plasticman Posted February 4, 2011 Share #21 Posted February 4, 2011 Advertisement (gone after registration) On film and with a subject like that, you'd run into the same kind of problems basically; it just would look differently. The highlights would also be burnt out and spill over in the adjacent areas—only the spills would look white, not violet, and hence appear less objectionable. It's not too hard to de-colourise or de-saturate those digital fringes in post-processing. Nothing like this would happen on film. Link to post Share on other sites More sharing options...
Advertisement Posted February 4, 2011 Posted February 4, 2011 Hi plasticman, Take a look here Is this an inordinate amount of CA?. I'm sure you'll find what you were looking for!
mjh Posted February 4, 2011 Share #22 Posted February 4, 2011 Nothing like this would happen on film. Well of course it does, it is a well-known effect. Light gets scattered and reflected within the emulsion, resulting in blurry halos around burned highlights. Like 01af said, the main difference is that on film these fringes are white (or actually black since it is a negative), rather than the purple we see in digital images. Link to post Share on other sites More sharing options...
AlanG Posted February 4, 2011 Share #23 Posted February 4, 2011 Nothing like this would happen on film. Well perhaps not exactly this much fringing, but purple fringing can happen on film too. (Maybe most people did not examine their film images closely enough to reveal it.) Below is an example that was shot on 6x6 film using the first model Zeiss 40mm Distagon on a Rollei 6006. It is not from the scanning process as I saw it on the film itself. But the example you posted is very extreme and would warrant a comparison of this lens vs. others on the same sensor. 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! Link to post Share on other sites Simply register for free here – We are always happy to welcome new members! ' data-webShareUrl='https://www.l-camera-forum.com/topic/143264-is-this-an-inordinate-amount-of-ca/?do=findComment&comment=1578329'>More sharing options...
Luke_Miller Posted February 4, 2011 Share #24 Posted February 4, 2011 Having shot the last year and a half with this lens on an M6 with B&W, I can't say I've seen anything like it, neither with any Canon L or Nikon lens I've ever shot with: I can assure you that had I taken this shot with one of my Nikon DSLRs using a fast lens and same exposure I would have the same results. Link to post Share on other sites More sharing options...
pico Posted February 4, 2011 Share #25 Posted February 4, 2011 Something else is going on, too. It appears that the software is repeating the vertical highlights. Patterns like that concern me. Link to post Share on other sites More sharing options...
adan Posted February 4, 2011 Share #26 Posted February 4, 2011 Nothing like this would happen on film. Never heard of halation, have we? http://www.sapiens.itgo.com/documents/foto/photo/images/halation.jpg BTW, the reason these effects go purple-blue in color images is that the photons at the blue-violet-ultraviolet end of the spectrum are the most energetic and thus bounce the hardest and furthest, and also retain more energy for exposure even after spreading. The red-yellow end of the spectrum is less energetic (which is why safelights work). _______________ @ pico: ??????? Link to post Share on other sites More sharing options...
AlanG Posted February 4, 2011 Share #27 Posted February 4, 2011 Advertisement (gone after registration) Never heard of halation, have we? BTW, the reason these effects go purple-blue in color images is that the photons at the blue-violet-ultraviolet end of the spectrum are the most energetic and thus bounce the hardest and furthest, and also retain more energy for exposure even after spreading. The red-yellow end of the spectrum is less energetic (which is why safelights work). So what would fast lenses have to do with this? By the way...I've shot many many night shots and have not encountered this problem on Canon DSLR cameras with any of my lenses. Even if this somehow is a common occurence on the M9, one can certainly not say that you have to be "stuck" with purple fringing when shooting these kinds of highlights on digital cameras. Of course specific lenses can have issues with how they depict highlights, but that has nothing to do with halation which was a characteristic of film - light going through the emulsion and bouncing back off the pressure plate or backing paper causing a glow. Anti halation coatings eliminated much of this. I've noticed that some lenses make much more of a star pattern than others - with longer "trails," when shooting specular highlights. But I don't get purple fringing around the highlights. These were shot with w/a lenses and are 100% crops. 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! Link to post Share on other sites Simply register for free here – We are always happy to welcome new members! ' data-webShareUrl='https://www.l-camera-forum.com/topic/143264-is-this-an-inordinate-amount-of-ca/?do=findComment&comment=1578440'>More sharing options...
Washington Posted February 4, 2011 Share #28 Posted February 4, 2011 Andy, darn good point! I had forgotten all about that as I used to run into in the darkroom way back when I did Cibachromes. Link to post Share on other sites More sharing options...
plasticman Posted February 4, 2011 Share #29 Posted February 4, 2011 Interesting to see everyone beating their own personal drum. Film would definitely not have given this particular type of abrupt and ugly blowout with distressingly ugly color fringing. Film halation can be very beautiful - and the transitions from blown highlights are handled with grace, which can't be said for sensor blooming that we can see here. Incidentally, the (very) oft-repeated canard that film users don't look so closely at their images as digital users doesn't apply anymore in reality, when a hybrid workflow includes scans (like mine from the LS9000) that are larger than the M9's image. Still, if it makes some of you feel better to say that sensor bloom also occurred on film (laughing) or that the problem is worse on the M9 than on a Canon, or vice versa, so be it. Whatever makes you happy eh. Link to post Share on other sites More sharing options...
AlanG Posted February 4, 2011 Share #30 Posted February 4, 2011 Still, if it makes some of you feel better to say that sensor bloom also occurred on film (laughing) or that the problem is worse on the M9 than on a Canon, or vice versa, so be it. Whatever makes you happy eh. Halation hasn't been a significant issue on film for a very long time. And while sensors can bloom, this also seems pretty well controlled on my cameras as my posted examples illustrate. I've shot the same types of dusk and night scenes countless times on transparency film. I certainly had much more trouble holding highlights (along with shadow detail) on transparency film than I do with digital. So while you may feel I am "beating my own drum," my opinions are based on many years of commercial architectural shooting and careful comparisons of the results. Why don't you show some controlled tests that compare slide film with various digital capture systems by shooting night shots that contain bright lights and shadow detail? When purple fringing occurs on film, it is a lens issue and not a film issue. Link to post Share on other sites More sharing options...
mjh Posted February 4, 2011 Share #31 Posted February 4, 2011 if it makes some of you feel better to say that sensor bloom also occurred on film Now that wouldn’t make any sense at all. Purple fringing has nothing to do with blooming. Link to post Share on other sites More sharing options...
mjh Posted February 4, 2011 Share #32 Posted February 4, 2011 Below is an example that was shot on 6x6 film using the first model Zeiss 40mm Distagon on a Rollei 6006. To me this looks more like lateral chromatic aberration. The violet fringes on one side seem to correspond to green fringes on the other (which can easily be overlooked since green fringes around green leaves don’t exactly stick out). Link to post Share on other sites More sharing options...
AlanG Posted February 4, 2011 Share #33 Posted February 4, 2011 To me this looks more like lateral chromatic aberration. The violet fringes on one side seem to correspond to green fringes on the other (which can easily be overlooked since green fringes around green leaves don’t exactly stick out). That is right. In this case, the c/a produces purple fringe. But c/a is often red/green. The old Zeiss 40mm was probably not such a great lens. That may be one reason why they replaced it. Likewise one can get purple fringing in digital systems due to c/a without the sensor being a factor either. It seems the OP's example shows an issue with the sensor. Link to post Share on other sites More sharing options...
mjh Posted February 4, 2011 Share #34 Posted February 4, 2011 I've noticed that some lenses make much more of a star pattern than others - with longer "trails," when shooting specular highlights. That would depend on the shape of the aperture blades as the star pattern is caused by diffraction at the aperture. If the aperture opening forms a proper circle, there will be no spikes around specular highlights, just some blur due to diffraction. A polygonal shape will produce as many spikes as there are aperture blades if that number is even; if it is odd there will be twice as many spikes. Link to post Share on other sites More sharing options...
mjh Posted February 4, 2011 Share #35 Posted February 4, 2011 That is right. In this case, the c/a produces purple fringe. But c/a is often red/green. Yes, it can be any pair of (roughly) complementary colours – red/green, purple/green, blue/yellow …. But if the fringes are all the same colour (such as purple), it cannot possibly be lateral CA. Link to post Share on other sites More sharing options...
plasticman Posted February 4, 2011 Share #36 Posted February 4, 2011 Now that wouldn’t make any sense at all. Purple fringing has nothing to do with blooming. Why do I often get the feeling that a discussion is going round in aimless circles, chasing its own tail? To me, the original image looks like a classic case of a combination of a few of the faults inherent in digital images: over-exposure to avoid noise and banding, with consequential blooming that accentuates the purple fringing. Now I'm wondering why I had to defend halation? (an effect that many film photographers use for artistic reasons. I look forward to these sensor shortcomings also being used to the same aesthetic purpose) Anyway, I will definitely be doing some extensive night-time comparison shooting to prove my point. Or I may possibly save that for another life. Link to post Share on other sites More sharing options...
AlanG Posted February 4, 2011 Share #37 Posted February 4, 2011 Why do I often get the feeling that a discussion is going round in aimless circles, chasing its own tail? I don't know why you feel this way. But it seems that some of us have reached the conclusion that the sensor is at fault in this case as we have discounted lens faults. I believe that other digital sensors could have done a better job on this type of subject. Perhaps film would have also done a better job with this image. What does that have to do with the question other than you "beating your drum" that film is the way to go? If you don't do the testing, how will you ever know? Link to post Share on other sites More sharing options...
AlanG Posted February 4, 2011 Share #38 Posted February 4, 2011 That would depend on the shape of the aperture blades as the star pattern is caused by diffraction at the aperture. If the aperture opening forms a proper circle, there will be no spikes around specular highlights, just some blur due to diffraction. A polygonal shape will produce as many spikes as there are aperture blades if that number is even; if it is odd there will be twice as many spikes. It isn't just the shape of the blade and the number of points of the "star." It is how far the trails extend into the scene. I tested the Canon 16-35 series 1 and series II lenses and did not buy the series II model because it produces very elongated star trails. Our favorite reviewer, Ken Rockwell, even pointed this out in his review, "Diaphragm: 7 blade, stopping down to f/22. It's round until f/4 and heptagonal from f/8. It gives great 14-pointed sunstars." The old lens had 7 blades too but perhaps forms a slightly different shape when stopped down. But I really don't think that was the issue. Both lenses had similar star patterns when stopped down, it is just that the star extended much further on the newer lens. So I believe something else is going on due to the design of the lens, other than the shape of the blades and their quantity. I don't think reviewers typically look for this but it important to me as when I shoot interiors that have small or point light sources, I really don't generally want to exaggerate any star pattern. Link to post Share on other sites More sharing options...
vonheim Posted February 4, 2011 Share #39 Posted February 4, 2011 I have seen this in varying degrees with my 50mm summilux pre-asph. However, I have also seen a big difference in this effect between opening a raw file in Aperture vs. Lightroom. In Aperture it feels "incurable" if there is a bit of red or purple in the light. In Lightroom such a raw file looks much better right when I open it. Just thought it worth mentioning. Cheers, Knut Link to post Share on other sites More sharing options...
vanhulsenbeek Posted February 5, 2011 Share #40 Posted February 5, 2011 I believe that other digital sensors could have done a better job on this type of subject. This Forum would be awash with doubt when someone would believe the opposite (any sensor fault applying) Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.