Hi, I'm writing a laser range finder simulator using the depth component of a scene.
I'm linearising the range in a fragment shader using.. linearRange = (2.0*near) /(far+near-(nonLinearRange*(far-near))) I was hoping to display the foggy depth image by allowing the rtt camera to calculate the near/far planes and pass these values to the shader for use in the linearisation. I had hoped that this would mean that the calculated range should remain the same whether DO_NOT_COMPUTE_NEAR_FAR was set or not. I'm getting the near/far numbers from rttCamera->getProjectionMatrixAs Frustum and passing these to the shaders in a viewer->frame loop using Uniform->set(), but the linearised range is different dependent on DO_NOT_COMPUTE_NEAR_FAR. I can confirm that the near/far values change within the loop as the scene changes. Is it valid to pass data to a shader using Uniform-set() in a viewer->frame loop? Is my assumption that linearised range should be independent of DO_NOT_COMPUTE_NEAR_FAR if the near/planes are passed to the linearisation algorithm correct? Can anyone comment on the algorithm used (which seems to be generally accepted)? Ta. John ------------------ Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=35796#35796 _______________________________________________ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
