AF Nikkor 28mm f/2.8D lens review

General

Before I even start looking at particularities, I can not stop myself from saying that my overall impression based on real-life shooting is that this lens, which I bought new for USD213, is simply superb! This, this, and this photograph was taken with it. Very few people seem to own it these days, though - some find its focal length too trivial, some do not bother as its focal length is within that of almost any wide-angle or normal zoom. I personally think that this is not fair, as this lens is very light, cheap, and never stops surprising me with the quality of the images it delivers. It has six elements in six groups, is 64*44.5mm in size, focuses up to 25cm, and weighs only 205g. It does not have the CRC (Close Range Correction) system - AF Nikkor 28mm f/1.4D does. Its build quality seems to be the same as that of AF Nikkor 50mm f/1.4D, which for this price is perfectly acceptable. On a Nikon F100, it auto-focuses very, very fast with practically no hunting (partly due to its relatively short focal length). If necessary, you can find simple explanations of the terms below here. MTF graphs for this lens can be found here.

You might also be interested to know how this lens compares against the AF-S Zoom-Nikkor 17-35mm f/2.8D IF-ED set at 28mm - please read here.

Sharpness

Photodo.com rates this lens at 3.2, which at first seemed quite strange to me as I found this lens particularly sharp. At least, having used it for over a year by now I have an impression that, when stopped down, it is as sharp as the AF Nikkor 50mm f/1.4D, which is rated by the same Web site at 4.2. Apparently, this is one of those cases when MTF graphs and overall ratings can not tell you everything about a lens for your particular practical uses and purposes - but more on this in the Conclusions part.

This lens is super sharp, period. If you insist on being fastidious, though, there is a discernable loss of sharpness in the corners at f/2.8 and, to a lesser extent, at f/4 if compared with f/8 (where the lens is sharpest). However, you have to look really hard to actually see any difference in sharpness at these apertures in the center. An important point here is that light fall-off is a much bigger problem at f/2.8 (see below), so that the issue of sharpness at this aperture, at least to me, becomes somewhat irrelevant.

It took me several minutes to actually find any difference in sharpness at f/22 and f/8 even in the furthest corners with a 10X loupe, but it was so insignificant that for practical purposes I would say that this lens is as sharp at f/22 as it is at f/8. Interestingly enough, respective shutter speed at f/22 was 1/8 seconds, and, as I read in many reviews, the Nikon F100's mirror slap (see my comments on it here) is supposed to be at its worst at this shutter speed. However, I got quite an outstanding result, which praises both the lens and the camera.

Light fall-off

The test images below are scans of top right quarters of full-frame slides (light fall-off, of course, is symmetrical against the center of the frame).

f/2.8	f/4
f/5.6	f/8

The lens exhibits very strong light fall-off at f/2.8. In fact, it is so bad at this aperture that it rather looks like an overall underexposure than just light fall-off. It is significantly better at f/4, yet I would not use it at this aperture if light conditions allow. Quite unexpectedly, minor light fall-off is still visible in the furthest corners at f/5.6, although I have to stress that I never noticed this in real photographs and became aware of it only through these tests. For real-life photography, one can consider the lens free from light fall-off from f/5.6 on (which is the aperture I tend to use this lens at most of the time anyway).

Distortion

As you can see in the above picture, the lens demonstrates some barrel distortion; however, it is really insignificant for most applications. As another example, you can see this real-life picture - you will see that distortion is almost not noticeable.

Flare and Ghosting

Flare is very well controlled - see the images below. You can also see this picture, which also indicates that you can put bright sources of light in your composition without worrying about flare or ghosts.

The sun is just outside the image's right top corner	The sun is in the image's right top corner	The sun is in the center of the image

Bokeh

This is a wide-angle lens and one probably should not worry about bokeh. At least, I seem to have very, very few pictures with an out-of-focus background taken with this lens (you can see one example here, though). Nevertheless, it was interesting to check the lens' bokeh anyway so I took several pictures - see below.

Bokeh at f/2.8	Bokeh at f/4	Bokeh at f/5.6

As expected, bokeh is not too good at f/2.8, fairly neutral at f/4 and quite good at f/5.6. Wonder what bokeh looks like at f/8? Try getting pictures with a blurred background at this aperture J!

Conclusions

A long time ago I wanted only one - good quality - wide-angle lens (which is different now). I had used the AF Zoom-Nikkor 24-85mm f/2.8-4D for over half a year by the time and decided that 28mm was the focal length of the only wide-angle lens I wanted. Well, with this lens I got the two things I needed - superb quality and the right single wide-angle focal length. As I said above, this lens still surprises me with the high technical quality of the images I take with it; on the other hand, having used it for over a year now, I think that 28mm was the right choice for the only wide-angle lens. (An afterthought here: ever wondered why it is only at 28mm that Nikon produced an AF f/1.4 lens?)

Now back to juxtaposing my overall impression and MTF tests. The lens does not perform well at f/2.8; also, it performs significantly better at the level of 10lp/mm (which is more important) than at 20lp/mm or 40lp/mm. Apparently, all this drags its overall rating down to 3.2. Speaking of f/2.8, though, how often does one need or have to use this kind of lens wide-open? If you think about it, not that often, really. On the one hand, it is mainly used for the types of photography (e.g. landscape photography) where you normally need a rather deep depth of field. On the other hand, at this focal length you can shoot hand-held at 1/20 seconds, or even slower, speeds and get sharp results, which is to say that using the lens at f/2.8 can seldom be imposed by ambient light conditions. Due to these reasons I almost always use it at f/5.6 or, preferably, at f/8 - the lens performs great in this range and hence gives such a good subjective impression.

Juxtaposition in the above paragraph draws on a very important point: what matters is not how a lens performs objectively at all apertures with respect to all factors, but how it performs at the apertures you use most often (or it is supposed to be used at) and with respect to the factors that are important to you personally. I guess that Nikon engineers had this in mind when designing the lens and did it for a typical user such as myself.

To sum it all up, the lens is not very impressive at f/2.8 and f/4 - consider getting the AF Nikkor 28mm f/1.4D instead if you shoot in dark ambient light often (do not forget about the price difference, though!). Nevertheless, it generally delivers stunning results at the apertures it is supposed and likely to be used at, i.e. from f/5.6 on; use it at f/8 if you want to get the best technical quality out of it.