Mako2011 wrote:Targets like this are easy:Are you sure?They may be easy, but are they always good?The AF array needs detail (assuming a cross type) along a horizontal or a vertical line - which this suggestion does not have.Thom Hogan, rightly in my experience, advises not using star wheels.It is about 7 years since the number 1 tech guy at Canon USA said "Chuck Westfall “I recommend using a flat, detailed target parallel to the focal plane. (snipped)  The nature of the AF sensors used by EOS digital SLR’s as well as those from other manufacturers is that they perform most reliably when the entire length of the focusing area sees readable detail”.Radiating lines usually do not contain good readable detail for consistent accurate AF when the camera is centred on the hole in the middleIf not centred the AF can assume there is a focus correction to make

Mako2011 wrote:1st... PDAF sensors are sensitive to red green and blue channels differently.Are you sure?Neither AF sensors, or camera pixels directly detect colour!Nikon camera sensor pixels have a red, a green, or a blue filter above each of them. The camera compares the individual filtered effect plus light density and colour at around 40 different surrounding pixels before assigning a density and accurate colour code to each pixel site. From this, in 14 bit RAW capture, the camera can record millions of different colours.AF works faster and in lower light than matrix metering by dispensing with the colour filters to let more light through. AF does not see red, green or blue effects created by filters as part of the colour process.Digressing slightly Leica made a B&W digital camera which is superior in resolution etc to RGB type sensors of the same MP which also record colour. The "old" yellow, green, orange and red B&W filters can be used in front of the lens to vary B&W contrast between different colours in the scene using this special Leica body.A single colour light source may have low contrast or a none continuous light output which reduces the AF sensor efficiency. Even so as a general rule as AF sensors do not use separate red, green or blue channels the colour of the light source (as distinct from subject contrast) should not affect AF performance.

Leonard Shepherd wrote:Mako2011 wrote:Targets like this are easy:Are you sure?They may be easy, but are they always good?Yes, when used as outlined in the AF testing technique in the link I provided.The AF array needs detail (assuming a cross type) along a horizontal or a vertical line - which this suggestion does not have.Not true. If that was absolutely true, you would find PDAF could not lock on to this target. Try it and you will see that is not the case. There are more optimal targets though. This one was simply used as a far better examaple than the OP's original. The type used by the Lensalign system would be a even better choice.

Leonard Shepherd wrote:Mako2011 wrote:1st... PDAF sensors are sensitive to red green and blue channels differently.Are you sure?YesNeither AF sensors, or camera pixels directly detect colour!True, but I never said they did. I said they can be affected by color. As shown in the article I linked, color can affect focus in certain situations. That's why it's not surprising OP got consistent but different results with a red vs a blue low contrast target.

Leonard Shepherd wrote:liwei zhen wrote:It is about 7 years since the number 1 tech guy at Canon USA said "Chuck Westfall “I recommend using a flat, detailed target parallel to the focal plane. After reading through the PDF linked from your message, it appears that the author has missed a major point, i.e., any individual focusing point in a digital SLR is much longer than the simple line he is using on his chart. The nature of the AF sensors used by EOS digital SLR’s as well as those from other manufacturers is that they perform most reliably when the entire length of the focusing area sees readable detail. This condition is not satisfied by a thin line on a piece of paper. It's OK to include an angled chart in a test photo. In fact, Canon Factory Service Centers always do this. But the test target is always separate from the angled chart, and parallel to the camera's focal plane”.A thin line scan can produce spurious resultsIt never ceases to amaze me why some Nikon users prefer to believe in fairy tales for testing AF instead of following the guidance from Nikon and Canon, and then holler without taking sensible tests to find out if the problem is them or the equipment.Hi, don't really understand, could be my poor English. But still could you help me by pointing out how the focus point (really a frame) is longer than what side of the line, and how the line be too thin compare to the focus frame. Because I sure believe less in mysteries like "you should follow an abstract rule", than "analyze the individual instance":In this instance I think the vertical scan line is given good contrast twice, aside from Mako's theory about the poor contrast between blue/red and white, which I don't totally learn.To myself, the constantly reproducible shift difference of the different color in different light, with different camera/lens makes me believe that it is the color that caused the problem, but not the pattern.

Mako2011 wrote:liwei zhen wrote:Mako2011 wrote:1st... PDAF sensors are sensitive to red green and blue channels differently. That is normal and your tests show it. It's normal2nd... your targets are poor target for PDAF testing as there really is poor areas of contrast (white to blue/border). This is reflected in the way results differ as you rotate the targets. Again, normal and to be expected when using poor contrast target for PDAF testing.3rd... There is a AF-sub mirror behind the main mirror that you can not see. Slight variations in it's calibration can affect focus in different light and color channels. That's normal4th... dirt/debris on the AF unit at the bottom of the mirror box can affect focus and cause inconsistencies in different light and color channels.5th... Lens design can affect focus. The 35mm f1.8g is often one that brings focus "issues" up.Based on your testing to date...I don't think you have an issue that requires service.See link below for a better explanation.affect of color on PDAFThanks, the color/light condition affecting PDAF thing is nice to know. Still I wonder if it's common in DSLR world, that they may affect the focus that much, like the one shown with the eye of my boy, at f4.5.Yes, a known and common phenomenon.I don't believe the issue of the OP's focus errors on an eye has much to do with the effect of color on the PDAF viewfinder focusing. The spectrum is diverse enough in the example photo of the baby's portrait that I doubt it played much role in the misfocus.OP: Nailing the focus on the leading eye with a large aperture can be difficult. In this case you were closed down to f/4.5 so the DOF was wider. If you get consistent errors at this setting then something is wrong with the equipment or your technique. When I use f/2.0 and try to get the leading eye in focus I expect to get some misses, but I get a lot of well focused leading eyes as well. There's always the chance the subject or camera moved between focus aquisition and shutter. Sometimes you don't target the eye steadily during the focus aquisition phase. But if you are reasonably skilled and deliberate you should get several images that are properly focused.Or if it's normal for 35 18g, then maybe I should just stop using the lens. But in the situation, it may be silly to buy another lens.Again, you're using targets and conditions known to cause focus inconsistencies. In some situations it's best to focus on a good spot/target then recompose.OP: The saturated blue and red targets are extreme cases. The blue might provide marginal contrast but if the focus error is consistent it's not a contrast issue. I used a few pastel blue Post-It notes on a white sheet of paper. The paper I used provided even less contrast that the blue target you used. Most of the time the D300 and 35 F/1.8 focused well. But a few times it missed. The camera was on a tripod and lighting was about a stop darker than your scenario. (Thanks for posting the new photos with all the metadata intact.)And if your point 3-4 are right, I do need some service for the camera.Not really, I said those issues can affect focus. Simply check the AF unit for dust...and try testing using better targets (black and white high contrast targets) in differing lighting conditions...to see if AF-Sub mirror adjustment is called forI agree that the OP should go back to a good B&W high contast target.OP: The AF color dependency issue is a very interesting topic. But I think you need to make sure that your camera can accurately focus on a good B&W target -- or at least one that doesn't have pure red or blue saturated color. The few images that you provide with a B&W target suggest the camera or lens needs calibration. If it's adjusted correctly the focus should be centered on the scale so that other variations like color won't cause an even greater error.Use a target with bold and irregular width B&W lines. The targets I recommend are shown in my recommended AF test technique:http://www.pbase.com/slockhart54/auto_focus_test_procedureYou should change the target's orientation to see if you can reproduce the different focusing error that you encountered with the orientation change of blue target. With the center cross-type AF points you should get close to the same focus solution whether the lines run vertically or horizontally. If you get the same large variation as illustrated in your supplied photos then that's another indication something is wrong with the camera or lens.Besides, for the bad tweak they gave me alone, I almost cannot take any in focus picture now.Not using the targets you're currently showing us...and that is expected. Your blue and red targets are very close to trying to focus on a blank white wall (as far as what the Arrays are seeing) and saying you have focus issues because it can't lock or is inconsistent when it does. That's expected and doesn't necessarily point to a focus problem. Same with your red and blue targets.I don't think the contrast is that low based on my test, but it's good to eliminate any doubt. The AF arrays should be sensitive to anything in the visible spectrum, but its typical for the sensitivity to start dropping in the blue end of the spectrum.Now try a newspaper on a wall compared to Liveview (in bright light). That's a better test to see if you actually do have a focus issue.As for the rotation thing: I tested later with vertical direction, and even 45 degree rotation with the blue paper, so long as I put the focus frame in the blue area, the behaviors are identical to the horizontal positioning.Again, the focus arrays have a vertical and horizontal component that looks for a contrast along the array. You are showing them a single tiny contrast (white to blue/red border) and not understanding how poor a focus contrast that is. Like trying to focus on a tiny black spot on a white wall. Tiny black spots are hard to focus on and that's expected.Targets like this are easy:And why they are used to actually see if there might be a need for service due to focus issues. You're, in a sense, asking the dealer to fix your car because it seems to pull to the left when you drive on the side of a hill. Not a perfect analogy but kind of demonstrates the issue.I often disagree with Leonard but I agree with him that this target could present a problem if the AF sensor happened to see just the center where the line spacing is extremely fine. In most cases the aim will be off enough that the wider lines will be seen by the AF detector and it will focus fine. But there's no point in introducing such a problematic condition. My eyes don't even like looking at the center of that target.Steve

SteveL54 wrote:Mako2011 wrote:liwei zhen wrote:Mako2011 wrote:1st... PDAF sensors are sensitive to red green and blue channels differently. That is normal and your tests show it. It's normal2nd... your targets are poor target for PDAF testing as there really is poor areas of contrast (white to blue/border). This is reflected in the way results differ as you rotate the targets. Again, normal and to be expected when using poor contrast target for PDAF testing.3rd... There is a AF-sub mirror behind the main mirror that you can not see. Slight variations in it's calibration can affect focus in different light and color channels. That's normal4th... dirt/debris on the AF unit at the bottom of the mirror box can affect focus and cause inconsistencies in different light and color channels.5th... Lens design can affect focus. The 35mm f1.8g is often one that brings focus "issues" up.Based on your testing to date...I don't think you have an issue that requires service.See link below for a better explanation.affect of color on PDAFThanks, the color/light condition affecting PDAF thing is nice to know. Still I wonder if it's common in DSLR world, that they may affect the focus that much, like the one shown with the eye of my boy, at f4.5.Yes, a known and common phenomenon.I don't believe the issue of the OP's focus errors on an eye has much to do with the effect of color on the PDAF viewfinder focusing.Me too...I was only speaking to the red blue target differences he was seeing.Now try a newspaper on a wall compared to Liveview (in bright light). That's a better test to see if you actually do have a focus issue.As for the rotation thing: I tested later with vertical direction, and even 45 degree rotation with the blue paper, so long as I put the focus frame in the blue area, the behaviors are identical to the horizontal positioning.Again, the focus arrays have a vertical and horizontal component that looks for a contrast along the array. You are showing them a single tiny contrast (white to blue/red border) and not understanding how poor a focus contrast that is. Like trying to focus on a tiny black spot on a white wall. Tiny black spots are hard to focus on and that's expected.Targets like this are easy:And why they are used to actually see if there might be a need for service due to focus issues. You're, in a sense, asking the dealer to fix your car because it seems to pull to the left when you drive on the side of a hill. Not a perfect analogy but kind of demonstrates the issue.I often disagree with Leonard but I agree with him that this target could present a problem if the AF sensor happened to see just the center where the line spacing is extremely fine.I see that point as well...but still a good target to see if further testing might be called for. Something like the lens align target would be even better.In most cases the aim will be off enough that the wider lines will be seen by the AF detector and it will focus fine. But there's no point in introducing such a problematic condition. My eyes don't even like looking at the center of that target.I've actully found using a screen (like on a window) formoire interference patternsis extremely accurate as well.

Leonard Shepherd wrote:A thin line scan can produce spurious resultsIt remains a complete mystery why you persistently keep on making this claim, whereas tests shown to you clearly indicate this is not the case. Even a small rectangle that pops up in a focus area already gives very accurate focus results.http://www.dpreview.com/forums/post/51284996http://www.dpreview.com/forums/post/42439983Whether you are ignoring, denying or simply lying is a complete mystery. What remains is that based on facts your statement is completely flawed.It never ceases to amaze me why some Nikon users prefer to believe in fairy tales for testing AF instead of following the guidance from Nikon and Canon, and then holler without taking sensible tests to find out if the problem is them or the equipment.Yeah, I imagine you in front of a mirror, telling yourself "mirror mirror on the wall, who is the best auto focus expert of them all"...If you cannot accomplish accurate focus with a target as the OP has used, I propose you take a course in photography. Your current approach to this topic is an insult to the reader on this forum as well to the AF capabilities of the Nikon brand.To the OP, AF is known to be sensitive to color. PDAF works by analyzing 2 different maps of an object. Wide angle lenses with large apertures suffer from chromatic aberrations, hence the contrast edges of objects are (1) blurred a bit and (2) displaced on the focus line pairs (e.g. seeFalk's sitereferring to a Hoya patent describing the issue):Canon refers to a new AF mechanism, where they claim to compensate for this sensitivity (Canon's white paper on the EOS 1D Mark IV). One page 16, they mention the following:I.e., the sensitivity to color for AF is expected behavior, and the fact that the OP sees this effect in his test speaks only in favor of the quality of the test targets.

Leonard Shepherd wrote:Mako2011 wrote:All Nikon DSLR camera instruction books caution geometric patterns may result in focus error.Geometric in the context of AF. Not all geometric patterns will create a geometric AF pattern; the above target is an example of that.Not aligning exactly on the centre of the red circle could confuse the camera into focussing at the wrong distance.This is a wrong assumption, there is no reason at all to centre object for correct autofocus. I think you don't understand the concept of a differential measurement.In case you are still convinced, can you please explain in detail, including the way the target is mapped to the AF lines why you should centre the red dot?I would not rely on this type of target for a 100% reliable test.I would. Practice shows that targets like this work sufficiently well for testing AF behavior. If you have problems with targets like this, I recommend to do a course in handling your camera.

SteveL54 wrote:I don't believe the issue of the OP's focus errors on an eye has much to do with the effect of color on the PDAF viewfinder focusing. The spectrum is diverse enough in the example photo of the baby's portrait that I doubt it played much role in the misfocus.OP: Nailing the focus on the leading eye with a large aperture can be difficult. In this case you were closed down to f/4.5 so the DOF was wider. If you get consistent errors at this setting then something is wrong with the equipment or your technique. When I use f/2.0 and try to get the leading eye in focus I expect to get some misses, but I get a lot of well focused leading eyes as well. There's always the chance the subject or camera moved between focus aquisition and shutter. Sometimes you don't target the eye steadily during the focus aquisition phase. But if you are reasonably skilled and deliberate you should get several images that are properly focused.Okay, then this means that the blue/red figure may not explain the shift on the eye which initiated my one month testing.The eye cases are like these: I may have like 100 or so pictures (didn't really count) that I kept of my son. more than 10 of them are like the one I posted, the rate is not high, but high enough to invoke my suspect.I'm not 100% sure there is no shake or something during focus, but I think not enough to move the frame far enough to cause that far a shift.And on the local forum Xitek here, I see at least 2 other people complain about the same back focus on people's eye, one with D300S+ 3518g, the other unknown.Or if it's normal for 35 18g, then maybe I should just stop using the lens. But in the situation, it may be silly to buy another lens.Again, you're using targets and conditions known to cause focus inconsistencies. In some situations it's best to focus on a good spot/target then recompose.OP: The saturated blue and red targets are extreme cases. The blue might provide marginal contrast but if the focus error is consistent it's not a contrast issue. I used a few pastel blue Post-It notes on a white sheet of paper. The paper I used provided even less contrast that the blue target you used. Most of the time the D300 and 35 F/1.8 focused well. But a few times it missed. The camera was on a tripod and lighting was about a stop darker than your scenario. (Thanks for posting the new photos with all the metadata intact.)After the first visit to Nikon, I too can get the blue pattern focus almost right under good sun light (a bit front focus) constantly, very much the same as black and white pattern. But the red figure cause big front focus. And I fancy with red + black lines, I got some big back focus, but didn't really tried much of that---in want of simplifying the test.While under some unnatural light (like the one at Nikons), the blue one cause big back focus, the red one relatively close to the target.Did not much tested black/white under all situations, but I vaguely remember in the tests before, they usually give front focus (after the tweak), with good light, it seems closer to the target, but still slightly front focus.Also please be noted again that my tests are only showing good shift difference at a far enough distance, like at least 1 meter.And if your point 3-4 are right, I do need some service for the camera.Not really, I said those issues can affect focus. Simply check the AF unit for dust...and try testing using better targets (black and white high contrast targets) in differing lighting conditions...to see if AF-Sub mirror adjustment is called forI agree that the OP should go back to a good B&W high contast target.OP: The AF color dependency issue is a very interesting topic. But I think you need to make sure that your camera can accurately focus on a good B&W target -- or at least one that doesn't have pure red or blue saturated color. The few images that you provide with a B&W target suggest the camera or lens needs calibration. If it's adjusted correctly the focus should be centered on the scale so that other variations like color won't cause an even greater error.Use a target with bold and irregular width B&W lines. The targets I recommend are shown in my recommended AF test technique:http://www.pbase.com/slockhart54/auto_focus_test_procedureYou should change the target's orientation to see if you can reproduce the different focusing error that you encountered with the orientation change of blue target. With the center cross-type AF points you should get close to the same focus solution whether the lines run vertically or horizontally. If you get the same large variation as illustrated in your supplied photos then that's another indication something is wrong with the camera or lens.The orientation change error with the blue target may not be about the orientation from the first place, but about the coverage of the color in the focus frame. Because I tested later with the same thing, the blue figure cause the same shift with which ever orientation, even if I rotate it at a 45 degree angle.As for the B/W target, didn't do much of "formal tests" recently, but followed Mako's advice of testing a paper, not on any pod, with rather good sunlight, it always be more blurry than the liveview focus, and with greenish edge on the font---a sign of front focus I think.The Nikon guy claim that the camera is accurate on their target under natural sun light, he offered me another tweak according to my usage convention, I wanted to tweak it back a little too, but did not get time, so just left it as is now.

Marc Heijligers wrote:To the OP, AF is known to be sensitive to color. PDAF works by analyzing 2 different maps of an object. Wide angle lenses with large apertures suffer from chromatic aberrations, hence the contrast edges of objects are (1) blurred a bit and (2) displaced on the focus line pairs (e.g. seeFalk's sitereferring to a Hoya patent describing the issue):Canon refers to a new AF mechanism, where they claim to compensate for this sensitivity (Canon's white paper on the EOS 1D Mark IV). One page 16, they mention the following:I.e., the sensitivity to color for AF is expected behavior, and the fact that the OP sees this effect in his test speaks only in favor of the quality of the test targets.Thanks, I think this is a theory I can understand and believe easier, than Mako's "not enough contrast" metaphor.And for the reason of this, I wonder if a lens with less CA like the new 35mm ED, or a lens with smaller biggest aperture, could alleviate the color related focus shift. I did found less evidence of this in the 18-55 VR II.

Mako2011 wrote:SteveL54 wrote:I don't believe the issue of the OP's focus errors on an eye has much to do with the effect of color on the PDAF viewfinder focusing.Me too...I was only speaking to the red blue target differences he was seeing.I do hope that they are of the same cause, so I can rest a little :-).The shift degrees are alike, there are basically 2 colors involved in the eye issue sample, dark brown in the apple of the eye and a little pink of the skin---With saturated red + black I did see some back focus, I fancy.And the rate of the eye issue being reproduced incline me to believe that the relatively random positioning of the focus frame may sometimes cover the "right" pattern around the eye, and cause the issue.Another evidence is that under sun light the eye focus issue are more frequent, fit the test results with saturated colors that are affected by light quite much.And the logical path that lead me to the color test: first seeing the eye focus issue at a rate, then clicked the button around to see if other target make shift too, and found the difference of red/blue cards on the wall, then some more "serious" test with drawings/printed figures. Cannot think of anything more now...

liwei zhen wrote:Marc Heijligers wrote:To the OP, AF is known to be sensitive to color. PDAF works by analyzing 2 different maps of an object. Wide angle lenses with large apertures suffer from chromatic aberrations, hence the contrast edges of objects are (1) blurred a bit and (2) displaced on the focus line pairs (e.g. seeFalk's sitereferring to a Hoya patent describing the issue):Canon refers to a new AF mechanism, where they claim to compensate for this sensitivity (Canon's white paper on the EOS 1D Mark IV). One page 16, they mention the following:I.e., the sensitivity to color for AF is expected behavior, and the fact that the OP sees this effect in his test speaks only in favor of the quality of the test targets.Thanks, I think this is a theory I can understand and believe easier, than Mako's "not enough contrast" metaphor.And for the reason of this, I wonder if a lens with less CA like the new 35mm ED, or a lens with smaller biggest aperture, could alleviate the color related focus shift. I did found less evidence of this in the 18-55 VR II.Most imaging lenses are at least achromatics which means they correct for chromatic aberration at two points in the spectrum. I believe much of the color focus problem is due to the optics within the PDAF. They are very simple lenses such as the separator lenses depicted in the diagram above. I don't believe they make any attempt to correct for color. The PDAF units used by Nikon only use the light from the f/5.6 or f/8 annulus of the imaging lens, so any chromatic aberration observed in very bright imaging lenses is not considered by the phase detection sensor. Still, an imagining lens with better control of CA may help. I just don't think it can be eliminated because of the simple PDAF optics.Your camera's mirrors may not be properly aligned. Next time you bring the camera in for servicing ask them to perform the mirror alignment inspections if they didn't do it on the first visit.I have a hypothesis as to why we see so many reports of sporadic PDAF failures with very bright lenses. In an attempt to improve tracking capability, the new PDAFs use dozens of tightly packed AF points instead of the half dozen or so points used in the older units. This means dispensing with the field mask. Without the field mask, stray light (flare) can get into one side of the phase detection sensor and thus create a non-symmetrical condition on the sensor. Ultra bright lenses are more prone to flare. This flare may not be that obvious in the final image but it will perturb the correlation of the pdaf sensor's paired images.Steve

liwei zhen wrote:SteveL54 wrote:I don't believe the issue of the OP's focus errors on an eye has much to do with the effect of color on the PDAF viewfinder focusing. The spectrum is diverse enough in the example photo of the baby's portrait that I doubt it played much role in the misfocus.OP: Nailing the focus on the leading eye with a large aperture can be difficult. In this case you were closed down to f/4.5 so the DOF was wider. If you get consistent errors at this setting then something is wrong with the equipment or your technique. When I use f/2.0 and try to get the leading eye in focus I expect to get some misses, but I get a lot of well focused leading eyes as well. There's always the chance the subject or camera moved between focus aquisition and shutter. Sometimes you don't target the eye steadily during the focus aquisition phase. But if you are reasonably skilled and deliberate you should get several images that are properly focused.Okay, then this means that the blue/red figure may not explain the shift on the eye which initiated my one month testing.The eye cases are like these: I may have like 100 or so pictures (didn't really count) that I kept of my son. more than 10 of them are like the one I posted, the rate is not high, but high enough to invoke my suspect.That's not too bad of a failure rate if you consider subject movement and other effects that can affect the fast measurements made by a PDAF unit. As I mentioned in my other recent post if there's a strong light source in or near the frame or a sceen with nonuniform lighting, flare might occasionally throw off the PDAF solution.As for the B/W target, didn't do much of "formal tests" recently, but followed Mako's advice of testing a paper, not on any pod, with rather good sunlight, it always be more blurry than the liveview focus, and with greenish edge on the font---a sign of front focus I think.Yes, that's what I saw in one of your earlier tests with a B&W target and the scale -- front focus. The green halo is longitudinal CA. The plane of focus was in front of the target.The Nikon guy claim that the camera is accurate on their target under natural sun light, he offered me another tweak according to my usage convention, I wanted to tweak it back a little too, but did not get time, so just left it as is now.I think they should check the camera again, and make sure they check the mirror alignment.Steve

Leonard Shepherd wrote:Mako2011 wrote:All Nikon DSLR camera instruction books caution geometric patterns may result in focus error.Leonard,You still don't understand the concept ofregularas it applies to geometric patterns. I don't want repeat myself so here's the history of my attempts to set you straight.http://www.dpreview.com/forums/post/41271055Not aligning exactly on the centre of the red circle could confuse the camera into focussing at the wrong distance.Not again! I thought you finally stopped posting your totally erroneous and flawed centering notion about PDAF!http://www.dpreview.com/forums/post/37475367Steve

SteveL54 wrote:You still don't understand the concept ofregularas it applies to geometric patterns. I don't want repeat myself so here's the history of my attempts to set you straight.http://www.dpreview.com/forums/post/41271055Not aligning exactly on the centre of the red circle could confuse the camera into focussing at the wrong distance.Not again! I thought you finally stopped posting your totally erroneous and flawed centering notion about PDAF!http://www.dpreview.com/forums/post/37475367SteveI did similar attempts, but it looks like any information provided to Leonard is ignored on purpose.Here I've explained what symmetry means in terms of AF:http://www.dpreview.com/forums/post/42399326Here i've explained how object map to the AF lines:http://www.dpreview.com/forums/post/50699353Here I've explained and proven that in practice the difference between so called "ideal targets" and "less ideal targets" is negligible:http://www.dpreview.com/forums/post/51284996Simple questions that show the flaws of his reasoning are ignored persistently:http://www.dpreview.com/forums/post/52771877Leonard does not respond at all, and simply re-starts his AF inquisition in another thread. For that reason, I have my doubts about his integrity towards others, and his posts are more about himself than about AF.The reason for keeping on posting counter arguments is not because I believe he can be convinced (he does not want to), but to provide confused dpreview readers (that may not know him) an indication about the true mechanisms that that matter for AF.

Marc Heijligers wrote:SteveL54 wrote:You still don't understand the concept ofregularas it applies to geometric patterns. I don't want repeat myself so here's the history of my attempts to set you straight.http://www.dpreview.com/forums/post/41271055Not aligning exactly on the centre of the red circle could confuse the camera into focussing at the wrong distance.Not again! I thought you finally stopped posting your totally erroneous and flawed centering notion about PDAF!http://www.dpreview.com/forums/post/37475367SteveI did similar attempts, but it looks like any information provided to Leonard is ignored on purpose.Here I've explained what symmetry means in terms of AF:http://www.dpreview.com/forums/post/42399326Here i've explained how object map to the AF lines:http://www.dpreview.com/forums/post/50699353Here I've explained and proven that in practice the difference between so called "ideal targets" and "less ideal targets" is negligible:http://www.dpreview.com/forums/post/51284996Simple questions that show the flaws of his reasoning are ignored persistently:http://www.dpreview.com/forums/post/52771877Leonard does not respond at all, and simply re-starts his AF inquisition in another thread. For that reason, I have my doubts about his integrity towards others, and his posts are more about himself than about AF.The reason for keeping on posting counter arguments is not because I believe he can be convinced (he does not want to), but to provide confused dpreview readers (that may not know him) an indication about the true mechanisms that that matter for AF.That poor original poster, he provided several pics and test targets that were perfectly fine, yet he was met with nonsense from some. The truth is that if his setup couldn't acquire perfect focus and Nikon fails to accecpt that there is a problem, then that setup is simply worth nothing and he would be better off with a Holga.(I returned my d5300 for the exact same reasonhttp://www.dpreview.com/forums/thread/3658259and have been following this tread in order to try to get a reasonable explanation why a 800 USD dslr fails so bad as it does)

SteveL54 wrote:That's not too bad of a failure rate if you consider subject movement and other effects that can affect the fast measurements made by a PDAF unit. As I mentioned in my other recent post if there's a strong light source in or near the frame or a sceen with nonuniform lighting, flare might occasionally throw off the PDAF solution.Thanks, that could be a good reason. I never had the hood on even when going out in sun. The failure rate was not too high, but still the same constant pattern of shift alarmed me. Sometimes at a point, I can get 3-4 similar OOF shots, where the move the target or my hand are rather less possible, or I should have been very aware.I think they should check the camera again, and make sure they check the mirror alignment.I also did the tests with different focus points, and with high contrasts lines at the different border inside the center focus point, all make no perceivable difference. Doesn't that mean the alignment is good?Also if you meant that the alignment could cause the red/blue focus issue (don't think you mean that, but rather the original eye focus issue), they give me 2 more bodies and 2 more lenses to try, similar thing.

Emil Jacobsen wrote:That poor original poster, he provided several pics and test targets that were perfectly fine, yet he was met with nonsense from some. The truth is that if his setup couldn't acquire perfect focus and Nikon fails to accecpt that there is a problem, then that setup is simply worth nothing and he would be better off with a Holga.(I returned my d5300 for the exact same reasonhttp://www.dpreview.com/forums/thread/3658259and have been following this tread in order to try to get a reasonable explanation why a 800 USD dslr fails so bad as it does)Hi, thanks for the sympathy. I read through your thread too. Looks something similar to mine, but of course we don't shoot in the very same way. And your failure rate are rather different from mine (before the first tweak). I noticed the failed examples are all taken in a longer distance, so could it just be a normal shift, that are not very obvious at a closer distance (so you have some in focus pics also)?Unfortunately for me, as said, we don't have that liberal a return policy here, with any manufacturers. Even the big international ones that may offer great service in some countries. Here they won't accept a opened product back for a reason like "I cannot focus like I used to", they first assume it's that "you don't know how to use it", then "it's common to have random error", or "the product works to our standard".---quite a complex issue for undeveloped society.The Nikon technician did offer me some extra devices to test on my red/blue setup, and that did prove the red/blue things are "common"---but I'm not saying "acceptable to me".I don't know if the color thing has to do with the original focus shift I (and some others) experienced---Steve and Mako are both experienced pros, who think they are not related. But the combo are rather unusable, in various light conditions, so am I impressed.

Hi, I went to Nikon again last Friday, had the lens tweaked again "according to my standard". The result is that most targets with good contrast (like a paper) are acceptable now.Took some pictures during the weekend, not too bad. But still get similar pattern of back focus with the kid's eyes. While now knowing things like the color caused focus shift are common, I think I can live with that.It was cloudy 5PM that I had the back focus, with the hood on the lens, so not much likely it was caused by flare. I still believe more in the color theory, or perhaps it could be about the eyelashes sticking out.By the way, I did get a a6000 like 2 month ago to compare with the D5300. But found it hunts quite much when focusing on the face, especially when the light is a bit low or when the kid is moving. So I guess you are right, Steve, I need the DSLR for photoing the kid. And thanks to Stacey for the suggestion of AF-C too.And a6000's NR (even if turned off) for jpeg and exposure in some bright scene, sometimes make me unhappy too. Feel like I lost some details (to high light most of the time). Could be I used it wrong too.Will try some more with that camera to see if things like AF-C or face detection, + different metering will work better.