Thanks Leonard. You are right, people need to know how their equipment works.While with this case, the equipment is really problematic and it's not the subject that makes the focus shift. And it's more like a good example of "the owner didn't make enough observation and gave misleading information to bother people". Sorry and thanks again, guys.And David, thanks for letting us know the possibility here. I learned. But still I think Stacey and Steve "guessed" correctly in practice, in this particular case.

Some update on the focus shift of my camera:I tried the combo quite some more, and found the focus shift happens mostly on some color pattern, like some blue figures printed on some big cards (they are big so I can make sure I focused on the card from 2-3 meters away). Otherwise there are some minor and random shift (which could be normal?)I figured it must be a defect of the camera since the pattern of target makes difference. Today I took it to the service (the bigger one out of 2 here in Shanghai). But they won't hear my point of the inconsistency in the focusing, and only assume it's a normal focus shift, and calibrate the lens for me.After the calibration, I think I saw some serious front focus, when focusing on the TV screens (outputing some status info from computer) hung there in the wall, but still they refuse to admit it's a defect. The defenses they use: different standard, reflection on the screen, too big a focus point---none of which really stand before this big a focus shift.When brought the Camera back, and did my normal tests again, I found yes, the back focus is mostly gone for the blue card I used, but with the red cards, I got a serious front focus. The cards are covered with very thin plastics, again, reflection? But other than that, now I'm getting OOF shots on more targets than before, even when getting closer.I'm not sure, is this really a defect of the camera, or am I just being too picky here?

liwei zhen wrote:Some update on the focus shift of my camera:I tried the combo quite some more, and found the focus shift happens mostly on some color pattern, like some blue figures printed on some big cards (they are big so I can make sure I focused on the card from 2-3 meters away). Otherwise there are some minor and random shift (which could be normal?)I figured it must be a defect of the camera since the pattern of target makes difference. Today I took it to the service (the bigger one out of 2 here in Shanghai). But they won't hear my point of the inconsistency in the focusing, and only assume it's a normal focus shift, and calibrate the lens for me.After the calibration, I think I saw some serious front focus, when focusing on the TV screens (outputing some status info from computer) hung there in the wall, but still they refuse to admit it's a defect. The defenses they use: different standard, reflection on the screen, too big a focus point---none of which really stand before this big a focus shift.When brought the Camera back, and did my normal tests again, I found yes, the back focus is mostly gone for the blue card I used, but with the red cards, I got a serious front focus. The cards are covered with very thin plastics, again, reflection? But other than that, now I'm getting OOF shots on more targets than before, even when getting closer.I'm not sure, is this really a defect of the camera, or am I just being too picky here?Sounds like you might be being a tad unrealistic with PDAF expectations and inconsistent with your focus target selection. The field of view of the focus arrays is different than the dipiction of the focus box in the viewfinder. Have you considered that? The PDAF unit also sees red/green/blue uniquely so using colored targets can affect results. Are you giving the camera a contrast to work with? If your refective targets are acting almost mirror like, then you can expect issues. In this case, you seem to be giving the camera targets that are not optimal for PDAF to get accurate lock, but expecting it to anyways.

Thank Mako. I really hope the behavior of the camera is alright, and it's only my own issue too. Then I can maybe resell it or give it to a friend knowing that I'm not cheating.But I did selected the target rather normally I think, good contrast in the focus point, the focus point on one single object only. Just the focus shift is a bit random. Now the curtain targets that will cause consistent shift are a set of kid's cards with blue/red figures (with different shift). And that TV displaying Chinese words/numbers at Nikon's.And the other sure thing is that it takes a rather long distance (2+ meters or so) to makes the focus shift obvious. What could this mean? The sensor failed to recognize the contrast at that distance? but there was no hunting at all. Or CA being too heavy?

liwei zhen wrote:Thank Mako. I really hope the behavior of the camera is alright, and it's only my own issue too. Then I can maybe resell it or give it to a friend knowing that I'm not cheating.But I did selected the target rather normally I think, good contrast in the focus point, the focus point on one single object only.Remember, the "Focus point" in the viewfinder is not the same as what the FoV of the actual focus array is seeing.Just the focus shift is a bit random. Now the curtain targets that will cause consistent shift are a set of kid's cards with blue/red figures (with different shift). And that TV displaying Chinese words/numbers at Nikon's.Again, those appear to be poor targets in that they are not optimal for checking focus accuracy. See the notes about focus on page 52 in the D5300 manual.And the other sure thing is that it takes a rather long distance (2+ meters or so) to makes the focus shift obvious. What could this mean? The sensor failed to recognize the contrast at that distance? but there was no hunting at all. Or CA being too heavy?Likely not optimal conditions. Hard to say without an actual example image. PDAF is fast not perfect

Thanks for being patient on this. And I was a bit busy recently to take more serious test. But could you please have a look at this? Thank you very much.OOF on the red figure.These are taken after the tweak. f1.8, iso 800, around 1/160s, center point AF-S. handhold, no re-compose. (but there could be very slight shock). It's more than 2 meters I think, and I aimed to the point at a less than 45 degrees angle.The cards are quite big, almost like a B5 paper (but thinner). I think the left one had been given quite some contrast. But it shows a big front shift, to the card 2 or 3 blocks to the left of the center point.The right picture is much better, even though seemingly it receives less contrast. These 2 are pretty much reproducible under different lights.And before the tweak, it's the blue figures under the red ones that causing a back focus at a similar degree.

How many of these threads have we seen involving the 35mm 1.8?

liwei zhen wrote:Thanks for being patient on this. And I was a bit busy recently to take more serious test. But could you please have a look at this? Thank you very much.OOF on the red figure.These are taken after the tweak. f1.8, iso 800, around 1/160s, center point AF-S. handhold, no re-compose. (but there could be very slight shock). It's more than 2 meters I think, and I aimed to the point at a less than 45 degrees angle.Might be some back focus but you can't judge that using a target that is angled. It can cause the results to be skewed badly. Target on the left might have enough contrast but the one on the right does not. Here's the best way to set up a target for testing focus...http://camerafocustest.blogspot.com/See how they use a flat target with an angled surface next to it to judge focus plane.The cards are quite big, almost like a B5 paper (but thinner). I think the left one had been given quite some contrast. But it shows a big front shift, to the card 2 or 3 blocks to the left of the center point.May be but you have to not use angled targets or the result can get messed up. This does mean I would check further though.The right picture is much better, even though seemingly it receives less contrast. These 2 are pretty much reproducible under different lights.And before the tweak, it's the blue figures under the red ones that causing a back focus at a similar degree.Might be being caused by dust/debris on the PDAF unit at the bottom of the mirror box. Hard to say

David Lal wrote:If focus has locked onto the reflection itself the camera to subject distance will be the camera to reflected object distance - same as focussing on a mirror.In fact, eye reflections have long since fascinated me and I seek the opportunities to make such images deliberately. Here's an example, the apparent subject is the eye (of a carrion crow) but the real subject is me and my house reflected in his eye.Very cool pics. Thanks for posting.

olliess wrote:Very cool pics. Thanks for posting.Thank you for your interest and kind remarks Olliess.

Hi, thanks for the helps. Cannot open blogspot here though (censored by GOV). Will have some more check later with some proxy.Points to add:The blue cards cause different focus shift before the tweak (now it's almost okay), while the red ones turned from okay to front focus. So I'm not sure it's about dust.I did try focusing on the blue cards before the tweak, just got blur pictures largely. Didn't take the card off the wall and put some other stuff around before.And again, the cards are covered with very thin plastics, which from the angle and the lighting condition, doesn't cause any perceivable reflection from my eye or in the taken picture. I tried to "skin" them but they are too tight.Will take some better target and continue testing.Let's hope like Jakes said: it's just the 35mm f1.8g. I did search the comments on focus shift of the lens before posting, don't recall to see one complaint exactly like mine.

Correction:"I did try focusing on the blue cardsright in front of thembefore the tweak"

Hi, Mako. I had some test with more correct approach and targets, could you please take a look and help (helps from other people are very much appreciated too):test with non-tilted targetsThis time I was shooting at the targets right in front of them. The distance was less then 2 meters; The light source fluorescent; af-s, center point; iso 400; f1.8; 1/50-60s; on tripod.The left most AA battery is right in front of the targets. the next one 5-6 cm closer to camera. The last one another 5-6 cm closer.The target in the left pic is a piece of normal A4 paper, draw with some mark pen. The on in right pic is a red envelope with no plastic cover at all. Also marked with the same pen.The result on the left was a 10cm or so shift to front. Pretty much reproducible. Didn't do real test in sun light though.The right pic shows focus much closer to the target, but the result is not stable, sometimes it even back focus quite a lot.So it seems that not the "red" figure that caused the front focus, but maybe white? Do the phenomena look normal to you?Also got a 18-55 vr II for test (because it's cheap), but the aperture is too small at 35mm (f4.5), the result is not much obvious. But I fancy I saw similar front focus too. Will check some more, maybe from farther distance and in better light.

liwei zhen wrote:Hi, Mako. I had some test with more correct approach and targets, could you please take a look and help (helps from other people are very much appreciated too):test with non-tilted targetsThis time I was shooting at the targets right in front of them. The distance was less then 2 meters; The light source fluorescent; af-s, center point; iso 400; f1.8; 1/50-60s; on tripod.That's pretty interesting. I assume you are still using the 35mm f1.8g. As a side note, that lens really comes up a lot when it relates to focus issues with the Nikon 39pt Multi-Cam 4800DX focus unit. Just something to consider. Be sure and test again in sunlight just to see.The left most AA battery is right in front of the targets. the next one 5-6 cm closer to camera. The last one another 5-6 cm closer.The target in the left pic is a piece of normal A4 paper, draw with some mark pen. The on in right pic is a red envelope with no plastic cover at all. Also marked with the same pen.Target might need a bit more contrast to see what might be happening. In your target it might be that only the vertical array is getting contrast to focus on and the black against red can actually be showing more contrast because of the Muti-cam design.Always better to test with a target that looks more like this and be sure the angled ruler (batteries) is well off to the side so it they can't interfere.The result on the left was a 10cm or so shift to front. Pretty much reproducible. Didn't do real test in sun light though.The right pic shows focus much closer to the target, but the result is not stable, sometimes it even back focus quite a lot.So it seems that not the "red" figure that caused the front focus, but maybe white? Do the phenomena look normal to you?There is a possibility the AF unit at the bottom of the mirror box needs cleaning or the AF sub mirrior needs adjusting by Nikon.Also got a 18-55 vr II for test (because it's cheap), but the aperture is too small at 35mm (f4.5), the result is not much obvious. But I fancy I saw similar front focus too. Will check some more, maybe from farther distance and in better light.I would go ahead and test with that in good light to see. You can use an angled ruler to make even f4.5 obvious. Did it like this:http://camerafocustest.blogspot.com/

Is it under warranty? Can you send it in to Nikon?I went through the exact same things you are with my d7000 for 2 months. (reading as much as possible and trying all those different tests with inconsistent results). Lost confidence in my camera and in my technique. Finally sent it in and they immediately determined there was a  autofocus calibration problem with the body and are fixing it.Seems to me that finding out for sure whether or not there is a problem would be worth the time/money just for the piece of mind.

Thanks, totally agree. I even lost confidence in DSLR VF focus system as a whole. But you guys saying "this is abnormal" brings me back a little.The camera/lens are still under warranty. And I took them to Nikon service a week ago. But they didn't much listen to my explanation, just said it's a focus shift, and adjusted the lens (where I don't know whether the camera was touched too). After the tweak, it's better on some previously back-focus target, but introduced quite some front focus on some other targets.I heard D5300 share the same PDAF module as D7000, then probably I'm experiencing the same thing as you. So your camera was just "adjusted", no "repairing"? Could that correct all the inconsistency?

Thanks Mako, will do some more test when the weather is better here. Some points of mine below:Target might need a bit more contrast to see what might be happening. In your target it might be that only the vertical array is getting contrast to focus on and the black against red can actually be showing more contrast because of the Muti-cam design.Always better to test with a target that looks more like this and be sure the angled ruler (batteries) is well off to the side so it they can't interfere.In reality we often are not able to provide as good a target as you pasted. If I bring my paper to Nikon, I don't think they should (they might, actually) say "our camera doesn't lock on that target".Of course I agree it works as a measure to help see better conclusion.Also you are suggesting that the batteries are too close to the AF point, that they might have be considered by the AF system? If yes, the AF system is terribly inaccurate, I should say. I understand there are 2 light paths, but still there should be a very high degree of consistency in the two.Of course again, it's a good idea to isolate possible problems. Thanks again.There is a possibility the AF unit at the bottom of the mirror box needs cleaning or the AF sub mirrior needs adjusting by Nikon.One other fact is that when I'm closer to the target, like in 0.5M, where the DOF is much shallower, the focus is fine. And with all the inconsistent/random result with different targets, I wonder if it's just a defect (or a defect in my brain), rather than something adjustable.

... the virtual image reflected from the curved surface of the eye would have to dominate any detail from the eye and its immediate area.liwei zhen wrote:...., so I wonder could it be the refection in the eye that shifted my focus?I don't see any practical case where such a reflection would dominate the iris, eyelids, eyelashes, eyebrow, etc. These provide much more contrast for the AF to work with than the typical reflection off the eye.I mean the reflection is like a reflection in a mirror, just the eye is with a outwardly curved surface, and with low contrast reflection. Then could it be the reflection that make the "back focus", like when focusing a subject in a mirror, the focus fall in the mirrored subject but not the mirror itself.Unlike a flat mirror, the virtual image cast off the highly curved eye's surface is not that far behind the eye's surface. So if the AF did focus exclusively on the virtual image, the error wouldn't be as large as the focus error you presented earlier of the baby's eye.Tried with some dolls with plastic eyes, cannot reproduce it---the eyes are not that smooth, and the scenario could be a little different.Look about on the web to see if there was any theory about this, but never found on answer. Any comments? Thanks a lot.Find something like a large polished metal bowl that has a near spherical section. This should create a virtual image of any bright object reflected off its outer convex surface that is just behind the curved surface. The reflection I chose to focus on was a double row of long flourescent lamps several feet away. This caused the camera to focus just a bit past the bowl itself. If I put some dark tape on the bowl then the camera would focus a bit closer on the actual tape stuck to the bowl's surface.Steve

liwei zhen wrote:Thanks Mako, will do some more test when the weather is better here. Some points of mine below:Target might need a bit more contrast to see what might be happening. In your target it might be that only the vertical array is getting contrast to focus on and the black against red can actually be showing more contrast because of the Muti-cam design.Always better to test with a target that looks more like this and be sure the angled ruler (batteries) is well off to the side so it they can't interfere.In reality we often are not able to provide as good a target as you pasted. If I bring my paper to Nikon, I don't think they should (they might, actually) say "our camera doesn't lock on that target".Of course I agree it works as a measure to help see better conclusion.Totally understand and agreeAlso you are suggesting that the batteries are too close to the AF point, that they might have be considered by the AF system? If yes, the AF system is terribly inaccurate, I should say.In the case of the Muti-Cam 4800DX unit...the FOV of the focus arrays can extend almost 50% beyond the boundaries of the focus box in the viewfinder. If the batteries are that close...then the array may pick them as the best contrast it can find. Something to remember and why I"m careful with what I lock onto when testing focus or trying to lock onto small objects. . Once you get used to that..It can be very accurate with focus. Also makes keeping lock in a dydnamic AF-C situation easier.  Pros and Cons. If I had to move up to the unit in the D7100...I'd have to retrain how I focus. Wouldn't take long and you'd get accurate with that system as well.I understand there are 2 light paths, but still there should be a very high degree of consistency in the two.They are very consistent. Problem is that most targets are not consistent when you look at them from the point of view of the arrays and from a contrast standpoint. If the contrast at the intersection is weak...they adjust, and then often the contrast along the FOV's differ so the algorithms have top make choices.If you superimpose the arrays FOV (purple) over the red box in your picture...you'll see the contrasts the arrays are working with can vary greatly. Just something to consider when learning how best to use what we have An example...note how each specific array (of 39) has way more than one contrast to work with in this sceneLook at just the center array.  It sees...top of box, bottom of box, lettering left, lettering right. And it's seeing in a different way than we are (RGB). Complicated stuff.Of course again, it's a good idea to isolate possible problems. Thanks again.There is a possibility the AF unit at the bottom of the mirror box needs cleaning or the AF sub mirrior needs adjusting by Nikon.One other fact is that when I'm closer to the target, like in 0.5M, where the DOF is much shallower, the focus is fine. And with all the inconsistent/random result with different targets, I wonder if it's just a defect (or a defect in my brain), rather than something adjustable.I do think a fair chance your unit might be off a bit. AF sub-mirror needs adjustment or unit is dirty/poor sample. Troubleshooting a tiny AF issue is really hard. Also the 35mm f1.8g is for some strange reason a magnifier of AF issues in many cases.

Thanks Mako for all the help. Glad I posted here. Please see some comments and a little progress below:Mako2011 wrote:If you superimpose the arrays FOV (purple) over the red box in your picture...you'll see the contrasts the arrays are working with can vary greatly. Just something to consider when learning how best to use what we have An example...note how each specific array (of 39) has way more than one contrast to work with in this sceneLook at just the center array. It sees...top of box, bottom of box, lettering left, lettering right. And it's seeing in a different way than we are (RGB). Complicated stuff.I see the point. While my thought is that when we make sure the frame of focus point (shown in view nx/capture nx) are on one surface (parallel to focal plane) only, then we are sure the focus should not be out of that surface. Of course the camera could still slightly shock on my tripod, but not much. In the case of the last pic I pasted, I think the frame is far enough from the battery.The "no 45 degrees" thing is actually the same, IMHO. 45 degree surface is not good, but the focus plane should still be within the scope of the nearest point in the focus frame to the farthest point, while in my previous cases, the plane was shifted way out of the same card. This I think is practical an approach to test the shift from a longer distance: when we need the large reference "ruler" (all the cards hung on the wall) to see the shift for the much thicker DOF.Those said, I did make a similar setup as the blog you posted---but much shabbier. The shift difference when focusing on the differently colored paper still stands. But with more detailed reference than batteries, I can see that the difference when shooting from nearer place too. So the shifts (and difference) do exist at closer distance, just not as much as standing farther. Here is an example of the small difference at close range (forgot how close):closer shot with smaller difference on differently colored paper. 35mm @f1.8. not so good sunlight.Please note the scale on it is not accurate (bigger than real ruler), not sure how long they are at horizontal direction, maybe closer to the real thing.And there are errors since the tool set is light, not well made and I might position it and the camera slightly differently. But still the error doesn't account for all the shift difference on the paper.Tried 18-55 too, but the bigger DOF forced me to try at a close distance, while at the close distance, the focus difference could be very small---at least smaller than 35 at f1.8, and I can not safely tell. So I still need to try some more with much bigger reference object. Will do that with even better light and better tool kit.