Uli - currently, both Schneider and Rodenstock build to a tolerance of +/- 0.5mm for the FFD specified. I have no data for Nikon or Fuji. In any case, the Fotoman HFM provides +/- 1mm of internal adjustment capability (Schneider & Rodenstock HFM's provide +/- 0.5mm) to compensate for the allowable manufacturing tolerance. From our experience calibrating MANY lenses, all of the makers most usually deviate on the longer (+) side of the stated FFD... probably due to stack-up. Regarding the use of a pre-marked or un-marked HFM, a pre-marked mount includes an accurate DOF scale for the focal length being mounted... a distinct asset.

GPS - HFM's from every maker are accurate (out of the box) for infinity focus with a given lens ONLY on a pot-luck basis... maybe 10% of the time. The reason is that every lens is different... even within a single production run. For this reason you must always find or verify infinity focus anyway. Assuming the components have been manufactured within tolerance, the final "calibration" of the HFM becomes the last fine tuning step in the "Harmonizing" effort.

Determining the exact length for each component of the assembly is not that difficult as long as you have precise measurements (see my earlier posting). Maintaining Parallelism across muliple components on the other hand is more difficult. However assuming one uses quality machine tools and proper machining practice, acceptable parrallelism is quite acheivable.