Re: Architectural shader and inverted faces.

["Fred Valleur" <fred.valleur(at)gmail.com>]

It had some of the polygons inverted.
As Hannes mentionned, you would also in that case get discontinuities with other surface shaders, my bad, I was trying to reproduce a problem I had on a production scene. 
The real problem originally happened on a CAD model with some inverted faces *and* user normals (my example oversimplified that fact),  but I can't  reproduce the problem now. 
So until the bug is repro I guess I will have to call it a closed case.

Thanks all for looking into this and sorry for the noise,
    
Frederic.

On Dec 22, 2007 3:08 PM, kim aldis < xsi(at)kim-aldis.co.uk> wrote:

Just to clarify, the sphere to the right has some or all polygons inverted?

 

From: owner-xsi(at)Softimage.COM [mailto:owner-xsi(at)Softimage.COM] On Behalf Of Fred Valleur
Sent: 20 December 2007 15:23
To: XSI(at)Softimage.COM
Subject: Re: Architectural shader and inverted faces.

 

Hi,

thanks for the prompt answer, but I don't think it solves the problem. When "weak internal reflections" is turned on, the reflections are totally removed as if there was some backface culling happening.  When turned off, the reflections are visible, but  incorrectly computed, unless I am mistaken (the right sphere on the attached image has inverted polygons and you can see the reflections are different). 
Hopefully I missed something between the documentation and Zap's blog.

  Cheers,
  Frederic.

On Dec 20, 2007 1:11 PM, Hannes < polygonislands(at)gmail.com> wrote:

There is a default for the shader under " additional options"  > turn off weak internal reflections.
if you turn it off it should behave correctly.

>>> http://mentalraytips.blogspot.com/2007_01_01_archive.html

Cheers, Hannes

On Dec 20, 2007 10:29 AM, Fred Valleur < fred.valleur(at)gmail.com > wrote:

Hi,

it seems the Architectural shader does not behave properly on geometries with inverted faces. I have a simple grid with half faces inverted, and although the "Visible Geometry Side" parameter is set to both in the mental ray optimization tab, the reflections are different on each half of the grid, as if only front facing polygons were taken into account.
it seems to be only happening in the reflection computation, the diffuse part seems to be using the properly inverted normals.

Does anyone know a workaround for this?

    Cheers,
    Frederic.


--
Frederic Valleur | LaMaison

--
Frederic Valleur | LaMaison

 

--
Frederic Valleur | LaMaison

-- 
Frederic Valleur | LaMaison