tom0511 Posted January 27, 2007 Share #1 Posted January 27, 2007 Advertisement (gone after registration) I just found out something which is new to me: I have to lenses where I realized that the tendancy to front focus changes with f-stop!!! My Noctilux seems to slightly front focus at f1.0 However at f1.4-f2.0 it seems to focus spot on! I am not talking about the larger DOF. My Summilux seems to also front focus just a little tiny bit at f1.4 but seems totally fine at f2.0 I can just recommend to do focus tests at different f-stops before you draw conclusions about a lens. The strange thing is that I wonder if the focus for my Nocti can be either optimized for f1.0 or f1.4 and not for both. Strange. Link to post Share on other sites More sharing options...
Advertisement Posted January 27, 2007 Posted January 27, 2007 Hi tom0511, Take a look here The focusing plane changes with f-stop!!!. I'm sure you'll find what you were looking for!
jaapv Posted January 27, 2007 Share #2 Posted January 27, 2007 Yes- focus shift. The Noctilux is well known for this. Many lenses do this to some extent. It has been extensively described in optical theory. You obviously cannot see this phenomen on an SLR. It is RF lore... Link to post Share on other sites More sharing options...
eronald Posted January 27, 2007 Share #3 Posted January 27, 2007 Yes- focus shift. The Noctilux is well known for this. Many lenses do this to some extent. It has been extensively described in optical theory. You obviously cannot see this phenomen on an SLR. It is RF lore... Why can't you see it on an SLR ? A modern SLR is autofocused, or focused, wide-open - while the image is captured stopped down. Edmund Link to post Share on other sites More sharing options...
jaapv Posted January 27, 2007 Share #4 Posted January 27, 2007 Because it uses the direct image projected on the matte screen/ sensor to focus. You focus accurately wide-open, and the shift on stopping down disappears into the DOF.Thus this does not show up. Link to post Share on other sites More sharing options...
tom0511 Posted January 27, 2007 Author Share #5 Posted January 27, 2007 So what to do about it? I would like my Noct ro focus precisly with f1.0 and with f1.4 ! Link to post Share on other sites More sharing options...
MP3 Posted January 28, 2007 Share #6 Posted January 28, 2007 I'm interested to know is this wide-open-front-focus issue a typical characterastic of all rangefinder lens or just your Noct and Lux? Best Matthew Link to post Share on other sites More sharing options...
wattsy Posted January 28, 2007 Share #7 Posted January 28, 2007 Advertisement (gone after registration) I'm interested to know is this wide-open-front-focus issue a typical characterastic of all rangefinder lens or just your Noct and Lux? My 35/1.4 ASPH and 50 ASPH have noticeable focus shift between F1.4 and F4. Since my M8 has come back from Solms it is backfocussing too much for comfort and the focus shift now comes into play. Link to post Share on other sites More sharing options...
jaapv Posted January 28, 2007 Share #8 Posted January 28, 2007 So what to do about it?I would like my Noct ro focus precisly with f1.0 and with f1.4 ! Nothing- except learn to compensate.... Link to post Share on other sites More sharing options...
ho_co Posted January 28, 2007 Share #9 Posted January 28, 2007 Why can't you see it on an SLR ? A modern SLR is autofocused, or focused, wide-open - while the image is captured stopped down. Edmund-- You're right, of course, technically. But in practice with an SLR, if you both focus and shoot wide open, you don't need to worry about focus shift because taking and shooting aperture are the same. And as you stop down, the increased depth of field will cover the focus shift. With a rangefinder, you're always relying on the same RF spot; depth of field will again cover the error when stopped down, but you're flying blind wide open. --HC Link to post Share on other sites More sharing options...
adan Posted January 28, 2007 Share #10 Posted January 28, 2007 Without getting into too long a diatribe on optics, let's just say that the larger the relative aperture (f/stop) of a lens, the more apparent it becomes that spherical lens surfaces are not the best way to focus light - they are just much cheaper and easier to make than the paraboloid "a-spheric" surfaces that would do the job right. The edges of the lens elements are moving further and further from an approximation of the ideal the wider they are. Therefore, with a lens like the Noctilux (f/1 and no aspheric corrections) when shooting at f/1 the edges of the glass are focusing light at a different point than the center of the glass. The best "apparent" focus will be at a point that averages together the mishmosh of light rays from all parts of the glass. Once you stop down to f/1.4, you have thrown away half the light rays, and in particular the half that were most aberrant (from the outer part of the glass). So the average point of best sharpness shifts closer to the point where an "ideal" perfect lens would focus. Stop down again to f/2 and you've tossed 3/4s of the light rays, again the "worst", and the average best focus point shifts even closer to the ideal point. When calculating the precise positioning of the Nocti glass and focus cam relative to the lens mount and film/sensor, Leica has to make a choice - calibrate everything for f/1, or for the more "ideal" focus locations at smaller apertures (or perhaps yet another "average" point in between the two). As a non-Nocti-user, I don't know which choice Leica makes, but Nocti-users may be able to help. Link to post Share on other sites More sharing options...
j. borger Posted January 28, 2007 Share #11 Posted January 28, 2007 So what to do about it?I would like my Noct ro focus precisly with f1.0 and with f1.4 ! I think you need your RF adjusted.... the Noctilux should focus spot on at F1 and f1.4 (its' most used apertures) and a focusshift should be noticable at f 2.0 and f 2,8 ..... after that dof is large enough you will not notice anyway..... this would be "normal" Noctilux behaviour! On the other hand .. if you use the Noctilux in real life shooting you will probably not notice a shift ... i never noticed untill i tried shooting rulers ... Link to post Share on other sites More sharing options...
tom0511 Posted January 28, 2007 Author Share #12 Posted January 28, 2007 I think you need your RF adjusted.... the Noctilux should focus spot on at F1 and f1.4 (its' most used apertures) and a focusshift should be noticable at f 2.0 and f 2,8 ..... after that dof is large enough you will not notice anyway..... this would be "normal" Noctilux behaviour! On the other hand .. if you use the Noctilux in real life shooting you will probably not notice a shift ... i never noticed untill i tried shooting rulers ... The thing is that in real life shot with the shallow DOF of the Noctilux its hard to say if images are not sharp because of movement, userfailure or maybe focus inaccurancys. In my experience with such fast lenses I allways have to take several images in order to get 1 or 2 which are focused spot on. I own over 10 M-lenses and did focus test with all of them yesterday. 2 showed slight back focus tendancy, some slight front focus, and some are just fine. I am just surprized if people do run focus test and say that all their lenses are totally precise. I think most of my lenses have focus inaccurancies which are not a real problem for using them. My 50/1.4asph which shows slight but consistent backfocus has been sent to Leica now for calibration. I am just beginning to wonder how reliable one can focus the fast lenses on a rangefinder. I also feel (as others have mentioned before in other threads) that the sharp-unsharp-tansition with digital is different (harder) compared to film (because the film-thicknes different from the sensor) and therefore minor focus inaccurancies show up faster/more in digital compared to film. The focus shift (and thanks for the explanations) was something new to me and makes it even harder. Link to post Share on other sites More sharing options...
tom0511 Posted January 28, 2007 Author Share #13 Posted January 28, 2007 I'm interested to know is this wide-open-front-focus issue a typical characterastic of all rangefinder lens or just your Noct and Lux? Best Matthew Not all, my 50luxasph for example shows slight backfocus The 75lux the front focus is minimal my 35lux asph slight front focus my 90/2.8 perfect focus ...... I am not talking about massive focus inaccurancies but small ones at short distance and wide open Link to post Share on other sites More sharing options...
fuchs Posted January 28, 2007 Share #14 Posted January 28, 2007 .... Therefore, with a lens like the Noctilux (f/1 and no aspheric corrections) .... Andy, the f/1.0 Noctilux (second version) actually is an asph lens, with one aspheric element as per the factory literature. The f/1.2 Noctilux (first version) had 2 aspherical elements, and the cost of producing them in the late 1960's was allegedly one of the causes its design was dismissed with the introduction of the f/1.0 version. It was indeed the first Leica aspherical lens ever to make it to the market, but the marketing designation "asph" in the lens' name was not yet being used at the time. best, Ed Link to post Share on other sites More sharing options...
adan Posted January 28, 2007 Share #15 Posted January 28, 2007 Eduardo: Can you provide a source for the info that the f/1 Nocti is ASPH? I believe that is not correct. Leica's Nocti .pdf does not mention aspheric surfaces (and they usually point them out in the optical cross-sections), E. Puts' lens compendium refers to the design "employing spherical surfaces" (as opposed to the hand-made aspheres in the previous f/1.2, as you mention), and the design of the lens dates to 1976, 15 years before Leica reintroduced aspheres in the first 35 f/1.4 Aspheric, and 20 years before they developed the (relatively) easier and less expensive blank-pressing technique for making aspheric surfaces. As to fast lenses and RFs, especially 50 and longer. Yes, RFs depend on a lot of moving parts ALL being in near-perfect calibration to one another. Including the link between the lens and the camera, which is being broken and restablished every time the lens is taken off and remounted. (whereas SLRs only need to keep 3-4 parts, and only one of them moving, in alignment: the image plane, the ground-glass (and/or AF sensor), and the mirror). Leica works to tight tolerances - but there are still tolerances. And if one happens to mount a Nocti or a 90 f/2 that left the factory at one end of the tolerance range on a body that happened to be just within the "opposite" end of factory specs, the combined error can be enough to foul the focusing. The rock photographer (and Leica user) Jim Marshall used to say that once he found a lens/body combo that worked - he never removed that lens again. Which may explain the picture of him at Woodstock with 6 M bodies strapped to HIS body. I'm not quite that extreme, but when I buy a lens longer than 50mm for an M, I always test the actual unit in question to make sure it is well-matched to my body(s). Link to post Share on other sites More sharing options...
jaapv Posted January 28, 2007 Share #16 Posted January 28, 2007 The Noctilux 1.0 is a non-aspherical lens, the Noctilux 1.2 was double-aspherical. The reason that Leica changed the Noctilux design was because at the time it was virtually impossible to produce costant quality within the tolerances on such a complicated design, especially the asperical part, making the number of rejects unacceptably high.Even then there is a considerable variation between individual lenses. It cost more to build than Leica could charge..... Link to post Share on other sites More sharing options...
fuchs Posted January 28, 2007 Share #17 Posted January 28, 2007 Eduardo: Can you provide a source for the info that the f/1 Nocti is ASPH? I believe that is not correct. Leica's Nocti .pdf does not mention aspheric surfaces (and they usually point them out in the optical cross-sections), E. Puts' lens compendium refers to the design "employing spherical surfaces" (as opposed to the hand-made aspheres in the previous f/1.2, as you mention), and the design of the lens dates to 1976, 15 years before Leica reintroduced aspheres in the first 35 f/1.4 Aspheric, and 20 years before they developed the (relatively) easier and less expensive blank-pressing technique for making aspheric surfaces. . You're right. The f/1 Noctilux is NOT an asph lens. I was thinking of the two versions of the Summilux 35mm asph and got somehow confused... I was just out of bed, before the rest of the family got up. I make the firm promise not to post anything again before a strong ristretto... :-) Sorry, Ed Link to post Share on other sites More sharing options...
stunsworth Posted January 28, 2007 Share #18 Posted January 28, 2007 Ed, you are forgiven this once, but next time we'll be round for a chat :-) Link to post Share on other sites More sharing options...
Jamie Roberts Posted January 29, 2007 Share #19 Posted January 29, 2007 Oddly enough, I'd say my hit rate wide open is about the same as it is on the DMR; though I have nothing longer or faster than the 50 1.4 Summilux, which I normally only shoot (for portraits) between f1.6 and f5.6... But when I have shot at 1.4, I've found the rangefinder easier to focus than the DMR. Is this because, as Andy was suggesting, Leica has set the lenses to be sharp wide open and the shift happens after DOF kicks in? BTW--what really used to kill me on the Canons was taking 3 shots in a burst to get one in focus on their fast lenses, even with enough light and autofocus. Hard to believe mirror slap is responsible for shake at 1/4000s.... Link to post Share on other sites More sharing options...
Jamie Roberts Posted January 29, 2007 Share #20 Posted January 29, 2007 Oddly enough, I'd say my hit rate wide open is about the same as it is on the DMR; though I have nothing longer or faster than the 50 1.4 Summilux, which I normally only shoot (for portraits) between f1.6 and f5.6... But when I have shot at 1.4, I've found the rangefinder easier to focus than the DMR. Is this because, as Andy was suggesting, Leica has set the lenses to be sharp wide open and the shift happens after DOF kicks in? BTW--what really used to kill me on the Canons was taking 3 shots in a burst to get one in focus on their fast lenses, even with enough light and autofocus. Hard to believe mirror slap is responsible for shake at 1/4000s Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.