Nikon AF-S DX Nikkor 18-300mm f/3.5-5.6G ED VR review
The Nikon 18-300mm F3.5-5.6 VR behaves much like other superzooms, which means that plenty of compromises have been made to accommodate the huge focal length range. Most obviously sharpness is relatively poor towards the corners at wideangle, and across the almost whole frame towards the tele end. Distortion is unusually high, and there's quite visible vignetting wide open at wideangle, too.
|Sharpness||At wideangle, only the centre of the frame is particularly sharp wide open, with the corners somewhat soft; best results are obtained at F5.6 - F8. As usual for a superzoom, things improve substantially on zooming in, with best results obtained around 50mm (where the lens performs very respectably indeed). Sharpness deteriorates substantially again towards the telephoto end, where stopping down to F8 or F11 will give the best results - light levels permitting.|
|Chromatic Aberration||Chromatic aberration is reasonably low for a superzoom. It's most visible as mainly red/cyan fringing at the two extremes of the zoom range, although there's also strong yellow/blue fringing around 50mm. Note that many Nikon SLRs will remove this in their JPEG processing anyway.|
|Vignetting||The lens shows quite strong vignetting wide open at wideangle. The problem here is not just the amount (~2 stops), but also the abruptness of the fall-off in illumination at the corners, which makes the effect more noticeable and visually unattractive. However this disappears quickly on stopping down, and is insignificant by F5.6. Vignetting is negligible at intermediate focal lengths, but becomes potentially visible again from 200-300mm, although it's unlikely to particularly problematic most of the time.|
|Distortion||Distortion is extremely high at almost all focal lengths. Like most superzooms, extreme barrel distortion at wideangle quickly gives way to intense pincushion distortion across the rests of the range, which is at its worst around the 105mm setting.|
The 18-300mm's close focusing ability is very respectable; Nikon's specified minimum focus distance is 0.45m, but our measurements indicate 37.5cm using manual focus. Focus confirmation in AF mode is (not unusually) limited to a slightly longer distance - around 40cm on the D3200 we used for testing.
Image quality at F5.6 is poor, with low contrast and strong halation around high contrast edges (often a sign of spherical aberration). But it improves dramatically at F8, and the best central sharpness is seen at F11. In our flat-field chart test, corners are soft wide open and improve gradually on stopping down, with optimal sharpness at F16. Distortion is minimal, but there's visible blue/yellow fringing due to lateral chromatic aberration.
The 18-300mm includes Nikon's 'Vibration Reduction' optical stabilization system, and the company claims that it allows hand-holding at shutter speeds up to four stop slower than usual without seeing the blurring effects of camera shake. The mechanism is effectively silent when operational, with only the stabilization of the viewfinder image betraying the fact that it's running.
To determine the effectiveness of the VR system we subjected the 18-300mm to our controlled image stabilization test, using the wideangle and telephoto settings plus one mid-range focal length (50mm). The subject distance for these tests was approximately 2.5m for 18mm and 50mm focal lengths, and 4m for 300mm; the test camera was the Nikon D3200.
We take 10 shots at each shutter speed and visually rate them for sharpness. Shots considered 'sharp' have no visible blur at the pixel level, and are therefore suitable for viewing or printing at the largest sizes, whereas files with 'mild blur' are only slightly soft, and entirely usable for less-critical applications.
|18mm, VR OFF||50mm, VR OFF||300mm, VR OFF|
|18mm, VR ON||50mm, VR ON||300mm, VR ON|
The 18-300mm's VR system gives mixed results in our stabilization tests. It works well at wideangle, where it only really falls over at the very-slow shutter speed of 1/2 sec. But it's less effective at the telephoto end, which unfortunately is where you need image stabilization the most; most other superzooms perform noticeably better here.
If you look at the graphs closely, one very specific problem becomes apparent, which is unexpected softness of images at shutter speeds around 1/80 sec. Our test shots suggest an inability of the VR system to fully correct shake, mainly in the vertical direction (when the camera is held in landscape format). The crops below illustrate how this looks:
|100% crops from test chart: 300mm, Nikon D800|
|1/160 sec||1/80 sec|
We get exactly the same effect using both the D3200 and D800 as test body, using two different copies of the lens and with several different photographers, so we're confident what we're seeing is real and typical of the lens. The effect starts to become visible at 1/125sec, peaks at about 1/80sec, then becomes less visible as shutter speeds fall further, such that images are only slightly-blurred at 1/40sec. It isn't mirror-induced vibration, by the way; tripod-mounted shots show no such problem.
The result of this is that at 300mm and 1/80 sec, we got no shots that we'd rate as truly usable. At 1/40 sec images look less-blurred, but we'd still not call them properly sharp. The same effect also has a negative impact at 50mm, although the blurring is weaker (due to the lower magnification of camera shake). Sadly the problem isn't confined to studio testing, and our real world images show exactly the same thing.
Real world examples
The examples below show real-world images, one at wideangle where the VR system has been effective, and one at telephoto where it's not done quite so well.
18mm, 1/10sec F3.5 ISO 200
300mm, 1/80sec F5.6 ISO 100
|Nikon D3200||Nikon D3200|
|100% crop||100% crop|
In the wideangle example at left, the image is reasonably sharp and detailed in the focus plane (at least within the context of being shot at F3.5, and the fact that we're looking at a 100% crop from a 24MP camera). Handheld without stabilization this would be hopelessly blurred, so the VR system is doing a decent job in this instance.
In the telephoto shot at 1/80 sec, though, the vertical double-imaging is typical of what we've seen from the 18-300mm, and indicative of under-corrected camera shake. In effect, this limits the usefulness of the lens in comparison to other superzooms. To keep shutter speeds up you have to either shoot wide open and accept the lens's softness, or boost the ISO to compensate (which inevitably means a drop in image quality).
|In the Big Girl Chair by BryanKing|
from Baby Photography
|Ruby Red Dress by cjf2|
from High key portrait with RED
|Follow me home by eaa|
from Shades of Blue in Nature