.5
> > being added?
> >
> > Thanks
> > Amaan
> > ========================================
> >
> > This e-mail (and any attachments) is confidential and may contain personal
> > views which are not the views of Men-from-Mars unless specifically stated.
> >
> > If you have received it in error, please delete it from your system. Do
> not
> > use, copy or disclose the information in any way nor act in reliance on it
> > and notify the sender immediately.
> > ---
> > Unsubscribe? Mail Majordomo(at)Softimage.COM with the following text in body:
> > unsubscribe xsi
> >
>
>
>
> --
> Guillaume Laforge
---
Unsubscribe? Mail Majordomo(at)Softimage.COM with the following text in body:
unsubscribe xsi


Search the XSI List archives here or use the advanced search form to search across mailing lists. Searching help is available.
This site supposedly brought to you by Benjamin Grosser and the Imaging Technology Group.