 |
 | Home | Contact Us | FAQ | Search & Site Map | Link to Us |
 | Sign In | Join | Other 45 Sites in Network |
| PhotoKB HomeDiscussion Groups Digital Photography Digital PhotoDSLR CamerasZLR CamerasPoint & Shoot CamerasFilm Photography 35 mmLarge FormatMedium formatDarkroomFilm and LabsOther EquipmentPhoto Technique Nature PhotographyPeople PhotographyTechnique GeneralGeneral Photo Topics General TopicsAustralian PhotographyUK PhotographyDirectoryPhoto Clubs |
Photo Forum / General Photo Topics / General Topics / November 2008Film versus digital sensor lenses question
Some years ago the manufacturers made statements to the effect that lenses for digital cameras needed to be designed differently than lenses for film due to the shape of the sensors. In particular, something in the sensor design, either the relative thickness of the sensing elements compared to emulsion thickness or the presence of micro lens, meant that the lens had to be designed differently. What in particular was the design parameter that had to be changed to improve performance? Does this still hold true today, when there are typically 10 million pixels instead of 1 million pixels? Thanks > Some years ago the manufacturers made statements to the effect that > lenses for digital cameras needed to be designed differently than [quoted text clipped - 13 lines] > > Thanks Digital SLR cameras use lenses that go back years before digital sensors and they work better than any point and shoot designed with special lenses. So I think its all rubbish? A good lense is still a good lense whatever the recording medium. > Some years ago the manufacturers made statements to the effect that > lenses for digital cameras needed to be designed differently than [quoted text clipped - 11 lines] > Does this still hold true today, when there are typically 10 million > pixels instead of 1 million pixels? There are two main aspects of lens design which have needed some rethinking to get the best results with digital sensors. The first is telecentricity, which is basically how close to right angles the rays of light from the back of the lens meet the sensor. Because due to the physical construction, the actual light sensitive photodiodes are effectively at the bottom of a well, light arriving at a shallower angle is partly masked. The micro lenses attempt to address this problem, but are not entirely effective. Another technique used to deal with this issue is the image processing, which can boost the brightness of pixels towards the corners of the frame, according to a map of the characteristics of that particular lens and sensor combination - I believe Nikon have started doing this with some of their recent top of the range DSLRs. The other issue is down the highly reflective surface of the sensor, which can cause ghosting and flare due to light bouncing back and forth between the lens' rear element and the sensor. This is addressed by anti reflective coatings to the rear element, and baffles. The number of pixels is largely irrelevant to the problems, although fewer, and hence larger for the same sensor size, pixels are likely to suffer less from the light falloff in the corners effect.  SignatureAlex Monro Exeter, UK Running on Linux (Kubuntu 7.1) >> Some years ago the manufacturers made statements to the effect that >> lenses for digital cameras needed to be designed differently than [quoted text clipped - 26 lines] > combination - I believe Nikon have started doing this with some of their > recent top of the range DSLRs. [...] And Leica have produced a digital version of the M series rangefinder camera, with the micro-lenses on the sensor off-set to match the direction from which light from rangefinder wide-angle lenses hits the sensor. Whereas wide-angle lenses for SLR cameras had to be built with the distance from the back of the lens to the film large enough to accomodate the mirror, that constraint doesn't apply to lenses for rangefinder cameras - so some of them get very close indeed to the film, with the light from them at a very shallow angle to the sensitive surface. So SLR wide-angle lenses are less affected by the characteristics of digital sensors than the lenses for range-finder models are. I think that the best results will eventually be obtained by having the optical and sensor parts of the system as a single unit, with the camera bodies containing only the controls, power, and storage. But then, I don't like SLRs. Signature-- ^^^^^^^^^^ -- Whiskers -- ~~~~~~~~~~ > > Some years ago the manufacturers made statements to the effect that > > lenses for digital cameras needed to be designed differently than [quoted text clipped - 20 lines] > photodiodes are effectively at the bottom of a well, light arriving > at a shallower angle is partly masked. This was the reason that I thought I remembered, but I didn't mention it since I didn't want to put incorrect ideas into other people's minds. > The micro lenses attempt to > address this problem, but are not entirely effective. I thought the reason for the micro lenses was to capture the light from above the non sensing parts of each cell, but certainly the micro lenses can address more than one problem. > Another technique > used to deal with this issue is the image processing, which can boost > the brightness of pixels towards the corners of the frame, according > to a map of the characteristics of that particular lens and sensor > combination - I believe Nikon have started doing this with some of their > recent top of the range DSLRs. I didn't realize that this was the case. This seems to mean that a lens review that measures a lens' response using accurate measuring equipment would give accurate but useless information since the corner fall off would normally be corrected by the camera or RAW processing profile for the lens. The flip side is that a solely optical lens design that provided for increased output on the edges would have poorer performance when used with film. > The other issue is down the highly reflective surface of the sensor, > which can cause ghosting and flare due to light bouncing back and forth > between the lens' rear element and the sensor. This is addressed by > anti reflective coatings to the rear element, and baffles. I thought that this issue was already addressed: there were film cameras that got light readings by reflection from the film so the issue was known. > The number of pixels is largely irrelevant to the problems, although > fewer, and hence larger for the same sensor size, pixels are likely to > suffer less from the light falloff in the corners effect. I guess my final question comes down to: How does the "thickness" of the part of the IC in front of the sensor scale with pixel size? (If it remains fixed than the effect becomes more pronounced with smaller pixels, if it scales with the other two dimensions then the effect remains constant.) Mark F wrote,on my timestamp of 14/10/2008 8:23 AM: > Some years ago the manufacturers made statements to the effect that > lenses for digital cameras needed to be designed differently than > lenses for film due to the shape of the sensors. Sure... > What in particular was the design parameter that had to be changed > to improve performance? The depth and capacity of your wallet. > Does this still hold true today, when there are typically 10 million > pixels instead of 1 million pixels? You bet. And wait for someone to claim you'll need all new lenses for the new 20 megapixel camera generation! > Some years ago the manufacturers made statements to the effect that > lenses for digital cameras needed to be designed differently than [quoted text clipped - 11 lines] > Does this still hold true today, when there are typically 10 million > pixels instead of 1 million pixels? Another thing I haven't seen mentioned in this thread yet is some older lenses show more chromatic aberration on digital. Presumably this is also a microlens issue. SignaturePaul Furman www.edgehill.net www.baynatives.com all google groups messages filtered due to spam >> Some years ago the manufacturers made statements to the effect that >> lenses for digital cameras needed to be designed differently than [quoted text clipped - 15 lines] > lenses show more chromatic aberration on digital. Presumably this is also > a microlens issue. Well on a D300 Nikon or a D700 for that matter as well as the D3 there is almost zero Crom. Aberat. on JPGs as the camera removes it as it processes the image. On a raw unprocessed file from these cameras there is some visible on some lenses both new and old (pre digital) but with good glass its pretty insignificant in the vast majority of images. It does not seem to be any worse on either film or digital with new or old lenses. In fact the sharpest lens I own on a digital camera is from the pre digital days. True of many of Nikons older full frame lenses. I dont think its an issue. Burgerman wrote,on my timestamp of 17/11/2008 9:15 AM: >> Another thing I haven't seen mentioned in this thread yet is some >> older lenses show more chromatic aberration on digital. Presumably [quoted text clipped - 12 lines] > the pre digital days. True of many of Nikons older full frame lenses. I > dont think its an issue. It never was an issue. It was always a fabrication from lens makers to convince the gullible they "needed" new lenses to go with the new dslr. In fact, what you'll find is that some lenses made in theory to take into account "digital" cameras will actually perform poorly with full frame sensors. Case in point, the current Nikkor 70-200/2.8 AF-S VR. Of course, what you said about software post-processing correction still applies. >> Some years ago the manufacturers made statements to the effect that >> lenses for digital cameras needed to be designed differently than [quoted text clipped - 8 lines] >> What in particular was the design parameter that had to be changed >> to improve performance? The problem was/is real enough; it is to do with the anti-reflection coating on the rear lens element. Film has a matt surface so there are no direct reflections from the film. Digital sensors - or at least the AA filter in front of the sensor is glass, and has direct reflections accordingly. The AA filter reflects light back to the lens, and with some lens designs, if the lens is not adequately coated, the light will be reflected back again to the sensor, causing ghosting and hot spots in the image. Not all lenses do this, at least not to the same extent. It depends mainly on the amount of curvature of the rear element outer face. Near-flat designs are worse than more convex lenses. Whatever, digital lenses have high-performance anti-reflective coatings on the rear element, whereas a lot of film lenses have minimal, or no coating on the rear element. >> Does this still hold true today, when there are typically 10 million >> pixels instead of 1 million pixels? Yes, it's nothing to do with pixels. > Another thing I haven't seen mentioned in this thread yet is some older > lenses show more chromatic aberration on digital. Presumably this is > also a microlens issue. CA from a lens is independent of the film or sensor, and a micro lens is a device to at least partly cancel the obliqueness of light rays that are off-axis. Each microlens is matched to a single pixel, and as each pixel is filtered to respond to only one primary color, CA cannot occur with microlenses Perhaps the better response of sensors to color separation may increase the apparent CA in terms of contrast. Colin D. >> Some years ago the manufacturers made statements to the effect that >> lenses for digital cameras needed to be designed differently than [quoted text clipped - 15 lines] > lenses show more chromatic aberration on digital. Presumably this is also > a microlens issue. This argument will aways go on sadly. Depends what you are after? Personally modern digital beats the pants of the equivilant film in terms of quality and believe it or not in my experience recently, resolution. |
Reply to this Message |
 Sign In
 Join
 My Latest Posts My Monitored Threads My Blog My Photo Gallery My Profile My Homepage Start New Thread Enable EMail Alerts Rate this Thread
|
©2010 Advenet LLC Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.