pico Posted June 23, 2012 Share #1 Â Posted June 23, 2012 Advertisement (gone after registration) Some our engineers might help with this idea. Â Might it be possible to correct for focus-shift by using a rotating prism lens such as found on the early Super Ikonta, or flat glass at a tilt (left or right) in front of the viewfinder's patch window? In each case the glass or prism would be adjustable and small so it can be flipped away from the patch window when working at an f-stop range that has no shift. Â Might an interference filter work, presuming we can get one that shifts the human eye's focus? (Idea taken from a Peak critical focusing scope for enlarger - it has a filter that supposedly shifts focus as it appears to the eye. I pulled mine off years ago.) Â In another thread I will pose a more likely, improved digital rangefinder for the M. Link to post Share on other sites More sharing options...
Advertisement Posted June 23, 2012 Posted June 23, 2012 Hi pico, Take a look here Focus Shift - a possible cure via the RF?. I'm sure you'll find what you were looking for!
Mark Pope Posted June 23, 2012 Share #2 Â Posted June 23, 2012 I think that it's an interesting idea I guess it's something that have to be integrated somehow into a camera's firmware. Wouldn't the camera need to know the following? 1. the lens being used (easy using coding) 2.aperture (also relatively easy, albeit a bit of a guess with the M8/9) 3. the focus distance, which would need some sort of means of detecting the position of the rangefinder focusing arm (or whatever that bit's called). Â A benefit of developing the last item is that in principlem it could be that it could be possible to provide framelines which adapted to the focus distance, eliminating framing errors, perhaps using LCDs to create the frame masks. Link to post Share on other sites More sharing options...
TomB_tx Posted June 23, 2012 Share #3 Â Posted June 23, 2012 Agreed, it is interesting. But since the amount of shift is a factor of the lens design, it would make more sense for the lens maker to build in correction. For example, the 35 f1.4 CV Nokton can be a great little lens, but it does focus progressively farther away as it is stopped down. Correcting by optical design would make it more complex and likely bigger and such. I'd bet they could build in a slight helical tied to the aperture ring, so that stopping down would move the optical unit slightly farther away by such a slight amount to correct the shift, while the focusing cam stayed put. Might be able to rotate the inner helicoid on the optical unit to do the same, rather than keeping it locked to the mount orientation, since it would only have to move a slight amount. For a rangefinder correction, recall the DOF marks at the top and bottom of the M3 patch? A narrow notch and a wider notch that indicate how much overlap you can have at two different f-stops for DOF. A similar mark might show the overlap needed to correct focus at a certain stop for a certain lens. I might see how that works with my M3 and 50 Heliar f2.0 - which shows quite a bit of shift. Link to post Share on other sites More sharing options...
pico Posted June 23, 2012 Author Share #4 Â Posted June 23, 2012 Agreed, it is interesting. But since the amount of shift is a factor of the lens design, it would make more sense for the lens maker to build in correction. [... snip - please see post ...] Â True. Such would require a custom focus ramp/cam for every single lens. Expensive, but considering how expensive Leica lenses are, perhaps warranted. A new method of inserting a custom machined focusing ramp could be made - it could be driven by computation controlled machining in conjunction with an optical tester. Maybe I'm being too futuristic. For a rangefinder correction, recall the DOF marks at the top and bottom of the M3 patch? [...] I remember it from the M2. Foggy memory. Maybe it was the M3. I have neither now. Very, very cool idea it was. Link to post Share on other sites More sharing options...
mjh Posted June 23, 2012 Share #5  Posted June 23, 2012 2.aperture (also relatively easy, albeit a bit of a guess with the M8/9) A guesstimated f-stop value won’t do and that is all you get with the M8 and M9 (and I suppose the M10 as well). The guesstimated f-stop could also change at any time: Suppose one was trying to achieve perfect focus while some correction element in the rangefinder is busily shifting the goalpost – it could drive one mad.  Leica’s strategy is to replace lens designs exhibiting a problematic amount of focus shift by new designs that don’t. Link to post Share on other sites More sharing options...
pico Posted June 23, 2012 Author Share #6  Posted June 23, 2012 Leica’s strategy is to replace lens designs exhibiting a problematic amount of focus shift by new designs that don’t.  And just what is a 'problematic amount', and how successful has the redesign and manufacturing been? Link to post Share on other sites More sharing options...
mjh Posted June 23, 2012 Share #7  Posted June 23, 2012 Advertisement (gone after registration) And just what is a 'problematic amount', and how successful has the redesign and manufacturing been? If people are talking about problems with focus shift then obviously it is problematic – you may remember the focus shift threads on lenses like the old Summilux 35 mm or the old Noctilux. As we all know their redesign has been quite successful in reducing focus shift to a barely noticable amount.  An in-camera compensation of focus shift would be an option for an autofocus camera; a similar solution for a rangefinder camera that doesn’t even know which aperture was selected would be extremely complex and costly and frankly not worth the effort. Link to post Share on other sites More sharing options...
jaapv Posted June 23, 2012 Share #8  Posted June 23, 2012 2.aperture (also relatively easy, albeit a bit of a guess with the M8/9) . And quite impossible with any film M. The uncertainty of this parameter alone would make the whole exercise pointless. Not to mention the extreme mechanical complexity of transferring the focus distance in such a system.On top of that a considerable number of shifting lenses have been adjusted to the owners’ taste, adding a further undefined parameter. Link to post Share on other sites More sharing options...
Michael Geschlecht Posted June 23, 2012 Share #9 Â Posted June 23, 2012 Hello pico, Â Too fiddly. M's are supposed to be ergonometric, not fiddly. Â Many people consider the goggles of a 35 for an M3 too fiddly. What you are suggesting some people might consider even more fiddly. Â Part of the success of the M is that you can pick it up & use it pretty much the way you think you can. Minimal fiddling. Â Best Regards, Â Michael Link to post Share on other sites More sharing options...
pico Posted June 24, 2012 Author Share #10 Â Posted June 24, 2012 Okay. I will drop the whole idea into the bin, and put the my thread into the kill-file. . Link to post Share on other sites More sharing options...
Michael Geschlecht Posted June 24, 2012 Share #11 Â Posted June 24, 2012 Hello pico, Â Don't drop the whole idea. Think M. Â What is an elegant soloution that maintains the Leitz/Leica mindset? Â If people gave up just because things were difficult there would be no gumdrops, escalators or full frame sensors. Â Best Regards, Â Michael Link to post Share on other sites More sharing options...
marknorton Posted June 24, 2012 Share #12 Â Posted June 24, 2012 Focus Shift can be cured instantly by sensing actual focus on the sensor. CMOS of course... See my post in your thread about a new rangefinder. Link to post Share on other sites More sharing options...
lars_bergquist Posted June 24, 2012 Share #13 Â Posted June 24, 2012 Focus Shift can be cured instantly by sensing actual focus on the sensor. CMOS of course... See my post in your thread about a new rangefinder. Â That presumes that focusing will be done at working aperture. This gets less precise the more you stop down, because the AF sensing has to contend with a larger depth of field. Â The dour old spoilsport from the Rangefinder Age Link to post Share on other sites More sharing options...
jaapv Posted June 24, 2012 Share #14 Â Posted June 24, 2012 Okay. I will drop the whole idea into the bin, and put the my thread into the kill-file. . Â Well, the whole point is that it is much, much simpler to design a Summilux 35 asph FLE. Link to post Share on other sites More sharing options...
marknorton Posted June 24, 2012 Share #15  Posted June 24, 2012 That presumes that focusing will be done at working aperture. This gets less precise the more you stop down, because the AF sensing has to contend with a larger depth of field. The dour old spoilsport from the Rangefinder Age  Indeed, a point I make in the thread I refer to. But that's the whole point. By focussing at working aperture, you are eliminating focus shift.  It's clear that the best way to evaluate focus is from the sensor image data itself using a live-view capable sensor, even if the way that information is conveyed to the user is through the viewfinder as I suggest on the other thread. Link to post Share on other sites More sharing options...
jaapv Posted June 24, 2012 Share #16 Â Posted June 24, 2012 Not eliminating. Masking. Up to a point. Link to post Share on other sites More sharing options...
mjh Posted June 24, 2012 Share #17  Posted June 24, 2012 If you focus based on a live-view image you can actually eliminate focus shift, rather than just masking its effects. Given the trend towards lens designs without discernible focus shift on the one hand and the likelihood of an M10 offering live-view on the other, is there a non-zero chance for Leica implementing an extremely complex system to counteract focus shift by correcting the rangefinder image? I don’t think so. Link to post Share on other sites More sharing options...
jaapv Posted June 24, 2012 Share #18 Â Posted June 24, 2012 If your liveview allows a focussing accuracy that is sufficient. I believe it is usually within 1/2 DOF, which is in the same magnitude as focus shift stopped down. Link to post Share on other sites More sharing options...
pico Posted June 24, 2012 Author Share #19  Posted June 24, 2012 If people gave up just because things were difficult there would be no gumdrops, escalators or [...]l  I have this vision of someone on a stopped escalator shouting, "Help! I'm trapped!" Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.