Announcement

Collapse
No announcement yet.

Light through glass, small explanation

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

  • Light through glass, small explanation

    Following the topic in problems thread I desided it would be a good idea to make a separate thread here so others can look for it is they have issues.

    I would like to explain what is affect shadows and why it is there, and how else light can pass through the glass, and what are the physical properties of that:
    Affect shadows is a "fake" or a "cheat" to allow fast renderings of light which has passed through glass. However its physicly incorrect. The reason for this is, every light energy that has passed through glass has become a caustic energy.
    Therefore you need caustics for the light to show from outside in interriors. For this the refractive caustics must be turned on, however because of often used biased method such as irradience map, caustics will never show, thus the illumination from passed light through glass will not appear, producing result shown in example 3 below.

    So, to see the light we eather turn on affect shadows, which is imo a good cheat for a fast and effective rendering time, or we turn off affect shadows, turn on refractive caustics and set illumination method to qmc/lc with very high sampling, or lc/lc in ppt mode and wait for many hours for the light to show.
    Here are some research tests I've done for maxwell forum, but they also apply to this:

    Image 1, shows maxwell caustics illumination through glass



    Image 2 shows vray glass with affect shadows



    image 3 shows when affect shadows is turned off



    And image 4 shows physical caustics in vray with affect shadows off, and refractive caustics on

    Dmitry Vinnik
    Silhouette Images Inc.
    ShowReel:
    https://www.youtube.com/watch?v=qxSJlvSwAhA
    https://www.linkedin.com/in/dmitry-v...-identity-name

  • #2
    Thanks for that explanation!!!

    Comment


    • #3
      thanks, good explanation!

      and the rendertimes?
      when the going gets weird, the weird turn pro - hunter s. thompson

      Comment


      • #4
        thanks. I have been wondering.
        If I dont use caustic photons and only in the indirerect light, check the reflect and refract caustics I never get the same result even when using ppt at a very high sampling level. it seems to me, the photon caustics are more "spectacular"?

        is the vray caustic image using photons?
        Daniel Westlund

        Comment


        • #5
          getting light to pass through glass

          so do I just check caustics in in the vray indirect illumination pulldown?

          I have plent y of light comming through with the caustics on. I have high settings all around and I'm getting these darn light spots again. I have these problems now and then. I tried all the usual stuff vray syas to do when these things show up.

          I have a very long render time as is with the HSph samples set to 50 andInterp samples set to 30 on HIGH. this scene at 480x360 took almost two hours on a dual opteron 270 node and there's till more detail that I'm going to add.



          http://static.flickr.com/122/257539694_a485512956_o.jpg
          res3d

          Comment


          • #6
            Try to play with the caustics only until you get the parameters right. This tutorial deals with the global photon map, but you can use exact same approach to caustics:

            http://www.spot3d.com/vray/help/150R...rials_pmap.htm

            Basically there are these things to consider:

            (*) Which lights are producing caustics - there is typically no use to make all lights to emit them as this wastes memory and render time.

            (*) If the light that should produce caustics is the VRaySun, adjust its photon radius - so that it just about encompasses the area where you want caustics to appear. Making the photon radius too large will spread the photons over a larger area and you will need more of them - which again increases memory usage and render times.

            (*) You should adjust the correct Max. density and Search distance parameters. If you are modelling in accurate units, maybe good starting values would be about 1cm for the Max. density and about 5cm for the Search distance.

            I hope this helps.

            Best regards,
            Vlado
            I only act like I know everything, Rogers.

            Comment


            • #7
              Is there a reason why this does not work with multi-sub materials
              Alain Blanchette
              www.pixistudio.com

              Comment


              • #8
                a small bug in rc3.
                Drop the glass sub-mat into a vrayblend, and it should sort it...

                Lele

                Comment


                • #9
                  The result is different, with the same glass, I do not have the same reflections.....



                  As shown, the multi sub version does reflect less, and it is the very same glass
                  Alain Blanchette
                  www.pixistudio.com

                  Comment


                  • #10
                    @ Morbid Angel
                    Hey...cool tests....
                    I am searching some page where a compares are shown Maxwell vs V-ray.
                    And some scientific proofs that are saying which Render Engine have to be bought.
                    I am working for CAAD Department at the TU Munich Germany and i am really searching the way to make my asisstents to see the true...
                    For me is basicly and fully clear...but...
                    Can u share render times please.
                    I wish u all the best!!!
                    Express Your Imagination - let it MAX!!!
                    www.3dkad.de

                    Comment


                    • #11
                      you may want to look at maxwell forum. Those guys love to compare to other renderers. However as a maxwell owner as well, I will say that most recent maxwell releases cannot match the quality of vray or of maxwell beta. Not to mention render times and other features. Look there. I my self have few samples to show.
                      Dmitry Vinnik
                      Silhouette Images Inc.
                      ShowReel:
                      https://www.youtube.com/watch?v=qxSJlvSwAhA
                      https://www.linkedin.com/in/dmitry-v...-identity-name

                      Comment

                      Working...
                      X