Steve Ash Posted June 29, 2010 Share #21 Posted June 29, 2010 Advertisement (gone after registration) Maybe just because the sensor draws more precisely? Maybe because the raw converters calculate unprecisely? Regards Steve Link to post Share on other sites More sharing options...
Advertisement Posted June 29, 2010 Posted June 29, 2010 Hi Steve Ash, Take a look here Noctilux 095, M9 and green fringes. I'm sure you'll find what you were looking for!
mjh Posted July 1, 2010 Share #22 Posted July 1, 2010 That is the CA connection I mentioned in my post, I suppose:o. But I cannot recall seeing anything like this on film, ever, and it seems to be quite common on sensors. Maybe just because the sensor draws more precisely? That’s certainly possible. I have found a rather similar example in Dr. Hubert Nasse’s recent article on depth of field and bokeh (http://www.zeiss.de/C12567A8003B8B6F/EmbedTitelIntern/CLN_35_Bokeh_EN/$File/CLN35_Bokeh_en.pdf, originally published in Zeiss’ Camera Lens News 35). It is found on page 40 and there it is convincingly explained as the effect of chromatic aberration. Dr. Nasse specifically refers to “the reversal of the colour effects in front of and behind the focus”, exactly as in your example. Link to post Share on other sites More sharing options...
jaapv Posted July 1, 2010 Share #23 Posted July 1, 2010 That article should be a gobstopper-sticky to be used in any DOF thread... Link to post Share on other sites More sharing options...
Mauribix Posted July 1, 2010 Share #24 Posted July 1, 2010 Anyway I'm sure the OP's nocti is faulty. I'd buy it as a wreck for 500 bucks. Link to post Share on other sites More sharing options...
Elsu Posted July 1, 2010 Author Share #25 Posted July 1, 2010 hehe, I dont think I would give it the NX095 just that fast Actually asked Leica Germany the same question with the sample images. Lets see what they reply. As for Lightroom 3, I only managed to import some images scanned from slides and a couple from M9. Those from M9 still showed the green fringes, and in the Chromatic Aberration correction menu, there is no option to correct GREEN. Now the beta version expired, I need to wait till Leica provides a LR3 download before I can try again. Link to post Share on other sites More sharing options...
johnbuckley Posted July 1, 2010 Share #26 Posted July 1, 2010 This may be a broader question than one specific to the Noctilux and the M9, but would seem to fit into the thread: in addition to what may or may not be done to remove CA from images in post-processing, are there techniques that can be employed when using a fast lens like the Noctilux to prevent either longitudinal or transverse CA? Any tricks in terms of focusing relative to light sources, etc. that might reduce the incidence of CA? Link to post Share on other sites More sharing options...
jaapv Posted July 1, 2010 Share #27 Posted July 1, 2010 Advertisement (gone after registration) You can use the CA sliders in RAW conversion, or you can use a number of photoshop techniques, of which the color relacement brush is perhaps the most practical. The best solution is given by Jamie Roberts in http://www.l-camera-forum.com/leica-forum/digital-post-processing-forum/115965-fringing-technical-problem.html this thread. Link to post Share on other sites More sharing options...
johnbuckley Posted July 1, 2010 Share #28 Posted July 1, 2010 Jaap - many thanks. The question, though, was whether there was anything to be done in composing the photograph that could minimize the need to eradicate CA in post-production. Link to post Share on other sites More sharing options...
jaapv Posted July 1, 2010 Share #29 Posted July 1, 2010 Yes- avoid extreme contrasts. Link to post Share on other sites More sharing options...
marmelade Posted July 1, 2010 Share #30 Posted July 1, 2010 And, Or step down the lens.. (yes i know its "best" wide open) but the pop at smaller apertures give quite a nice feel to the images.. Seems the CA is less apparent/minimal from 2.8 and "down".. best, Are Link to post Share on other sites More sharing options...
Elsu Posted July 1, 2010 Author Share #31 Posted July 1, 2010 And, Or step down the lens.. (yes i know its "best" wide open) but the pop at smaller apertures give quite a nice feel to the images.. Seems the CA is less apparent/minimal from 2.8 and "down".. best, Are That is the general answer from Leica at least, that if you shoot from F4 upwards, you will not see such fringes. My first question is "then why pay thousand of dollars for fast lenses with 1.2, 1.4 aperture or NX095??" After studying this issue for a while, it seems a rather common issue for all fast (and expensive) lenses: Canon, Zeiss and Leica for example. It is a relief that apparently it is not particularly a fault of my NX095 to produce all those green fringes. It is definitely some extra work to run those images through Photoshop but I dont think I would slow down the lens just to avoid the fringes, simply because I got this special lens to shoot at the special aperture. I might just as well turn to B/W and make the fringes irrelevant. Link to post Share on other sites More sharing options...
Elsu Posted July 5, 2010 Author Share #32 Posted July 5, 2010 Got an answer from Leica: "{The green fringe} is a color bug. This bug is normal by high opening lenses. The bug goes smaller when you close the diaphragm. The bug goes from green to blue and is controlled from the object. The picture are normal." So I guess this is an accepted "feature" of digital photography. Link to post Share on other sites More sharing options...
hoppyman Posted July 5, 2010 Share #33 Posted July 5, 2010 Elsu, more accurately it is an optical issue which has always been present. However with the advent of digital people can more easily see it when it occurs . That is largely due to being able to magnify images down to individual pixel level. I have read that described as looking at a 2 metre wide print from about 30cm away. Of course there are many other lens characteristics that are more apparent when the images are viewed at very high magnification as well. I think also worth noting that no lens provides the best possible quality at very large apertures. Fair to say though that modern Leica lenses are very much better than most in that respect. Link to post Share on other sites More sharing options...
marmelade Posted July 6, 2010 Share #34 Posted July 6, 2010 Elsu, more accurately it is an optical issue which has always been present. However with the advent of digital people can more easily see it when it occurs . That is largely due to being able to magnify images down to individual pixel level. I have read that described as looking at a 2 metre wide print from about 30cm away.Of course there are many other lens characteristics that are more apparent when the images are viewed at very high magnification as well. I think also worth noting that no lens provides the best possible quality at very large apertures. Fair to say though that modern Leica lenses are very much better than most in that respect. Hmmm 2 metre wide print from about 30 cm... that may be a slight exaggeration in this case i would say.. Have you shot with the NX0.95 ? Included here is an image not cropped at all in height, only in width due to client content, but you quite clearly see the CA i would presume.. so even though if we now can zoom in the CA is also apparent sometimes without it.. Welcome, dear visitor! As registered member you'd see an image here… Simply register for free here – We are always happy to welcome new members! Discussing focus on the other hand, may suit the 2 metre saying better Link to post Share on other sites Simply register for free here – We are always happy to welcome new members! Discussing focus on the other hand, may suit the 2 metre saying better ' data-webShareUrl='https://www.l-camera-forum.com/topic/124754-noctilux-095-m9-and-green-fringes/?do=findComment&comment=1371304'>More sharing options...
hoppyman Posted July 6, 2010 Share #35 Posted July 6, 2010 Marmelade people can look at higher magnification more readily on their computer screen than on a more common print size which is not reproduced at your screen resolution either. I just calculated for you and an M9 file at my standard monitor resolution of 93 pixels per inch would require a monitor just over 1.5 metres wide to display the entire image at 100%. Not much exaggeration involved I did look at the earlier examples and made comment amongst others regarding some of the conditions that may make those aberrations more apparent. Just my suggestions and opinions in the discussion. Maybe you could post some examples of focussed areas from some files if someone wanted to look or offer comparisons or other opinions for you.The ultimate check naturally is to send your lens for testing if you are not satisfied. Link to post Share on other sites More sharing options...
mjh Posted July 6, 2010 Share #36 Posted July 6, 2010 Included here is an image not cropped at all in height, only in width due to client content, but you quite clearly see the CA i would presume.. But is it CA? I am not so sure in this case. When there is turquoise in an area where unsharp pictures of blue sky and green leaves intermingle, isn’t that to be expected, even from a perfect lens? Link to post Share on other sites More sharing options...
marmelade Posted July 6, 2010 Share #37 Posted July 6, 2010 Hello there! Good question, ive almost similiar shoot except the DOF from a Canon EOS 5D2 without this artifact and here there wasnt used the most "exotic" glass either.. What do you suggest it may be ? (thanks for your 2 cents, or 4) best, Link to post Share on other sites More sharing options...
adan Posted July 6, 2010 Share #38 Posted July 6, 2010 Color fringes in out-of-focus areas are a natural effect of how lenses work - just like blurriness in OOF areas. Lens are designed on the basis of how they perform when FOCUSED - that's what designers study when computing the paths of the light rays. If all the colors focus at the same point (no Longitudinal CA)** with an image the same size (no Lateral CA) - then the lens has (technically) no CA. But - purple (red + blue) and green light rays may not track together in front of or behind the plane of focus (at the film or sensor). In which case, if they meet at the focus point, they must also cross over at the focus point, leading to purple fringes in front of or behind the focus plane and green fringes behind or in front of the focus plane. The fringes are "inside" and "outside" the blurred object or edge - so one or the other may not be very visible. Usually it is the outside fringe that is more visible. Bokeh works in the same asymmetrical way - a lens that produces smooth backgrounds will produce "busy" foregrounds. A lens that produces poor background bokeh will be smoother for OOF foregrounds. _______________________________________________ ** BTW, this is why Longitudinal CA is very hard to correct - one of the colors (usually red) is actually producing a fuzzy image where the others are sharp. The blur circle of the fuzzy color puts a colored rim around sharp objects, and there just is no way to really make a blurred image sharp after the fact. Link to post Share on other sites More sharing options...
zlatkob Posted July 6, 2010 Share #39 Posted July 6, 2010 Color fringes in out-of-focus areas are a natural effect of how lenses work - just like blurriness in OOF areas. Andy, that explanation makes a lot of sense. I imagine that if one is shooting at or near f/0.95, it is to use that lens's special qualities, including super-shallow depth of field and amazing light-gathering. If one needs very accurate reproduction, then one is shooting at or near f/8, probably with a less expensive lens. Link to post Share on other sites More sharing options...
plasticman Posted July 8, 2010 Share #40 Posted July 8, 2010 Even though I have Hoppyman on my ignore list, I saw this passage quoted above: Originally Posted by hoppyman Elsu, more accurately it is an optical issue which has always been present. However with the advent of digital people can more easily see it when it occurs . That is largely due to being able to magnify images down to individual pixel level. Naturally, film images can also be seen 'at the pixel level' (I'm speaking metaphorically - I don't need anyone to point-out that film doesn't have pixels, as happened the last time I discussed scanned film images). I have the Noctilux f1, and don't see these sorts of artefacts on film, even when scanning at 4000dpi (which gives a slightly larger final image than the M9 sensor). In any case, film treats the abrupt transition between highly contrasting areas with far more grace than digital. But that is a whole other discussion.* * On second thoughts, it probably isn't "a whole other discussion" - I think it's this very graceful transition that probably masks the unwanted effects of chromatic aberration or color fringes on film. What's more, the physical depth of film, and the layering of the color grains probably also plays its part in softening an effect that is glaringly obvious from a digital sensor. Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.