Announcement

Collapse
No announcement yet.

Test QMC + LC #2

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

  • Test QMC + LC #2

    It is possible faster.
    Dual Xeon 3.06 4Gb
    http://img291.imageshack.us/my.php?i...rsonalv8gd.jpg

  • #2
    That is a pretty good combination, looks great, how many subs for the QMC?
    Eric Boer
    Dev

    Comment


    • #3
      30 subdivs. It is possible and less.

      Comment


      • #4
        i dont get it.

        This combination (QMC + LC) has been used by Natty, and quite a few other for along time.
        Whats the post proving?
        That it can be rendered faster?.... well then tell us how you rendered it faster.

        Whats the post about? that you can get good results with that combo, or that you find a way to make it render faster?

        Nice render BTW

        Comment


        • #5
          Only that I like this combination. And it is more than anything.

          Comment


          • #6
            ahh ok.

            It does work very well doesnt it

            Comment


            • #7
              I'm curious about by this, I have test it and the only thing that I have found is waste of time But very good quality of GI.

              Could you share your settings: QMC AA/ QMC GI and QMC sampler. Because it will be very cool if someone know how to render faster with this combination, because the test are very long and fastidious .

              thanks.
              =:-/
              Laurent

              Comment


              • #8
                ...

                QMC AA:
                Min 1 Max 4

                QMC GI:
                Subdivs 30

                QMC sampler:
                Adaptive amount 0.8
                Noise thresh 0.001 (Very much influences quality and speed )
                Min samples 16
                Global subdivs 1.0

                Speed most of all depends on quantity of light sources.

                Comment


                • #9
                  I heard somewhere, i think it was Morbid Angel, that if you have the Global subdivs set to 4 or something that it actually rendered faster... maybe that was QMC/QMC tho.

                  Might be worth a shot, of course i could be completely mixing things up

                  Comment


                  • #10
                    Yes it's Morbid, and he says also that he uses this combination all the time in animation.
                    Thanks for sharing! very appreciate
                    =:-/
                    Laurent

                    Comment


                    • #11
                      nice render!

                      I like the cables and the cds hidden under the desk
                      Chris Jackson
                      Shiftmedia
                      www.shiftmedia.sydney

                      Comment


                      • #12
                        I like the detail. verry nice and a good way to render. Thanks
                        Daniel Westlund

                        Comment


                        • #13
                          nice render
                          what your saying is that qmc + lc is faster then im + lc?
                          what is the render time on that scene with im for first bounce?
                          http://www.3dvision.co.il

                          Comment


                          • #14
                            Not faster, but more than fine shadows and caustics.
                            With a Ir map + LC, will be no more than 3 hours.

                            Comment


                            • #15
                              very nice rendering, I like the subtle DOF, but did you place the right back leg of the chair in the left of the image like this on purpose? Could well be since it matches the Escher poster on the wall. Otherwise it is strangely out of perspective.
                              You can contact StudioGijs for 3D visualization and 3D modeling related services and on-site training.

                              Comment

                              Working...
                              X