Announcement

Collapse
No announcement yet.

Tutorial Needed for VRay Dome Light + HDRI

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

  • #46
    Thank you Paul.

    I was never too impressed with those direct captures of sunlight for rendering. I think it's mostly Debevec's theoretical practice to demonstrate that it is possible.
    LunarStudio Architectural Renderings
    HDRSource HDR & sIBL Libraries
    Lunarlog - LunarStudio and HDRSource Blog

    Comment


    • #47
      Originally posted by Paul Oblomov
      The main cause - is overexposed HDRi. It's just wrong to do in that way...
      Well yeah.. but the gamma 2.2 broke the HDR even further so it wasnt able to produce sharp shadows.

      The scene was abit of a mess really..but it probabaly all came around from the crap HDR

      Comment


      • #48
        Do tell us how you fixed it tho Paul... would be really handy.

        Comment


        • #49
          Well - open it in PS. You wont' see anything but white - that's wrong. So we have to fix it - go to image->adjustments->exposure... Then make whatever you like to see not only white. For this particular case - I made gamma 2.2 + lowered exposure level, so I saw bright enough sun. That's it. Then just save it and use VRay default HDR Map parameters...

          You can say "that could be done in VRay...", well you're right, but that overexposured HDRi - it's just wrong Hope, you got what I was talking about
          I just can't seem to trust myself
          So what chance does that leave, for anyone else?
          ---------------------------------------------------------
          CG Artist

          Comment


          • #50
            I think Vlado mentioned the overly bright multipliers before with these particular probes. The other option (not using the physical camera) was to lower the multiplier to an extremely low amount like .01.
            LunarStudio Architectural Renderings
            HDRSource HDR & sIBL Libraries
            Lunarlog - LunarStudio and HDRSource Blog

            Comment


            • #51
              the hdr is correct being blown out, if you reduce the exposure in photochop you lose some of the range of the image. You are better off keeping it as is and reduce the multiplier of it
              Chris Jackson
              Shiftmedia
              www.shiftmedia.sydney

              Comment


              • #52
                You actually loose no range by lowering the exposure and saving the file. You're basically just balancing it out into a visible range. The floating point values are still embedded into the HDR file.
                LunarStudio Architectural Renderings
                HDRSource HDR & sIBL Libraries
                Lunarlog - LunarStudio and HDRSource Blog

                Comment


                • #53
                  the hdr is correct being blown out, if you reduce the exposure in photochop you lose some of the range of the image. You are better off keeping it as is and reduce the multiplier of it
                  No. You're wrong - PS can work with HDR images with needed precision. And it doesn't replace float (0.xxxxx) values with integer (I mean RGB 255 max)... It's just lowering the values. That's it. So - you lose nothing.

                  P.S. I'm about PS CS2 - it can work with HDR images (it can read and modify images higher, than 16-bit per channel).

                  edit- jujubee were first
                  I just can't seem to trust myself
                  So what chance does that leave, for anyone else?
                  ---------------------------------------------------------
                  CG Artist

                  Comment


                  • #54
                    the debevec skyprobes are angle maps, not mirror balls.
                    Just make sure the VrayHDR map is instanced to the texture of the dome light, and you should get nice sharp (er) shadows.

                    Comment


                    • #55
                      Yeah thats all sorted SV.
                      He had it set to mirrored ball not angular, but it was instance properly. But he also had gamma 2.2 in the vrayhdr as well which killed the HDRI totally.

                      All sorted now tho

                      Comment


                      • #56
                        OK guys, it looks like I have finally found some success

                        Here is the result:



                        It looks like, as was previously mentioned, that the problem has the gamma setting for the hdri map. It seems to me that the sun doesn't really begin to show until the gamma setting is at 1.2 (the above example was done at gamma = 1). Why is this so. I set it to 2.2 earlier because that is the gamma correction value for my output device. What affect does the gamma value have on vray hdri maps? Is it mentioned anywhere in the vray manual?

                        I have another question concerning hdri maps. I have a the following hdri map based upon a hemispherical fish-eye projection (no, it is not the earth)



                        What is the proper map type in the hdri settings. Should any of the horizontal and/or vertical transformation should be used with the appropriate map type?


                        Thanks

                        Marcus

                        Comment


                        • #57
                          You generally dont need to correct HDRI as they are already in linear work space (I believe) or something along those lines.

                          That one is probabably classed as a light prob.
                          Get yourself a copy of HDRshop V1 and in that you can convert all HDRI types into what ever type you want. I usually convert them into latitude/longiture maps, as they are easy to visualise when you see them and are often easier to work with.

                          Comment

                          Working...
                          X