3.5-4
(D2X)

Chromatic aberration, mainly blue fringing, can occur and is much more visible than with the 12-24 Nikkor at the wide end. However, judicious post-processing can deal with this problem so it is not a huge issue. Otherwise, image sharpness is good or excellent on-axis, while the corners need some stopping down to get into their stride. This is pronounced towards the wide end of the zoom range, and occurs despite the strongly aspherical front element. Setting the lens to f/8 or f/11 assures good image quality across the imaged field. Focusing with HSM (AFS equivalent) is quite snappy, but not blindingly fast.

The external finish of this lens looks nice, but will wear more quickly than the Nikkors.

To sum up, a very acceptable quality at a lower price makes it an attractive alternative to Nikon DX-format users.

3.5
(@ 12 mm)

4
(@ 18 mm)

5
(@ 24 mm)

(D70, D1X, D2H, D2X)

IR:

1-5
(D1, D1H, D70, D2H, D200, S2Pro, S3Pro UVIR)

Despite Nikon's advertised claims about DX lenses being small and light, the 12-24 DX fulfils neither, in fact it is similar to say the 14 mm f/2.8 in size and heft (about 0.5 kg). The lens is quite nicely built, but on my review sample the focusing collar rattled somewhat. The lens takes 77 mm filters and shares the hood HB-23 with current 17-35 and 18-35 Nikkors. Focusing down to 0.3m, the 12-24 DX operates fast and silently as expected from its AFS lineage. It is a true IF design and thus there is no front rotation or change of total lens length during focusing.

The wildly exotic optical design, with plenty of aspherical and ED elements, produces high-contrast images with vividly saturated colours and low levels of flare and ghosting. Field curvature is negligible and evenness of illumination across the entire frame is remarkable. Set wide open at 12 mm, light fall-off into the corners is detectable, but a far cry from being troublesome. Under all other conditions there is virtually no vignetting. A spectacular achievement indeed.

Geometric distortion is low for such a wide lens and although the predictable barrel distortion does show up at 12 mm, the amount is not alarming and in fact much less than found in the 17-35 AFS (@17 mm). At 24 mm, a slight pincushion distortion can be seen.

At 24 mm, very sharp and crisp images resulted at nearly all aperture settings. Only the f/4 and f/22 were slightly less sharply etched. Peak performance occurred at f/8-f/11, but the high image quality held up nearly everywhere else too. Colour fringing was virtually non-existent at this end of the focal range. No Nikkor 24 mm prime lens gives this kind of quality images.

However, towards the 12 mm setting more chromatic aberration occurred and image sharpness in general declined. Thus, at 12 mm the corners were distinctly soft unless the lens was stopped down beyond f/8. Peak performance still was in the f/8-f/11 range.

A direct comparison (@17mm) with the AFS 17-35 mm Nikkor indicated the 17-35 had more geometric distortion than the 12-24 DX, but exhibited slighly sharper images (at f/8). I hadn't the opportunity to repeat comparison at other common focal lengths, but wouldn't be surprised if the 12-24DX got the better of 17-35 at the 24mm end.

The 12-24 DX behaved impeccably similar on both my D1X and D1H, so no nasty surprises like those found with the 14 mm should be expected here. On D2X however, the occasional need for additional processing of the image is evident.

(The full review is here)

IR performance: This lens is frequently severely flawed by a dominant hot-spot. The tendency for hot spotting develops towards the shorter focal settings and is exacerbated when the lens is stopped down. You have to try with your own camera to see if the combination works satisfactorily, the chances are against it though.

5
(FX:DX)

A zoom design encompassing up to 114 degrees picture angle can hardly be expected to show perfectly straight lines at all focal settings. Thus, used up close towards the wide end, barrel distortion is very visible, but for more normal shooting distances the distortion is much reduced. I shot cityscapes without undue distortion (you do get severe keystoning of course which is a matter of image geometry more than a lens flaw). Around 20mm there is a cross-over to pincushion distortion, but this never gets really troublesome.

The rear element moves a lot while the lens zooms, but stays put during focusing (the 14-24 is an IF design). Since there is a collet around the rear element daredevils might try taping a gel filter there and this should work. However, there obviously are reasons why Nikon didn't provide a filter slot on the rear, so I suspect you jeopardise image quality by putting even a gel at the rear.

The sophisticated nano-crystal coating applied on the inside of the front element helps reduce flare and ghosting, but eliminating these issues completely is an impossibly tall order. Having said that, I got much less flare and ghosts than with virtually any other wide lens I've used so far.

4
(DX:D2X, D200)

IR:

4-4.5
(DX: D200 modified)

At the widest setting, there is modest barrel distortion but huge amounts of corner darkening (vignetting), the latter needs massive stopping down to clear, to beyond f/8 when the lens is set to 16 mm. The geometric distortion is minimal around 40 mm and then crosses over in the usual manner to become a distinct pincushion towards the 85 mm focal length. I observed a slight field curvature too, so refraining from using the lens wide open seems a prudent move anyway. Image detail, contrast, and colour saturation, are all good and in the peak range from f/5.6 to f/11 (wide end) to f/8-f/11 (long end), excellent image quality can be acquired. The lens achieves around 1:4 magnification when focused close at its longestend. Not "macro" by any stretch of imagination, but nifty for close portraits, animals, larger flowers, and suchlike.

The filter threads are 67 mm and you need to keep the quite short and wide lens hood attached since the front element is very exposed and thus potentially prone to damage. Flare and ghosting are not troublesome with this lens, but care is needed when you point the lens towards the sun or other strong point-light sources.

AF speed is more than adequate on my pro cameras and VR lends a supporting hand to make amends for the slow speeds brought about by the smallish aperture of this lens. You can shoot down to about 1/15 sec with VR activated.

The 16-85 positions itself as a successor of the 18-70 Nikkor. It provides better image quality, adds a useful extension of range, and there is VR too for those in need of such devices. With all the new features factored into the equation, the loss of 2/3 stop "speed" at the long end becomes tolerable.

Finally, in case anyone wonders, the projected image circle will not cover the FX/24x36 mm format at any focal setting.

IR: I was surprised to learn that quite excellent IR pictures could be acquired with the 16-85 VR. No hot spots were observed in IR. As also seen with visible-light photography, vignetting at the shortest end of the zooming range was plainly visibe in IR. When the lens was zoomed a little away from the 16 mm position, the vignetting issue diminished rapidly.

5
(F5)

5
(D1)

5
(D1X)

5
(D2X, D200)

4.5-5
(FX: D3)

IR:

5
(D1, D2H, D70, S3Pro UVIR)

Quick shooting with it on an F5 left me with an impression of extreme optical quality. I also checked the lens on a D1 and drew similar conclusions there. Sharpness, contrast, and colour saturation all are superb across the entire focal range. Vignetting in the frame corners is moderate at f/2.8@17 mm and largely disappears at f/4. On the D1/D1X, there is no perceivable fall-off at all at the wide end. At the opposite focal end, negligible fall-off is seen even at f/2.8@35 mm on the F5, and none on D1/D1X. There is some barrel distortion in the wide-angle settings, which however is kept under good control even at 17 mm. By contrast, set at 35 mm the lens exhibits a slight degree of pincushion distortion.

Virtually gone is the colour fringing that haunts the AF 20-35 lens. This gives the 17-35 Nikkor a significant advantage over all other similar lenses on the market (all exhibit plentiful of lateral colour aberration towards their shorter end, even the expensive Canon 17-35/2.8 L). This by the way was an improvement necessitated by the D1, which would not have taken the colour fringing of the 20-35 in its stride. The vestiges of colour fringing that may occur is outside the coverage of the lens when mounted on the D1. To put this into perspective for use of the lens on a full-frame 35 mm camera, I had to scrutinise my test shots at 40X magnification to detect the minute traces of residual lateral colour. For all practical and ordinary purposes, this lens is devoid of colour flaws. However, if you shoot close-ups, there will be observable (but slight) blue fringing towards the corners of the image.

The 17-35 performs extremely well when shooting into bright light, in fact its performance in this respects surpasses most prime lenses. Flare and ghosting evidently are strictly controlled. I've never used a zoom with this degree of superior flare and ghosting control before. Likely the fancy optical formula and the bizarre front element pay dividends in this respect, too. However, an early report by "Moose" Peterson claims this lens flares easily. His sample may differ from mine, or test conditions may be quite different. Leaving a UV filter on will make the ghosting much more visible so any filter should be removed before shooting into the sun.

Curvature of field for the 17-35 was very low, so it is eminently suitable for shooting flat as well as 3D subjects. It is quite uncommon for a wide zoom to perform in this way.

A full test of the 17-35 is given here. This review now includes news about production variability of the 17-35.

Added after having used this lens professionally for nearly 3 years: A heavily used lens will get more dusty in its innards and accordingly, be more prone to flare and ghosting. I've seen this occuring with my own sample, so be warned. On a more positive note, my 17-35 has taken a lot of beating without any other ill effects. The surface finish seems to stand well up to wear, too.

On the D3, the 17-35 behaves in an exemplary fashion. Only a small amount of vignetting into the extreme corners occurs at the widest end, and stopping down helps mitigate the issue. However, the age of the design is shown by corner sharpness @17mm being less crisp unless you stop down well beyond f/5.6. So I have reduced the rating ever so slightly to cater for this observation.

IR performance: No hot-spots seen with any camera tested so far.

The AFS 17-35 Nikkor is rapidly becoming one of the Nikon legends. You cannot go wrong with this lens.

4.5 - 5
(D2H)

5
(close range; D2X, D200)

4
(distant; D2X)

4
(FX:D3)

IR:

3
(S3Pro UVIR)

The lens for once lives up to the DX slogan of "smaller, lighter" being a trifle smaller than the 17-35 Nikkor. That is, until you add the impressive, scalloped lens hood HB-31. The hood by the way has a much improved locking feature and stays reasonably put during field use. The zooming control is in front and the focusing collar in rear, just as with the 17-35 and 28-70 Nikkors.

Optical performance clearly shows the clever engineering efforts gone into developing this lens. Thus, with help of its 14 elements (3 of which are ED, and another 3 are aspheric) in 10 groups, the detail sharpness, colour saturation, and image contrast all add up to an excellent imaging quality. Under direct side-by-side comparison with the 17-35 and 28-70 Nikkors, the new DX lens gives virtually identical results to these old champions in the longer range of each of them (28 to 35mm and 35 to 55 mm, respectively). Chromatic aberrations are kept under tight control and this leads to the perceived cleanliness of the images. There is slightly more pronounced geometric distortion, however, begining with a barrel-type at 17 mm and going to mild pincushion at the other end. Around 35-40 mm, negligible geometric distortion is found. Field curvature is, as usual for new optical designs sthese days, minimal and light fall-off into the corners is negligible even with the lens set wide open. Image sharpness is very good at f/2.8 and picks up even more in the f/4-f/8 range.

AFS speed is about equal to the others, and the near limit of the DX at 0.36 m makes for some nice closely cropped shots.

The real issue with this design is its propensity to flare and ghosting. The AFS 17-35 is very resistant to both areas (at least when the lens is new and dirt-free inside), whilst the 28-70 is troubled with flare. The 17-55 DX exhibits quite nasty ghosting under strongly backlit conditions, and there is significant flare shown as well. I guess this is the inevitable price to be paid for the added versatility and the bigger zooming range. If your typical sessions include shooting into the sun, you might be advised to use another lens. However, if you are basically lazy like me, you will be endeared by the charm and versatility of the 17-55. So I ended up (for a while) shooting with 17-55 as my main bread-and-butter lens on D2H and D2X. Now it is retired in favour of prime lenses. But I might reconsider once again.

On the D3, the 17-55 won't cover the entire FX frame below approx. 28mm setting, but you can use it from 24mm if a slight vignetting is allowed. The centre of the image is very crisp and sharp, but the corners do suffer loss of contrast and detail sharpness all the way up to 55mm. So I have reduced the FX rating a little according to this fact. Please note that full FX coverage is not provided at the shortest end, so deploying this lens only on the D3 camera is not recommended.

IR performance (DX): Images come out quite sharp, but there is a fair risk of getting the dreaded hot-spot smack in the middle of the frame.

3.5 - 4.5
(on F5)

At the wide end, the 18-35 has quite visible corner light fall-of wide open, which largely disappears at f/5.6. Field curvature is much more pronounced, too, and the extreme corners (on F5) are quite soft, much less so on the D1. Stopping down to f/8 gives very good centre sharpness and the quality rendition holds up well to f/16. Some colour fringing is evident into the corners and in the out-of-focus areas, but not to a detrimental level so colour rendition is crisp and well saturated. There is some tendency to ghosting and flare under critical light situations.

Image quality increases when the lens is set to longer focal lengths. In fact, the 18-35 attains nearly the same quality at the 35 mm setting as that of the expensive 17-35 AFS. You would indeed be hard pressed to tell the difference here.

I expect the 18-35 Nikkor to become a very popular option for those lusting for the 17-35 AFS, but lacking the powers of persuasion to convince their spouses or bank managers of the need for buying the professional model. It is a worthy alternative, though.

3-
(D2X)

3.5
(D40)

IR:
4+
(D200 modified)

However, if one disregards the obvious "cheapness" of the design, and starts shooting with it, this humble lens has several surprises up its plastic barrel. It handles shooting into the sun with less flare and ghosting much better than some highly touted and very expensive Nikkors (the 17-55 AFS comes to mind). At close range barrel distortion at the 18mm end is quite pronounced but this problem manifests itself less conspicuously for distant shots such as of architectural subjects. Towards the long end some mild pincushion distortion creeps in, but far worse has been seen.

Image quality is slightly below average on the D2X, but improves markedly on its dedicated companion the D40. With the D2X, mild CA of the blue fringing kind can be seen at the longer end, whilst the short end has some purple/cyan fringes. Both are entirely correctable in post-processing. The colour rendition is quite good no doubt thanks to the ED glass inside. For best results, shoot around f/8 at the short and f/11 or so at the long end of the focal range.

IR performance: A really big surprise comes when this lens is used for IR photography. There is no hot spot issues that I could detect, and the images came out very sharp indeed with my D200 [modified] using the rather dark 87C-type filter. Obviously the limited bandwidth of IR alleviated many of the issues shown in ordinary usage in the visible spectrum. Goes to show that one should not have preconceived views of "cheap" gear.

4
(D1X,
D2H,
D70, D80)

3.5-4
(D2X)

IR:
3
(Fuji S3 UVIR)

The lens design comprises ED and aspherical elements and is surprisingly compact and light-weight. Some constructional short-cuts have obviously been employed to get the lens this small, so it cannot take equal beating to, say, a 17-35 Nikkor. On the other hand, the lens has a rear rubber gasket so it prevents intrusion of dirt better than some of the "pro" lens.

Images are crisply defined with a trace of softness into the corners at wide settings, all of which have disappeared by f/5.6 - f/8. Image fall-off into the corners is barely visible at 18 mm when the lens is set wide open. Contrast is high, colours are vividly saturated, and image sharpness simply is stunning compared to the dirt cheap price of this little gem. True, you do get noticeable barrel distortion at the wide end and some pincushion at the other, with a touch of wavy high-order distortion thrown in for a good measure. The really nit-picking users (those who don't ever take photographs, just do test shots) cetainly will enjoy the occasional slight trace of chromatic aberration within high-contrast areas, but the practical importance is negligible both for the nit-pickers and the rest of us. Also, the DX design this time really means "DX", thus the lens cannot be deployed on full-frame cameras to give an image free of vignetting at any focal settings. The image circle projected by the lens is by the way at a maximum around 50 mm.

Flare and ghosting resistance is well above average and much better again that you'd expect from such a low-priced item.

Looks like Nikon has come forward with a real winner this time. For the high-resolution cameras such as D2X, however, the weaknesses of the design are easier to detect, but not all applications to which this lens is put will show them.

IR performance: There is, unfortunately, persistent issues with hot spots at all focal lengths with the Fuji S3Pro UVIR camera. I would not recommend the 18-70 DX for critical IR work.

4
(D1)

The main problem with this lens, and the reason I swapped it for the AFS 28-70, is simply that it shows the ill effects from IF, viz. colour fringing in the periphery and out-of-focus areas. In fact, it never imparts high sharpness when it is applied to landscape-type photography because there remains a tangible fuzziness within the DOF zone. So, although theory predicts a large DOF, it cannot be realised due to the chromatic errors (mostly lateral colour). It is a small consolation for Nikon users that the other brand lenses are even worse far as colour fringing is concerned.

If you insist on using this lens, put the apertures within the f/5.6-f/11 range for best results. I kept mine for just 2 months before I got rid of it - can't understand why I purchased it in the first place. Sometimes you do act like a fool.

(5)

2.5-3
(D1X)

After some haggling, I obtained my sample for free from the odds-and-ends bin of my photo store. I then went on to prove it could be used on my D1X, before it ended its life when I extracted its matrix chip for a better purpose.

4

(D1X)

The design uses aspherical elements but no ED glass, giving it an excellent image rendition with high contrast and vividly saturated colours. There is a trace of corner fall-of at wide apertures at 24 mm, but stopping down to f/5.6 largely cures this problem. Some field curvature leads to soft corners at the wide end, but again stopping down helps. Central image sharpness is excellent and by f/8, you obtain surprisingly sharp images that hold their quality down to f/16. Despite the lack of ED glass, colour fringing is negligible in most situations.

You will, however, have to accept substantial geometric distortion at either end of the zooming range. At 24 mm, a peculiar kind of barrel distortion, aptly named "moustache" distortion is evident on full-frame images, while D1 users will see mild barrel distortion only. Strong pincushion distortion occurs at the long end of the focal settings.

In common with the trend set by the AFS 17-35 and now followed by 18-35 Nikkor, the new 24-85 becomes an even better performer towards the long end. Results obtained at 85 mm showed just a trace of softness into the corners at f/4, and stopping the lens down a little more resulted in remarkably sharp images with all the snap and bite one could ask for.

3.5-5
(D1X,
D100)

3.5-4
(close ups,
D1X,
D100)

Now, turning to the AFS 24-85 itself, it focuses down to 0.38m and incorporates an IF design, so it stays the same size over its focusing range. Neither does the front turn when the lens is zoomed, a desirable trait. However, barrel distortion is evident at 24 mm to change into pincushion at 85 mm. Outside the central image, there is some waviness of the straight line rendition at all focal lengths. This is not a perfect candidate for shooting architecture. Light fall-off is very well controlled across the zooming range, thus it is hardly noticeable even when the lens is used wide-open on a D-series camera. On my F5, corner fall-off however is evident towards the wide end.

Images are rendered with vividly saturated colours, typical for ED designs. Chromatic aberration is absolutely minimal in the 50 to 85 mm range, but can be plainly seen towards 24 mm. Contrast is lowered by internal flare at 24 mm and wide apertures, and some ghosting may occur as reddish spots. Towards the middle and long zoom settings, image contrast increases, lens flare is better controlled, and little ghosting can be observed even when the lens is pointed towards strong light sources.

Sharpness is good at 24 mm, with some softening in the extreme corners unless the lens is stopped down to f/5.6. Peak quality is exhibited towards the long end and at apertures between f/5.6 and f/11. Quality keeps up quite well at f/16, but beyond that point the image softens.

In contrast to many zooms, field flatness is excellent when the lens is focused on near or distant subjects. However, at the near limit the image sharpness declines, field curvature occurs and colour fringing reappears, this time with bluish fringes instead of the purple-cyan type shown at normal distances. Putting the zoom to 85 mm will give 1:4.7 magnification, handy for the occasional flower close-up.

The rating range for this lens is caused by its uneven performance; medium at 24 mm, excellent at 85 mm.

3.5

3
(24mm)

3.5
(50 mm)

4-4.5
(120 mm)

Tested on D1X and D1H

Geometric distortion goes from quite noticeable barrel at 24 mm to pincushion at 120 mm. Around 50 mm lines are kept straight all over the frame.

The VR feature is really useful and I obtained quite sharp images down to 1/8 sec @120 mm. There is an ON/OFF switch for VR and no "Active" mode as on the 70-200VR. A slight softening of the image may occur when VR is not turned off for a tripod-mounted camera, but the deterioration may go unnoticed by many users.

ED glass in the optics manifested itself in good image contrast and highly saturated colours. Just as expected.

Optical performance was very good to excellent at 120 mm, with no field curvature and negligible corner fall-off. Sharpness peaked at f/11 and the entire f/8-f/16 range was very good. Beyond f/16, image contrast and sharpness declined gracefully. Chromatic aberration (CA) occurred only in minute traces towards the image corners.

A quite similar performance was observed at 50mm, but wide-open at f/4.8 the image went quite soft. Again, the best performance was at f/11.

However, at 24 mm the optical performance was quite uneven and non-spectacular. The corner were quite sharp at f/3.5, but image contrast was low and CA plainly visible. At f/5.6 to f/11, much of the image field off-axis softened (astigmatism or decentered elements?) and the lens had to be stopped down to f/16 to get corner sharpness. I cannot say whether this behaviour was confined only to my review sample, or is inherent to the optical design. I'd like to add that VR was turned off for all tests, and better results might result with VR operative.

[AI, AIS]

4
(FX: D3)

IR:

2
(DX: D200 modified)

Later development shows this even might apply to the D2X for landscape work. Thus, the venerable 25-50 is able to produce silky-smooth images when it is attached to the D2X. There is a roundness and tactility to the image that makes the 25-50 downright enchanting. You might get slightly sharper images with other lenses, but hardly more pleasing to the photographic, inner eye. The virtual lack of distortion, vignetting, and chromatic aberration helps the 25-50 to project its endearing image quality onto the imaging sensor of the D2X.

On the D3, the 25-50 interacts less gracefully and if you intend to use it on the FX camera, near subjects probably will come out better than landscapes.

IR: Its IR performance is troubled by a tendency for a hot spot. Also, some image degradation is seen, plus severe focus shifts along the zoom range. I would not recommend this zoom classic for IR use.

| Best of the Best Nikkors | Lens Survey | Zoom Lenses Overview | My Equipment |

| To Top | Far Side | Gallery | UV | IR | Links | Personal | Professional | Reviews | Start |

Last Update 19 April, 2008