Loading ...
Sorry, an error occurred while loading the content.
 

Testing Fulldome Animations...

Expand Messages
  • Daniel Ott
    Hello, I was wondering if there has been any discussion about techniques for testing full dome animations in 3D software packages, such as Maya? I m currently
    Message 1 of 2 , Jun 7, 2004
      Hello,

      I was wondering if there has been any discussion about techniques for
      testing full dome animations in 3D software packages, such as Maya? I'm
      currently working on my thesis which implements a lens shader to be use
      within mental ray for Maya; however, I'm trying to find the best way to test
      my output since I don't have access to a dome. I have tried using a
      spotlight gobo technique that projects the image onto a hemisphere and
      planar texture mapping on a hemisphere, but both appear to be improper.
      There are perspective distortions when viewing from dome center. (I know
      there should be distortions along the outer edges, but how much spherical
      distortion should there be when viewing from dome center?) Does anyone have
      any other suggestions?

      Thanks,

      Daniel Ott
      dott@...
    • Don Davis
      Hello, I was wondering if there has been any discussion about techniques for testing full dome animations in 3D software packages, such as Maya? I m currently
      Message 2 of 2 , Jun 7, 2004
        Hello,

        I was wondering if there has been any discussion about techniques for testing full dome animations in 3D software packages, such as Maya? I'm currently working on my thesis which implements a lens shader to be use within mental ray for Maya; however, I'm trying to find the best way to test my output since I don't have access to a dome. I have tried using a spotlight gobo technique that projects the image onto a hemisphere and planar texture mapping on a hemisphere, but both appear to be improper. There are perspective distortions when viewing from dome center. (I know there should be distortions along the outer edges, but how much spherical distortion should there be when viewing from dome center?) Does anyone have any other suggestions?

        ===

        Solving this problem may be very specific to the software used. General advice would be to try texturing an accurate grid on a sphere and placing the camera at the center aimed at one of the 'poles', or one could also model a 10 degree per segment sphere and render it as a wireframe with your 'fisheye'. The distance between the 'latitude' concentric circles should be even and consistent from center to edge in the final circular rendered image. Carefully examining your final image in detail from center to edge should give a decent idea of how the image quality holds up with suitable test subjects.

        Ideally a hemispheric projection, for instance with spheres shown in different parts of the dome, will appear to a viewer at the center of the dome as an accurate recreation of the viewing geometry with all spheres appearing round. An architectural simulation rendered and projected in true hemispheric perspective is an amazing thing to behold. Many 'flat dome' facilities sometimes compress the rendered enviornment to bring the horizon above the high cove line, but this introduces distortion which prevents round objects from appearing so in any location within the dome. Rendering seperately the landscapes and planets, etc. would be a possible compromise.

        Don Davis
      Your message has been successfully submitted and would be delivered to recipients shortly.