Agree that instructions printed on the template itself, on the "old" Feickert protractor, are very vague. In fact, you sort of have to know what to do a priori in order to understand the instructions. For example, you set up the protractor based on the range of the expected P2S (where S = Spindle) distance, and then you set the stylus on an arc calibrated based on P2S, when in fact you are setting pivot to stylus distance, in that first step. But, at least with my original version Feickert, Steps 1, 2 AND 3 ought to be spot on, as I think it was Lutke said. If step 3 is not spot on, there was a small error in step 1 or 2. A small error in the first two steps is magnified in step 3, which is why step 3 is offered. This, my friends, is why I never bother with Step 3; I don’t give a hoot about a small error.
I agree with Chakster on the need for magnification and a high intensity light, if you wanna get crazy with the Feickert or most other protractors. That is exactly what is offered for the premium price you have to pay to get the SMARTractor, which is otherwise based on the Dennesen, which is also the basis for the Feickert. In addition, the Smartie gives you perfect accommodation for any one of three spindle diameters. (I own a UNItractor, the precursor to the Smartie.)