Nikon 40mm/f2.8 DX AF-S Micro Lens, First Impressions

[cc_chang2]

<p>Here are a couple of close up of people taken by the 40/2.8 from Nikon:<br>

http://www.nikonusa.com/Nikon-Products/Product/Camera-Lenses/2200/AF-S-DX-Micro-NIKKOR-40mm-f%252F2.8G.html#lightbox/carousel/media=2200_ER_land_port_01.jpg<br>

Can this be done easily with the 35/1.8?</p>

[ShunCheung]

<p>CC, please take a look at Jose Angel's explanation on the "full length portrait lens" <a href="00ZHry">http://www.photo.net/nikon-camera-forum/00ZHry</a>, where you posted the same incorrect information:</p>

<blockquote>

<p ><a name="00ZI8Z"></a><a href="../photodb/user?user_id=499258">Jose Angel</a> <a href="../member-status-icons"><img title="Hero" src="../v3graphics/member-status-icons/hero.gif" alt="" /><img title="Frequent poster" src="../v3graphics/member-status-icons/1roll.gif" alt="" /></a>, Sep 06, 2011; 10:07 a.m.</p>

 

<p>Perspective is only distance dependant (in this case, camera to subject distance).<br />Focal lenght doesn`t have anything to do with it. It doesn`t matter if 35, 50 or whatever.</p>

</blockquote>

<p>In other words, as long as your distance (i.e. the photographer's distance or actually the camera's distance) from your subject (the person) does not change, you can use the 35mm/f1.8 DX AF-S or 40mm/f2.8 DX AF-S micro and the perspective is the same; i.e. any perspective distortion will also be the same. The 35mm lens will provide a slighter wide angle of view so that your'll get more surroundings/background than you will with the 40mm lens, and the subject's face will occupy a smaller percentage of the entire frame.</p>

<p>However, if you move closer to your subject so that your subject's head will look bigger to compensate for the wider angle of view from the 35mm lens, your subject will look more distorted because, again, the nose is now proportionally closer to the camera then the ears.</p>

<p>Therefore, if you want people portraits to look "right" on a DX body, you'll still need something like a 60mm to 75mm lens, or perhaps a bit longer depending on your preference. The fact that this 40mm lens is a macro will not change that equation.</p>

<p>The 40mm DX micro happens to have very little geometric distortion, but that is a totally different type of distortion. See the "brick wall" image below.</p>

<P>

<CENTER>

<IMG SRC="http://d6d2h4gfvy8t8.cloudfront.net/14153394-lg.jpg">

</CENTER>

</P>

[walterh]

<p>The bokeh and rendering of the macro examples from above link are pretty. So far the only really positive aspect for me.</p>

<p>By the way: one reason to design short macro lenses is the need for a large opening angle to obtain high resolution at magnification (above 1:1). My bet is that it is not better in resolution or distortion than the old Micro Nikkor 55mm f3.5 or similar older designs. The fact that it is a G lens points to a more consumer use where such parameters are of minor importance.</p>

 

[walterh]

<p>Shun we doubled. Now I see distortion at that range seems not perfect but quite ok.<br>

And thanks for taking the extra effort to answer al burning questions before the final test :-)</p>

 

[cc_chang2]

<blockquote>

<p>However, if you move closer to your subject so that your subject's head will look bigger to compensate for the wider angle of view from the 35mm lens, your subject will look more distorted because, again, the nose is now proportionally closer to the camera then the ears.</p>

</blockquote>

<p>Jose and I agree with one another because when I made my statement, it was made in the context of achieving more or less the same subject coverage using lens of different focal length. If you wish to say fill the frame with a face, with a 85mm lens, you can do so farther way from the subject, thus resulting in less distortion. However if you were to do the same with a 35m lens, you need to move in closer, resulting in more distortion.</p>

<p>If you look at the #5 picture on Nikon's web site, it is a very close up of a woman's face. As far as I can see, her nose and ears and other features are of the correct portion. It is highly unlikely that the regular 35/1.8 can do the same, although a 85mm lens can do so with ease. In fact this would be something you could include in your lens test. Take head and shoulder shots of your son using both lenses, and show us the results. We have seen enough charts and MTFs ...</p>

[walterh]

<p>The difference between 40mm and 35mm is there but it is small.<br>

The fact that the 40mm lens is a macro lens has nothing to do with perspective distortion.<br>

http://en.wikipedia.org/wiki/Perspective_distortion_%28photography%29<br>

Different lens designs may have a larger impact in cases where the focal length changes with focusing e.g. internal focusing design or floating elements. This is more apparent in zoom lenses.</p>

[cc_chang2]

<blockquote>

<p>The difference between 40mm and 35mm is there but it is small.</p>

</blockquote>

<p>The key question in my mind is that for <strong>head-shoulder shots,</strong> is the new 40mm micro much better than the 35/1.8? Nikon's marketing is telling us that the 40mm micro can do a very good job on that, and stresses the point of versatility. So far the discussion here focuses entirely on its performance at 1:1. Considering its potential users, 1:1 may not be where it will be asked to perform. We should perhaps think of this as a lens that can focus really really close to the subjects without a great deal of distortion, and one can take advantage of this to take close ups of flowers and food, or photographing people's face/shoulder, all typical subjects for vacationers, in a light and small package ...</p>

[andylynn]

<p>Don't confuse distortion that is caused by the optics' failure to render straight lines (barrel and pincushion distortion, usually) with perspective distortion that is caused under certain circumstances shooting subjects that are close to the camera. Being close exaggerates the third dimension and elongates the sides of the frame. The wider the lens, the closer you get to the subject, the more of this distortion. This is not a function of the quality of the lens - you will get the same perspective distortion in a 3D render simulating a portrait from up close, that simulates an optically perfect lens.</p>

<p>So the type of distortion you're trying to avoid by using a long lens to shoot portraits is not a function of the particular lens, but of the field of view and subject distance. There's not much difference between 40mm and 35mm. There's not going to be much difference in the portraits you shoot. (If you have a normal zoom, try shooting at 35 and 40 and see if it really changes much.)</p>

[jose_angel]

<p>I read time ago (I think it was close to the announcement on this lens on topic) something that confused me and is probably the origin of this confusion (I guess). It was about the flatness of field of macro lenses and their supposedly inherent ability to "avoid" that perspective distortion. I have tried to find more info about this topic on the web and on my books, unsucessfully.</p>

<p>I`m struggling my head this days trying to understand how a theoretically perfect flat field lens could benefit that kind of "nose distortion" in comparison to an ordinary curved field lens.<br /> From what I have read, I have to be a bit skeptic about the field flatness of Micro-Nikkors.</p>

<p>My knowledge in optics is very limited; the only difference I see is about the distance of the different points in the scene to the focus plane, shorter in the flat field lens. Theoretically, I guess, it should affect out of focus blur in this points, or even the size, but I also wonder if it could be really noticeable.</p>

<p>In fact, I has been tempted to ask Shun to check for it, taking a couple of pics with the new 40Micro and another with a 35AFD or 45P, or even a 50mm prime, to see at least if there is even a subtle difference. I was wondering too if this effect could be noticed only with a tilted or leveled camera.</p>

<p>It will be unnecessary if someone experienced in this topic post the answer... !</p>

[jose_angel]

<p><em>"the distance of the different points in the scene to the focus plane, shorter in the flat field lens... " </em>Behind that plane, I mean. Also, considering the focus plane of a non-flat field lens as a curved plane (!).<br /> Well, I`m totally lost.</p>

[ShunCheung]

<p>Jose, two weeks before the 40mm DX Micro was officially announced, Josh Root and I had a conference call with Nikon USA's Steve Heiner, who is the general manager of their DSLR systems. Steve gave us heads up on this then up-coming lens so that I could write a preview.</p>

<p>During that call, I specifically asked Steve about flat field since Nikon's 60mm macro lens is well known to be a flat field macro, which means the focus plane itself is flat. This type of macro is optimized for copying work, such as copying books, paintings, etc. Steven pointed out that the 40mm DX macro is also a flat field lens. I have a few examples showing that it has very little geometric distortion so that the edges of your paintings will look straight and it is sharp from corner to corner.</p>

<p>Perspective distortion is something totally different. When we shoot "head and shoulder" type portraits, we are typically 2, 3 meters from the subject so that the proportion of the nose, eyes, ears, etc. look "right." When you use too short a lens and you still want to cover the head and shoulder, you have to get much closer. All of a sudden the nose is now a lot closer to the camera than the ear, and typically the face will look fat and the nose looks big. There is no lens that can automatically correct that optically.</p>

<p>Here are a couple of (poorly captured) samples. The subject is a Epson 8.5x11" paper box. I captured the first image with a D700, full FX frame with the 40mm DX macro. Even from just 1 foot away, you can see darken corners. However, the edges of the box remain quite straight.</p><div></div>

[ShunCheung]

<p>Now, I captured a similar image with the 35mm/f1.4 AI-S (manual focus) on the D700. From about a foot away, the 35mm AI-S is showing obvious barrel distortion, but since it is a lens intended for FX, there are no darken corners.</p><div></div>

[cc_chang2]

<p>I dragged the image from Nikon's web site to show what Nikon says the 40 micro can do when photographing people. I hope I am not violating any copy right issue ... As you can see, there is a close head-shoulder shot from the 40 micro lens. I do not see any objectionable distortion in this picture. May be this is the result of a "flat field"/micro lens or what ever you want to call it? or is it due to the way that her head is laying horizontally ... (or what ever) By contrast, the promotional pictures for the 35/1.8 lens contain only an environmental portrait which was shot perhaps 5-6 ft away from a person. There is no head shots provided and we all know from our experience that using the 35/1.8 for head shots will results in distortion.</p><div></div>

[ShunCheung]

<p>Sorry CC, you are in violation of photo.net's copyright rule; you are supposed to only post images you yourself captured, as I did for each one of the images I posted to this thread.</p>

<p>But concerning the image you posted, do you notice that the top of the subject's head/forehead looks awfully big? The subject's face is not parallel to the camera; therefore, the area that is closest to the camera, namely the forehead, is exaggerated. No lens can violate the laws of optics. I still have Nikon's 40mm DX sample around. When I find a siutable subject, I'll take some portrait shots to show you the results.</p>

<p>Again, flat field merely means this lens can focus on a flat plane, as a lot of lenses focus on a curved plane so that when you shoot a flat wall or a painting, you cannot get a sharp center and sharp corners at the same time until you stop way down to gain a lot of depth of field. It has nothing to do with perspective distortion.</p>

[cc_chang2]

<blockquote>

<p>So the type of distortion you're trying to avoid by using a long lens to shoot portraits is not a function of the particular lens, but of the field of view and subject distance. There's not much difference between 40mm and 35mm.</p>

 

</blockquote>

<p>Andy: It is understood already that a 85mm lens, but not a 40mm lens, is typically used in the film days for head-shoulder shots primarily because it allows one to shoot at a comfortable working distance that is just far away enough as to not cause distortion. But the discussion here is not about JUST a 40 mm lens, but about the new Nikon micro 40mm lens, which Nikon claims that it CAN shoot head-shoulder shots. Is this true, which we hopefully will soon find out from Shun. If true, why? Is it because it is a micro/macro lens that all micro lenses are designed to shoot close without distortion or is it because this is a Nikon flat-field micro lens? I am clearly not the expert here in terms of lens design but I am simply intrigued by the potential usage of this lens for head-should shots.</p>

[cc_chang2]

<p>Sorry Shun about the picture, won't do it again.</p>

<p>I too noticed the way the shot was framed with her head tilted so a "large" nose would not be so noticeable. As for the portion of her features, Nikon posted another shot of her that was taken a bit farther way, and it shows a similar portion of her forehead. May be she does look like that or may be it too is caused by distortion. Any way, your test will clarify it all. Thank you for being patient.</p>

[ShunCheung]

<p>CC, you CAN shoot portraits with the 40mm macro, but you still need to stand 2, 3 meters away as you do with a 85mm or 105mm portrait lens on FX. Remember, perspective is determined by camera-to-subject distance, not focal length. Obviously a 40mm, even on DX, will cover more than a 85mm or 105mm on FX, so you have to crop to get just the head and shoulder. We have no idea how much cropping Nikon has done on their sample images.</p>

<p>I am in the office right now; while I have the 40mm lens with me, taking distorted images of co-workers and then post them to the web is not a way to make friends. There is a guy who brings his dog to the office every day, so I was going to use his dog as the model. But somehow today is the one day he doesn't have his dog with him. So I have to wait, perhaps a few days.</p>

[jose_angel]

<p>Thanks Shun. I now know that this lens is a flat field one.</p>

<p>My cake is complex. I was wondering about the effects of a flat field vs curved field lens on a 3D subject like that on the Nikon site. I`ll leave it here.</p>

<p>I understand the effect you mention, <em>"... top of the subject's head/forehead looks awfully big... ",</em> are both perspective distortion and volume anamorphosis (the image gets thicker near the limits of the image circle) typical of low geometrical distortion lenses. With a high (geometrical) distortion lens the subject will look more "rounded" (at least in the corners), I guess. This is why I prefer a fisheye over a extreme wide angle for photographing people in confined spaces (at least until mine was stolen... ).</p>

<p>To CC: I wonder if the promotional phrase in the Nikon site is confusing you; <em>"The AF-S DX Micro-NIKKOR 40mm f/2.8G is compact, lightweight and versatile enough for everyday photography, <strong>including flattering people shots</strong> and breathtaking landscapes, so you get brilliant results in most situations without switching lenses."</em><em><br /></em><br /> I think this lens doesn`t have any special feature that make faces to look flatter at closer distances. Maybe they want to mean that it`s long enough (focal lenght) to be used for portraiture (head shots) with acceptable results (longer focus distance than a 50mm on FX -doesn`t look too bad-).</p>

[andylynn]

<p>CC, I already explained, and Shun explained better, that there are two <em>entirely separate and unrelated</em> kinds of distortion being discussed on this thread - though I think we've now expanded it to three.<br>

1. The type the Nikon told Shun it had designed this lens to correct, that is shown in Shun's photos of the paper box, is barrel distortion. As you can see in Shun's shots, an ordinary lens used close (even a very good one like that 35/1.4) will fail to make straight lines straight - they'll be curved. <a href="http://en.wikipedia.org/wiki/File:Barrel_distortion.svg">You can see a simple drawing of that here</a> and <a href="http://en.wikipedia.org/wiki/Distortion_%28optics%29">an explanation here</a>.<br>

2. A related issue, technically referred to as field curvature, that I don't think is technically distortion. <a href="http://en.wikipedia.org/wiki/Petzval_field_curvature">See here for a drawing and explanation</a>. This lens has been designed to render a flat field, meaning that the lens focuses on a plane parallel to the plane of the sensor. A normal lens renders anything <em>at the distance of focus</em> as in focus. If you've taken geometry, you know that the set of points in three dimensional space that are equidistant at some particular distance from a point is a sphere. If you are standing in the center of a spherical object and you have a regular lens, you can photograph the inside wall and put the whole frame in focus, but you can't photograph a flat wall and put the whole frame in perfect focus. A flat field lens can put a flat wall in perfect focus.<br>

3. Perspective distortion, which is the type of distortion that makes people want to not shoot portraits with short lenses, <a href="http://en.wikipedia.org/wiki/Perspective_distortion_%28photography%29">is described here</a>. It is a function of differential foreshortening, which is <a href="http://en.wikipedia.org/wiki/Foreshortening">described here</a>. None of these lenses can correct this problem. I guess it's possible to make a lens that would, but it would be huge, weird, unreasonably expensive and nobody would want it because you could just use a longer lens and stand back further.</p>

<p>Whatever Nikon is saying should not be construed to mean that Nikon has fixed the third type of distortion with this lens.</p>

[ShunCheung]

<p>Here is a portrait captured with the 40mm/f2.8 DX Micro on a D7000. I was, of course, unusually close to the subject. See the exaggerated nose?</p>

<p>This is the entire frame, uncropped, only scaled down to 700 pixels.</p><div></div>

[ShunCheung]

<p>The review is now out. Most of the topics and pros/cons we have already discussed in this thread: <a href="../equipment/nikon/lenses/40mm-f2.8-af-s-dx-micro/review/">http://www.photo.net/equipment/nikon/lenses/40mm-f2.8-af-s-dx-micro/review/</a></p>

<p>Again, I am still hoping that someone else who has this lens can provide additional comments and opinions.</p>

[kumar_b]

<p>I will be buying the 40mm Micro Nikkor sometime soon. This will be my first real macro lens and I have thus been reading around to get the best out of it. So, here is a question for the experts. <br />Is there something to quantify the extent to which macro lenses correct for perspective distortion when up close ? Consider the following hypothetical scenarios for taking head shots (on DX sensors) : <br />a) 40mm macro close to the subject vs 40mm non-macro at same distance - <br />b) 40mm macro close to subject vs 85mm macro further from subject <br />c) 40mm macro close to subject vs 85 mm non-macro further from the subject<br />Let us assume that there is enough light that the possible difference in max appertures for macro and non-macro lenses is not a factor here. <br />Now, my guesses are <br />a) 40mm macro is better<br />b) 85mm macro is better<br />c) Not sure<br />Am I right on a) and b ) and what is the best way to figure out the answer for c) ?<br>

Thanks in advance. This is my first post here.<br>

(I posted this as a separate question and then realized this thread might be a better place since my question is related to the discussions here) </p>