Announcement

Collapse
No announcement yet.

long calculating times for Light cache

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • long calculating times for Light cache

    I have quite a large file which has very long light cache calculating times...longer than the IR map and actual rendering. I am currently working on an animation (only 30 sec) and left the office last night with the computer doing the LC....expecting it to be done today...well after 18 hrs it is still working on it! Even for long animations the LC has always been quite quick. Subdivisions 1500, sample size 0,01, scale set to screen. I see that the time remaining is currently going up and in about 3 hrs I have not seen the progress bar make any more progress...it is a bit more than half way. I'm guessing I should just cancel it now.....but do not know what to do with it. What could be causing this long LC time?? Never done this, but what would happen if I only used the IR map for the animation and no LC?? Do not know how to work around this! Any suggestions???? Thanks.

  • #2
    Long LC calculation times are usually a material problem. Mostly when an object's reflection or refraction is set to pure white 255,255,255. I've attached a script you can run that will change 255,255,255 color swatches to 250,250,250. Run it real quick on your scene and let me know if it improves your LC calculations!

    Code:
    global materialWhiteToAlmostWhite
    
    fn materialWhiteToAlmostWhite mat = (
     if (isProperty mat #reflection) and mat.reflection == (color 255 255 255) then (
      mat.reflection = (color 250 250 250)
     )
     if (isProperty mat #refraction) and mat.refraction == (color 255 255 255) then (
      mat.reflection = (color 250 250 250)
     )
    )
    
    (
     for m in sceneMaterials do (
      if (isProperty m #materialList) then (
       -- multisub
       for i = 1 to m.materialList.count do (
        -- for each sub material
        materialWhiteToAlmostWhite m.materialList[i]
       )
      ) else (
       -- not multisub
       materialWhiteToAlmostWhite m
      )
     )
     
     for v in (getclassinstances VRayColor) where v.color == (color 255 255 255) do v.color = (color 250 250 250)
     
     twosided = getClassInstances VRay2SidedMtl
     
     for m in twosided do (
      try (
          if (m.texmap_translucency.color == (color 255 255 255)) then m.texmap_translucency.color = (color 250 250 250) 
      ) catch ()
      try materialWhiteToAlmostWhite m.frontMtl catch ()
      try materialWhiteToAlmostWhite m.backMtl catch ()
     )
     
     falloffs = getClassInstances Falloff
     
     for f in falloffs where (f.color1 == (color 255 255 255)) do f.color1 = (color 250 250 250)
     for f in falloffs where (f.color2 == (color 255 255 255)) do f.color2 = (color 250 250 250)
    )
    Colin Senner

    Comment


    • #3
      Hey thanks! I will try that out Monday! I just cancelled the previous LC and redid it with a sample size of .05......got througyh it in 3 hrs...still way too long...but at least it got there....now it is working on the IR map. Thanks again! I will try it monday!

      Comment


      • #4
        excellent, let me know. The script I just threw together real quick, but I tested it and it works.
        Colin Senner

        Comment

        Working...
        X