Results 1 to 3 of 3

Thread: Separating GI Rays when through dielectric ?

  1. #1
    geo messy madno's Avatar
    Join Date
    Feb 2009
    Location
    Germany
    Posts
    799

    Separating GI Rays when through dielectric ?

    Hi,

    this question is difficult to explain. So it might be a little longish.

    I have a node setup that alows me to control illumination values comming from an inside pointing sphere which is textured with a hdr image (the node setup is made of pieces borrowed from all over the forum). No LW lights are used.

    Click image for larger version. 

Name:	separate_rays.png 
Views:	99 
Size:	200.6 KB 
ID:	126505

    First the explanation:

    GI Saturation = controls how strong the color of the sphere bleeds into other surfaces

    Reflection Intensity = how strong the sphere is reflected on other surfaces

    Direct Intensitiy = how strong the diffuse value of the sphere is if seen directly

    GI Intensity = how strong the sphere contributes to the radiosity illumination

    The cool thing is, that I can control all the values individually.

    Click image for larger version. 

Name:	GI-Render_001.png 
Views:	94 
Size:	786.1 KB 
ID:	126504 Click image for larger version. 

Name:	GI-Render_002.png 
Views:	81 
Size:	794.1 KB 
ID:	126503 Click image for larger version. 

Name:	GI-Render_003.png 
Views:	85 
Size:	742.9 KB 
ID:	126502 Click image for larger version. 

Name:	GI-Render_004.png 
Views:	96 
Size:	554.6 KB 
ID:	126501 Click image for larger version. 

Name:	GI-Render_005.png 
Views:	85 
Size:	538.8 KB 
ID:	126506

    All surfaces except of the sphere are made with delta material (blurry reflection active).


    BUT and this does not work when I have a dielectric material for the window glass.
    In that case the Direct Intensity has no effect anymore. Seems so that the Reflection takes over its part. So I assume that a surface seen through a dielectric is based on bounced rays and not direct ones.

    Does anybody have an idea how to enhance this node setup so that it can work through a dielectric as well?

  2. #2
    TrueArt Support
    Join Date
    Feb 2003
    Location
    Poland
    Posts
    7,981
    Bounces = 0 will be when initial ray from camera hit visible directly from camera polygon.
    It's not just "reflections counter"..
    Refracted or other ray will also have it incremented.

    Make Gradient, that depends on Bounces value, and attach it to reflective/refractive/materials scene.. You will see what colors will be on different surfaces.
    Color in gradient at 0 bounces will be only on not reflective, not refractive, pretty basic surface.

  3. #3
    geo messy madno's Avatar
    Join Date
    Feb 2009
    Location
    Germany
    Posts
    799
    Got it I think.

    In case of my windows the directly visible background is in bounce no 2 (= the third bounce). I can just change the logic node accordingly and get back the option to control the direct visibility. This won't work if there were cascaded transparent surfaces (this would require a stack of logic nodes, if possible at all), but for this type of scenes it's ok.

    Click image for larger version. 

Name:	GI-Render_007.png 
Views:	64 
Size:	143.1 KB 
ID:	126517 Click image for larger version. 

Name:	GI-Render_008.png 
Views:	70 
Size:	444.9 KB 
ID:	126516 Click image for larger version. 

Name:	GI-Render_009.png 
Views:	58 
Size:	577.0 KB 
ID:	126515

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •