[Leica] Rangefinder error on M9

[hlritter at bex.net (Howard Ritter)]

Frank, Marty, Mark?

Well, I don't know whether this is focus shift with stopping down or not, 
and I doubt I could determine this with tests. One would have to step the 
lens through the indicated focus point, examine the images highly magnified, 
and determine where the maximum sharpness occurred. At f/16, it would 
probably take a mathematical analysis of the images and plotting of a curve 
to determine within a few percent where perfect focus is.

My intuition is that the degree of defocusing is more than aperture-related 
focus shift would produce in any design that Leica would let into 
production. (Always was a pollyanna.)

I think I'll just have to remember to rotate the barrel a millimeter to the 
near side when shooting at medium distances and wide apertures. And do some 
testing to see how far I have to stop down to be happy settling for what the 
RF gives me.

?howard


On Nov 7, 2010, at 6:02 AM, Frank Dernie wrote:

> Hi Howard,
> I think what you are seeing here is the effect of "focus shift". It seems 
> many lenses have a shift in focus with aperture which is normally masked 
> by the increasing depth of field with aperture closing, and, on film 
> bodies, by the effectively thick sensitive layer, since film emulsion is 
> fairly thick, and the film rarely flat anyway, the focus will fall 
> -somewhere- within the sensitive volume (it is not that much). A digital 
> sensor is flat and therefore less tolerant of focus shift, the 90mm f2 apo 
> has a tiny depth of field and at some apertures the plane of focus falls 
> outside the depth of field due to focus shift.
> It is also difficult to get perfectly in focus anyway, I use mine with a 
> viewfinder magnifier.
> A quick search on the internet will turn up quite a lot of discussion on 
> this and the 90mm f2 apo-asph lens is one of the most sensitive lens out 
> there for this problem - sadly.
> The new 35mm f1.4 asph came out because focus shift on the existing lens 
> was being noticed on the M8 and M9 when it was not using film bodies.
> cheers,
> Frank Dernie