kanner@tymix.UUCP (Herb Kanner) (03/22/86)
I am belatedly adding a bit of fuel to the fire on the subject of rangefinder cameras vs slr cameras. Contradicting a recent contribution to this newsgroup, I have read in several books and columns written by professional photo-journalists the statement that they use a rangefinder for candid work in poor light. They gave two reasons: one apparently psychological--that the absence of the flipping mirror gave them a feeling of continuity and intimacy with their subject, the other that focussing was easier. I want to talk about focussing. Reactions to equipment are very personal. For example, about ten years ago, when I bought a Nikormat, several reviewers claimed that the OM1 was outstandingly easy to use when wearing glasses with respect to seeing into the corners. For me, the camera was outstandingly bad in that respect. Now, when a person claims that a camera is hard or easy to focus in comparison to some other camera, I feel that we may be dealing with a purely subjective perception. Let me explain. Suppose we have cameras A and B. Camera A has a rangefinder with a four inch base and a lousy optical system in the rangefinder. Camera B has a rangefinder with a one-half inch base and a superb optical system in the rangefinder. Now, the person trying his damndest to make the overlapping images coincide would conclude that camera A is hard to focus. But, if you conducted a test similar to "timed" fire on a pistol range: give the user one second in which to focus on an arbitrary target and snap, I'll bet that camera A gives a higher proportion of well focussed pictures. With this in mind, let me relate a calculation I did some years ago on the fake rangefinder screen that you find in most SLRs. These are made by imbedding two crossed prisms in the glass. The amount of image shift for a given focussing error is determined by the angle between these two prisms. The limitation put on this angle in manufacture is set by the requirement that it be possible to find a position for the eye at which both halves of the area be illuminated. You all know, of course, that as you stop down, particularly with long focus lenses, that half of your split image thingy tends to go dark. There is a certain fixed diameter D, for a given split image prism, such that if the lens is stopped down below D, one half will always go dark. The relation D=F/f holds, where F is focal length and f is f-number. So, the manufacturer designs the prism for a D value which is reasonable for his line of longer lenses. I.e., if it is to work with a 100 mm f3.5 lens, then D must be set at about one inch. Now, what I calculated is that the image shift you get as you change focus is exactly that which you would get from a coupled range-finder whose base distance is D. That is not very impressive! I discovered something interesting about micro-prisms. These are usually in the area surrounding the split image. They are merely a lattice work of miniature split-image pairs. A slightly out of focus image shimmers. The interesting fact is that the shimmer is caused by camera vibration. Put the camera on a tripod and tighten all the clamps and the micro-prism becomes useless. Enough already ... -- Herb Kanner McDonnell Douglas (TYMNET) ...!hplabs!oliveb!tymix!kanner
kanner@tymix.UUCP (Herb Kanner) (03/22/86)
In article <707@tymix.UUCP> kanner@tymix.UUCP (Herb Kanner) writes: > >With this in mind, let me relate a calculation I did some years ago on the >fake rangefinder screen that you find in most SLRs. These are made by >imbedding two crossed prisms in the glass. The amount of image shift for a >given focussing error is determined by the angle between these two prisms. >The limitation put on this angle in manufacture is set by the requirement >that it be possible to find a position for the eye at which both halves >of the area be illuminated. You all know, of course, that as you stop >down, particularly with long focus lenses, that half of your split image >thingy tends to go dark. There is a certain fixed diameter D, for a given >split image prism, such that if the lens is stopped down below D, one half >will always go dark. The relation D=F/f holds, where F is focal length and >f is f-number. So, the manufacturer designs the prism for a D value which >is reasonable for his line of longer lenses. I.e., if it is to work with a >100 mm f3.5 lens, then D must be set at about one inch. > This posting is to correct some errata before I get flamed. I did the above derivation that establishes the relation between a SLR split image and an equivalent rangefinder some ten years ago, showed it to a couple of interested people, and threw it away; I have been writing from memory. "Particularly with long focus lenses" should read "... with short ..." ^^^^ "line of longer lenses" should read "... shorter ..." ^^^^^^ A more precise statement would be that D is the diameter of the exit pupil of the lens. The D value for the shortest focal length lens made by the manufacturer when the lens is wide open determines the maximum angle between the crossed prisms. Peace. -- Herb Kanner McDonnell Douglas (TYMNET) ...!hplabs!oliveb!tymix!kanner
smh@mhuxl.UUCP (henning) (03/28/86)
**** **** From the keys of Steve Henning, AT&T Bell Labs, Reading, PA mhuxl!smh > belatedly adding a bit of fuel to the fire on the subject of > rangefinder cameras vs slr cameras. 1) split images are just a cruch for cameras that can't look through a REAL lens... 2) a rangefinder is only worthwhile if it has a leaf shutter... 3) rangefinders with focal plane shutters have all the disadvantages and none of the advantages of any camera but they make good tie bars... Seriously, if you do much flash macro photography you learn to love SLR leaf shutter cameras which have a sync speed of 1/1000 second and super critical focusing over the entire frame, not just in the center.