To create a photonic transistor, two input holograms are needed.
    One input hologram is on the left side and the other is on the right. 
    They project light onto the standard output square so that when the
    light at both input squares is on, the combined image relocates  
    the photons to produce an output image quite different from the image 
    produced by either one of the input holograms by itself. 
 
       The fringe component separating mask, then separates the light 
    so as to produce the desired outputs, for each of the 4 possible input 
    states.  
 
       As an example of how this works, let's make a hologram for the right
    side of the transistor. This one is to project a similar, but not a
    duplicate, of the first hologram's image.  We will do this by using the
    program FIMAGE to flip the wavefront pattern left for right. This
    causes the projected image to flip also, when projected from the
    right side onto the output square.  The many results from image flipping 
    are greatly simplified in this case because the projected image is a 
    single point, and therefore a symmetrical image that can be flipped
    this way.