Is there a problem with my 70-180mm micro ? (see pic)

[fredscal]

<p>Hello.</p>

<p>I recently shot this pic: and I wonder wether there is a problem with my lens. It is a pre-digital lens so maybe the percieved defect still fall within the 1990s' tolerance ? Every insight welcome, but could you please also tell me whether you've personnally used this lens at some point, so I can put every opinion into perspective ? Thanks.</p>

<p>So here's what bothers me: the background wall, +/- parallel to the sensor, dramatically looses sharpness as we get away from the center, towards the right edge. That could be considered normal, except the left-side wall is more than reasonably sharp all across -- even though not parallel to the focus plane (the aperture is F11 and everything was more than 40 feet away). Isn't that puzzling? Moreover, the blur near the right edge doesn't look like out-of-focus blur to me. It looks... weirder.</p>

<p>So, what do you think ? Should I get my lens repaired, or is its design just too old to use with a D600 ?</p>

<p>(I've used this lens for years on a D700 and never noticed anything of the sort... So I don't know whether the defect appeared following a possible shock or sustained vibrations (I carry it around everywhere), or if it was always there but only got revealed by the more modern sensor.) The lens' exact reference is Nikkor AF 70-180/4.5-5.6 D Micro ( http://www.photosynthesis.co.nz/nikon/afd70180.jpg ), it was introduced is 1997 and discontinued in 2005.</p>

[michael_bradtke]

<p>It might be decentered. If it was mine I would send it in and have it checked out.</p>

[lornesunley]

<p>neither the coping on the top of the right hand wall or the pipe running across it appear to go "dramatically" out of focus across the photo.<br>

I also do not see how that right hand wall can be parallel to the plane of focus.</p>

[keirst]

<p>Lorne is right; this is not a planar subject, nor is it parallel to the imaging plane of the sensor. I think you need to be more methodical about your testing before assuming there is a problem. F/11 isn’t going to give deep focus (especially examining the image at the pixel level) with a telephoto focal length, even at 70mm zoom. Did you use a tripod or monopod support? If not, there could be motion blur added to out of focus blur. Have you AF fine tuned the lens? Did you try focusing in Live View? If the the focus is substantially different between Live View and focusing using the prism viewfinder, then the lens needs AF fine tuning.</p>

[rodeo_joe1]

<p>I'm afraid that very little can be told about the lens from that image Fred. The plane of focus is tilted upward for a start, and no part of the image is parallel to the focus plane.</p>

<p>Visualise the back of your camera defining the angle and inclination of the plane of focus, and that in actual fact there is no such thing as depth-of-field. Unless the subject and focal plane are absolutely parallel to each other, there can only be one, or a small number of points in the subject that are absolutely sharp and in focus; hence the much-maligned boring old brick wall test for lenses. I suggest you find a boring old brick wall and set the camera up precisely parallel to it, in both the vertical and horizontal planes. Only then can you decide whether there's some sort of fault or misalignment of the lens.</p>

<p>Another quick test is to take two shots with the exact same framing, one with the camera right way up, and the other with the camera held upside down. Obviously the focus setting should remain fixed between the two shots. If the two images vary in the sharpness of each corner or edge when rotated to the same orientation, then there's a good chance the lens is decentred.</p>

<p>Decentring can be a tricky thing to pin down, because if the decentring or tilt is mainly in a part of the lens that rotates during focusing, then the decentring shifts position with focus as well. IME there aren't many lenses that <em>don't </em>exhibit a small amount of decentring if you look hard enough, especially complex zooms and lenses containing aspherical elements, which are extremely difficult to mount exactly coaxially. The amount of decentring shouldn't be visible in "normal" use however.</p>

[fredscal]

<p>What I meant about the wall being parallel to the plane of focus is just that it is "more parallel" than that other, sharper wall ;-).</p>

<p>Thank you for your inputs on how to conduct a more rigorous test. I appreciate the education. Any other advice welcome.</p>

[keirst]

<p>One thing that can complicate testing of some lenses is field curvature, which I read that this model of lens exhibits to some degree. Lenses with field curvature don’t actually focus a perfect plane of subject onto the film or sensor, but instead the area off perfect focus resembles a parabolic dish, where the corners may focus either closer to the camera than the center of the image (forward field curvature, example 28mm/1.8G AFS Nikkor) or further away (rearward field curvature, example 28mm/2 Zeiss ZF.2). <br /><br />Lenses that exhibit no field curvature are called flat field designs, such as the 55mm/2.8 Micro Nikkor. Flat field lenses are very good at projecting a planar subject such as a brick wall or 2D artwork onto a sensor. Curved field lenses are not as good at this and the corners will be soft when the center is focused perfectly, and vice versa, focusing for the corners will soften the center of the captured image, unless you can stop down far enough to bring the every part of the subject plane into depth of field.<br /><br />In photographing 3D objects, field curvature is not really much of a problem, and may be even helpful. For example, I find the 28mm/1.8G Nikkor’s forward field curvature helps increase foreground detail in my use of the lens for vertical landscape frames, whereas the Zeiss 28mm gives slightly softer foreground corners in the same kind of picture, even stopped down to f/11. <br /><br />I have not used the 70-18mm Micro lens, so I’m not sure how much the field curvature issue affects it, but you probably want to look for it as you test. You might want to include both flat test targets and 3d subjects at different distances in your evaluations, and try different apertures from wide open to the minimum aperture. Also be aware that diffraction will start to blur the image as you stop down past f/8, so even as you gain depth of field there is a trade off resulting in less resolution at the point of focus.<br /><br />Hope things work out as you get to know the lens.</p>

[fredscal]

<p>@Steven: I thought of that too, but would be surprised because the blur ressembles a defect more than regular bokeh. But you may be right (because I may be wrong), and anyway, thanks for your very clear explanation of field curvature: I'll save it as one of the best concise articles on the matter.</p>

[Edwin Barkdoll]

<p>Fred, I've used the lens quite a bit primarily on a D300 and more recently on a D600. I love the lens for its macro capability (micro) and its versatility. I've never been terribly impressed with its non-micro performance, however, and therefore do not use it as much in a non-micro situation unless I have to. As as been mentioned above, f/11 is not going to achieve great depth of field particularly at the longer focal length (I didn't see what focal length was used).</p>

<p>Metadata have been stripped from the file so I cannot tell where the focus point was but this may also have some bearing.<br>

Looking at some images of planar objects I've taken with this lens I do not see field curvature as a significant enough problem to cause the effect you are seeing i your photo<br>

Finally, as Steven says, a more rigorous test is in order before you send it in for repair - use a level and make sure the focal plane is actually parallel to the object plane.</p>

[fredscal]

<p>@Edwin: oops, the host site must have stripped the metadata. The focal length is 70mm, and as for the focus point, it is always central, even if it means I need to reframe after focusing. I know this is not optimal technique but I find the D600's DX AF matrix, combined with its FX sensor, about useless. (This is one of the many little reasons why I probably won't be keeping this body much longer.)</p>

<p>PS: I agree far focusing is not this lens' strongest suit. But still, I like its vintage feel -- except on this pic it got a little too vintage ;-).</p>

<p>Anyway, tomorrow I'll go ask Nikon how much they ask for a diagnostic. I neither have the infrastructure nor the patience for rigorous testing.</p>

[fredscal]

<p>Ok, here's how it is: Nikon will test it thoroughly for (only!) 30€. Then I'll know if there's something to be done, and at which price.<br>

Thank you all for your inputs.</p>