Erwin Putts and Dante Stella have each added new and valuable articles to this debate on their sites. Erwin's is a re-write of previous stuff that is set out in a clearer and more concise form. Dante's takes the same basic data and adds in the effects of film bulge and some good depth of focus calculations. The only 'flaw' I see in Dante's method is an assumption of film thickness (too thick by at least .05mm) and no calculations for wide lenses which have much shallower depths of focus and therfore show up more readily these errors . BTW Dante does a good job at pulling to bits the article on Leica/hexar compatibility by Mr Keppler which I got all worked up about a while back - nice one Dante!

-- Johann F (johannfuller@hotmail.com), June 07, 2002

Answers

The film thickness issue is interesting. Erwin Puts used 0.125mm base and 0.085mm emulsion, which I just adopted.

As an aside, wideangles do not have their own DOF calculations; as long as lenses are focused properly at infinity, any two focal length lenses of the same aperture have the same DOFocus (the factor that changes between wides, normals and teles closeup is magnification, but with any lens at infinity, magnification=0). The article, like that principle, assumes that there are no errors in the *lens* collimation (safe assumption with Leica, right?!). If the mechanical aspect of the lens is not up to snuff, it is true that wideangles do suffer more.

But what I can't wait for is that I have a Hexar RF body at exactly spec, a Leica M3 at exactly spec, and I am having a fast wideangle set exactly to Leica. When these come back from the respective service facilities, I will shoot some comparative tests - - and evaluate some tricks some people have suggested for moving the plane of focus.

-- Dante (dante@dantestella.com), June 07, 2002.

dante -- you failed to consider magnification. even if the defocus effects are the same, a long lens will magnify the error much more. thus, wideangles, as a practical matter (i.e. where you are making prints or projecting slides) have a much greater tolerance for this type of error.

-- roger michel (michel@tcn.org), June 07, 2002.

Roger, It's the other way round (we have been here before) - a small error in lens to film distance is much more significant with a wide lens than a long lens - increase the film to lens distance of a very wide wide (eg a 21mm) by even less than 1mm and you go from infinity to near a meter - on a tele (say a 135) you may not even notice a change in RF alignment!

-- Johann F (johannfuller@hotmail.com), June 07, 2002.

johann -- my point was only that for a given amount of DEFOCUS (which of course will be greater for a specific FFD error in a wide than in a long -- we never disagreed about that), the practical impact will be greater in a long lens due to increased mag.

-- roger michel (michel@tcn.org), June 07, 2002.

johann -- i should add that, as a practical matter, the magnification issue appears to swamp the comparative defocus issue. case in point: contax ad nikon RF lenses have a different helical pitch (but the same BFD). nonetheless, the wides may be inerchanged on the two cameras because DOF (a phenomenon that is STRICTLY tied to mag) masks the defocus errors. the longs, however, are not interchangeable. that is why voigtlander/cosina has made its CL mount lenses available only in f = 21-35. you can say DOFocus is not the same as DOF, but both are simply an issue of nodal point to film distance for a given focal point. thus a back focus distance error, can always also be viewed as a simple focus error, making DOF relevant to considering whether it wll be visible.

-- roger michel (michel@tcn.org), June 07, 2002.

Roger, Very valid points but I think we are going off at tangents - I'm not that concerned with the actual RF acuracy in this debate - errors induced by out of spec lens cam rates, misaligned RF arms, insufficient effective baselines and inaccurate use etc - these can all to some extent be delt with by calibration and appropriate choice of body and lens. What concerns me is that a body which measures 'thick' can never focus a Leica lens at infinity and that error becomes greater the wider the lens.I would have no hesitation in buying a Hexar RF if it was only the cam rate and starting point that was different - I could adjust that and match it to the Leica spec. However the Hexar seems idealy suited to wide angle work with it's low mag finder but I wouldn't touch it with a barge pole if the only way I could get it to focus a wide angle at infinity was to machine down the lens flange or film rails because as it stands that's your only option.

-- Johann F (johannfuller@hotmail.com), June 07, 2002.

i agree in all respects. if i had concerns about my hexar rf, i would just send it to sk grimes in RI, have him mount an M lens, plop the rig on his collimator and then have him shave/shim the lens flange until it is perfect. and BTW, he WILL do this kind of thing (and not charge all that much. akin possibly to the added performance that can be obtained by taking a stock engine and porting/polishing the cylinders.

-- roger michel (michel@tcn.org), June 07, 2002.