Hello Hairy,

    Good points. When I wrote the script, I did some research on the net to find the difference between the parallel and convergent settings. Apparently, the parallel adjustment is better because convergence can generate Keystone effect. However, my tests in Poser didn't reveal big difference. Moreover, I read that keystone can be corrected by modern software. So, practically, the two approaches are acceptable. Personnaly, I prefer to have a choice of the stereobase (see below my point of view on that)

  About the stereobase adjustment. First, parallax and stereo base are not the same thing. They are related by the fact that parallax increases when stereobase increases for the same scene setting (Bercovitz formula). You can have the same parallax for different stereobase by working also with the distance between camera and objects in the scene (distance as such or focal length).

   A large stereobase didn't imply bad viewing. For example, at IMAX or on some still photos, you have probably seen landscape (Example: the Grand Canyon) with depth details that human eye cannot perceive. They are confortable and interesting to see. In the case of Poser, increasing the stereobase can be useful if you want to reveal depth of various body parts relative to each other. For example, if you look at my figures in this thread, the stereobase was 20 cm. For me, these anaglyphs are quite confortable to see.  One basic rule in stereography is to select a stereobase equal to 1/30 of the nearest object in the scene. Obviously, the stereobase cannot be changed for a physical camera (Fudji), but in Poser no problem at all.