Zeiss paper

[T3D john bercovitz <bercov@xxxxxxxxxxx>]

I reread the Zeiss paper from years back and things were not 
precisely as I recalled.  8-)  First, the paper was about 
contrast and resolving power, not acutance and resolving power.  
So that explains that confusion.  The paper is illustrated with 
four photos and since it's an old xerox, the contrast is jacked 
up and it's hard to see from the figures what they were talking 
about.  Oh, by the way, the paper was a reprint from a Zeiss #51 
technical information book.  Next, the graphs were called 
contrast transfer function and not MTF.  They look for all the 
world like MTF curves, though.  Percentage on the ordinate and 
spatial frequency on the abscissa. The contrast transfer curves 
are for photos, according to the text, so that would include 
acutance of the chemistry in the curves.  They did not separate 
this effect out.  Also, they used black and white bars instead 
of sine curves for test objects.  This part is OK, it's an 
accepted method, because there is a simple transform that takes 
care of the difference.  So what did they say?  Basically that 
resolution beyond what the eye can perceive is useless so you 
might as well cut the maximum resolution down and pump up the 
contrast at the lower, useable, spatial frequencies.  In graph 
form:


      100|*             l
   %     |  x  *        l
 con-    |    x   *     l
 trast   |     x    *   l
         |      x    *  l
         |       x    * l
         |        x    *l
         |         x    *
         |          x   l*
         |            x l *
         |              x  *
         |              l  x*
         |              l    *   x
         |              l      *        x
         |              l         *             x
         |              l             *                    x
       0 |__________________________________________________ 
         0        1        2        3        4        5
                 spatial frequency


So the curve marked by asterisks is better than the curve marked 
by exes if the visual acuity limit (for the distance from which 
the photo is being viewed) is marked by the ells.  They 
mentioned that under the best of conditions, a contrast of 5% 
was sufficient for resolution by the human eye.  Sure makes 120 
lines per mm sound pretty weak if it's at 5% contrast, eh?

John B


------------------------------