Announcement

Collapse
No announcement yet.

AA, MB and DOF vs. rendertime

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

  • AA, MB and DOF vs. rendertime

    hello all

    I have a new project. It's an animation about blood, cells and a new drug. so we have to work with a lot DOF and motionblur. my question now, what's the best balance of quality and rendertime? wich AA is the best for DOF and noisefree frames and of course no flickering? any experiences? the other method would be scanline with finalDOF. is it working with vray, too. there was some problems with earlier versions of vray, what's about 1.5?

    here are some images from the tv serie "dr. house".


    many questions, I hope for many answers.
    thanks in advance.

    best regards
    themaxxer
    Pixelschmiede GmbH
    www.pixelschmiede.ch

  • #2
    You dont need much AA when you have copious amounts of DOF and Moblur. Fixed at like 2 maybe three would probably be best. But I dont have alot of experience with it.
    ____________________________________

    "Sometimes life leaves a hundred dollar bill on your dresser, and you don't realize until later that it's because it fu**ed you."

    Comment


    • #3
      I've been using frischluft lenscare in after effects a lot recent;y and it looks really really good - well worth a look and total control of dof in post.

      Comment


      • #4
        I second using lenscare in After Effects. good quality and you can make adjustments to your dof really fast unlike going through another round of 3d renderings with dof baked in.

        another option is Richard Rosenmann's dof plugin for photoshop. Its a photoshop plugin but you can animate its functions and apply the effects to image sequences.

        V Miller

        Comment


        • #5
          I have a magic number when doing hard core QMC stuff like DOF and/or MB. More of a ballpark.

          for low level fixed 4 to 6
          for mid level qmc 1,16 at 0.005
          for high level qmc 2,25 at 0.002
          in rare cases, I have to go with something like 2,50 or 2,80 at 0.001

          A lot of this depends on if you are in pure linear data, how much darks you have, etc...

          Comment

          Working...