jaapv Posted February 15, 2012 Share #61 Â Posted February 15, 2012 Advertisement (gone after registration) I mean, there's the artistic attraction to false detail, but does not having an AA filter offer any tech advantages like better high incident angle correction? Yes it does -as an AA filter is basically a piece of matte glass in front of your sensor, it will reduce the nominal resolution by about 30%. It is not just aliasing - it is real detail that gets revealed by removing it. Link to post Share on other sites More sharing options...
Advertisement Posted February 15, 2012 Posted February 15, 2012 Hi jaapv, Take a look here Help needed: real world M9 Moire problems. I'm sure you'll find what you were looking for!
nikonuser2012 Posted February 15, 2012 Author Share #62 Â Posted February 15, 2012 Yes it does -as an AA filter is basically a piece of matte glass in front of your sensor, it will reduce the nominal resolution by about 30%. It is not just aliasing - it is real detail that gets revealed by removing it. Â 30%, are you sure? Also, I thought that not all AA filters were created equal, some stronger than others, so putting a value on it is not that easy. Â How is this measured and do you have any info? Link to post Share on other sites More sharing options...
jaapv Posted February 15, 2012 Share #63 Â Posted February 15, 2012 I'm trying to remember where I found that Measuring is easy - just compare a camera with its filter to the same camera with the filter removed. Yes - obviously the strength (read opaqueness ) of the AA filter would influence that number. Link to post Share on other sites More sharing options...
nikonuser2012 Posted February 15, 2012 Author Share #64 Â Posted February 15, 2012 I'm trying to remember where I found that Measuring is easy - just compare a camera with its filter to the same camera with the filter removed. Yes - obviously the strength (read opaqueness ) of the AA filter would influence that number. Â Measure what exactly? What value do you measure for objective results? Link to post Share on other sites More sharing options...
Sp12 Posted February 16, 2012 Share #65 Â Posted February 16, 2012 Yes it does -as an AA filter is basically a piece of matte glass in front of your sensor, it will reduce the nominal resolution by about 30%. It is not just aliasing - it is real detail that gets revealed by removing it. Â From an information theory standpoint that's incorrect (the AA filtered camera will always provide a more accurate rendition of a scene), but I was referring to any advantages in other areas like a thinner optical block making for more accurate incident angle correction. Link to post Share on other sites More sharing options...
Erik Gunst Lund Posted February 16, 2012 Share #66 Â Posted February 16, 2012 The D800E has got a low-pass filter (AA filter) it's just another type of filter; there are two filters, the first correct the light in one direction the second correct it back to the same path. Â Normal low-pass filters are also a pair for normal RGB Bayer pattern sensors, first filter split the light vertical and the second split horizontal to avoid color moire. Link to post Share on other sites More sharing options...
jaapv Posted February 16, 2012 Share #67 Â Posted February 16, 2012 Advertisement (gone after registration) Measure what exactly? What value do you measure for objective results? Edge contrast? Link to post Share on other sites More sharing options...
nikonuser2012 Posted February 16, 2012 Author Share #68 Â Posted February 16, 2012 Edge contrast? Â Maybe, those details are beyond me but I would like to know if anyone has experience with this. Link to post Share on other sites More sharing options...
jaapv Posted February 16, 2012 Share #69 Â Posted February 16, 2012 Whatever else, it might be a very interesting camera for Leitaxed R lenses. Link to post Share on other sites More sharing options...
macusque Posted February 16, 2012 Share #70  Posted February 16, 2012 I was always told that moiré tends to show when lenses outresolve sensor.  I guess that D800E 36 Mp 24x36mm sensor will show less moiré than Leica 18 Mp 24x36mm sensor for a given lens.  But I just saw samples at dpreview from the new little Canon G1x, which despite its almost APS-C size sensor has a pixel pitch that translates to over 45 Mp on 24x36mm and guess what? Moiré... So who knows? Link to post Share on other sites More sharing options...
jankap Posted February 16, 2012 Share #71  Posted February 16, 2012 I was always told that moiré tends to show when lenses outresolve sensor. Lenses outresolve etc. means, that the lenses perform the AA-function. The problem is, that some structures of the pictures have the same resolution as the sensor and so interfere. The bayer sensor has regular structures. Fuji has offered a sensor with a different structure. How is this with a foveon sensor? No problem, I think. Because the colors are not separated onthe surface. Jan Link to post Share on other sites More sharing options...
Erik Gunst Lund Posted February 16, 2012 Share #72  Posted February 16, 2012 ...How is this with a foveon sensor? No problem, I think. Because the colors are not separated onthe surface. Jan  The Foveon types doesn't have color moire, otherwise they have the same problems Link to post Share on other sites More sharing options...
nikonuser2012 Posted February 16, 2012 Author Share #73  Posted February 16, 2012 Lenses outresolve etc. means, that the lenses perform the AA-function. The problem is, that some structures of the pictures have the same resolution as the sensor and so interfere. The bayer sensor has regular structures. Fuji has offered a sensor with a different structure. How is this with a foveon sensor? No problem, I think. Because the colors are not separated onthe surface. Jan  Unless we are talking some terrible lenses, don't all lenses essentially outresolve the sensor? From my reading, the lens is not the limiting factor in resolution, it's the sensor. Even on the D800/E. Link to post Share on other sites More sharing options...
macusque Posted February 16, 2012 Share #74 Â Posted February 16, 2012 Not at all. There are very few lenses out there which outresolve D3x sensor. D800/800E will challenge even the best lenses, Leica included. That is not to say you won't see more resolution (compared for instance to the D700) with all but coke bottle glass, but most lenses at 36 Mp FF will actually act like an AA filter which to me is a good thing for the D800E. Link to post Share on other sites More sharing options...
haroldp Posted February 17, 2012 Share #75 Â Posted February 17, 2012 Unless we are talking some terrible lenses, don't all lenses essentially outresolve the sensor? From my reading, the lens is not the limiting factor in resolution, it's the sensor. Even on the D800/E. Â Not really, to outresolve the D800 sensor, a lens would need about 110 lp/mm, which is a very good lens, on a tripod, mirror up, or at least 1/1000 sec. for a 50mm range lens. Â The very best commercial lenses probably get to about 144 lp/mm, again under ideal conditions. Â The big difference with modern lenses is that they can get to these resolutions at wider apertures and over a wider section of the frame. At optimum apertures and in the center, some older lenses are amazingly sharp. Â Even the slightest bit out of perfect focus will take it out of that range, depth of field does not apply at this level of resolution. Â To get moire or other artifacts on a D800E we would need a repeating parren of exactly the right size at the right distance in perfect focus with a really good lens at optimum aperture. Will this happen, I am sure that it will occasionally because people using a D800E (or an M9) hopefully are using good glass and know what they are doing. I personally have had none with my M9 at 18 mpx which is more susceptible because of larger pixel pitch. Â Like many decisions it is a trade-off based on one priorities and circumstances. Â While all of my Leica lenses are very good, the 18 mpx sensor has enough resolution to show me that a 50 summilux is quite special (only one example). Â Regards ... H Link to post Share on other sites More sharing options...
01af Posted February 17, 2012 Share #76 Â Posted February 17, 2012 There are very few lenses out there which outresolve D3x sensor. Actually there are very few which don't. Â Â ... but most lenses at 36 MP FF will actually act like an AA filter ... Isn't it funny how this myth will always pop up when a new digital camera arrives on the scene that has more megapixels than the competition ... be it 8 MP, 12 MP, 24 MP, or now 36 MP? Link to post Share on other sites More sharing options...
pgk Posted February 17, 2012 Share #77 Â Posted February 17, 2012 Isn't it funny how this myth will always pop up when a new digital camera arrives on the scene that has more megapixels than the competition ... be it 8 MP, 12 MP, 24 MP, or now 36 MP? Well now that depends very much on the lenses being used and how they are used. My Canon 5D2 for example shows significant softening when stopping down from around f/11. As I shoot a lot of macro material this is problem which is solvable only by using techniques like focus stacking, which are inappropriate for anything other than still subjects. So more MPixels are simple not going to help although they will probably produce more detailed images at say f/8ish - is this useful I wonder. [in fact I still find macro files from my original 5D to be extremely good and lacking very little in comparison to the 5D2). Â Fast lenses too are problematic, in as much as their performance wide open is often (but not with all lenses - the newer Leicas and Canon's 85/1.2 L come to mind) already showing signs of stress, with images shot wide open lacking real sharpness and showing colour fringing at times. More MPixels won't help here - I could go on. Â But I suspect that in the real world many people will not notice such problems, because they are too fixated on MPixels, but in reality the 'myth' is not entirely a myth. More Mpixels may be useful to those able to extract every last nuance from their lenses and whose technique is both relevant and immaculate, but there will be many situations where the greater MPixels are not an advantage. Link to post Share on other sites More sharing options...
jaapv Posted February 17, 2012 Share #78 Â Posted February 17, 2012 I am not convinced that more Mp will solve diffraction. When you stop down to f11 the aperture gets so small that it "bends"the light, resulting in a softer image. Your camera is just recording what the lens does... Link to post Share on other sites More sharing options...
macusque Posted February 17, 2012 Share #79  Posted February 17, 2012 Actually there are very few which don't.... Isn't it funny how this myth will always pop up when a new digital camera arrives on the scene that has more megapixels than the competition ... be it 8 MP, 12 MP, 24 MP, or now 36 MP?  What is funny is people who sarcastically respond without knowing we talk about moiré here, not if more Mp resolve better, which is always true, regarding of the lens used in front of the sensor. Link to post Share on other sites More sharing options...
pedaes Posted February 17, 2012 Share #80 Â Posted February 17, 2012 Sean Reid has just posted a essay about this topic, and the advantages of not having a AA filter, on The Luminous Landscape website. Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.