Sigma 18-300 vs. Nikon 18-300 vs. old Tamron
I ordered the Sigma 18-300mm /3.5-6.3 Macro OS HSM C Lens, but noticed that the images didn't seem to be sharp across the entire frame, so I ordered the Nikon AF-S DX Nikkor 18-300mm f/3.5-6.3G ED VR.
To cut to the chase, the Nikon has better image quality if you care about anything other than the center of the frame. This conclusion applies to the specific copies of the lenses I have when used with my camera.
I did some fairly careful testing at the long end, so I'm posting some specifics here. I shot the photos below at ISO 100 on a Nikon D7200 camera, processed the raw images using dcraw with identical white balance and brightness levels for each shot. I subtracted out a dark frame, but otherwise there is no noise reduction. These are all 100% crops from shots of a target about 32 feet away from the camera, 1 second exposure, f/6.3.
We start with center crops at 300mm from both cameras:
Both lenses look good here. The Nikon has a slightly larger focal length, even though both lenses are set to their maximum focal length which is nominally 300mm for both lenses. The Nikon also manages to collect a bit more light at the center even though both lenses are nominally at f/6.3.
To see how the lenses perform away from the center I tried shooting the same target, but placed it under the leftmost focusing point on the D7200 camera. Here are the results:
Both images are darker because of vignetting. It looks like the Sigma has the edge here; the Nikon seems to have problems with chromatic aberration. For completeness, we also show crops of images using the rightmost focus point:
Not so good. If the lenses were symmetrical, the right and left sides would be equally sharp. Obviously, neither of these lenses are symmetrical, although the Nikon does better than the Sigma.
The D7200 doesn't have focus points at the edge of the frame, so to get samples of the edge of the frame I used the camera's central focus point to focus on an object the same distance away as the target, getting these results. The right hand edge of the following crops is the edge of the image:
These images are not of the same portion of the target because (as noted above) the Nikon has a longer focal length than the Sigma even though both are nominally the same. The Sigma gets worse at the edge of the frame. The Nikon on the other hand, improves a bit over the preceding image.
When I was researching these lenses, I looked at the DxOMark measurements, which indicated that the Nikon was sharper than the Sigma, but the Nikon had a high degree of chromatic aberration at the 200mm and 300mm focal lengths. That made it hard to tell which lens was better at those focal lengths, because it depends on the relative importance of sharpness versus chromatic aberration, which depends on what you are shooting. Looking at the above crops, though, I'd say that the chromatic aberration on the Nikon is not a big problem.
Now let's turn to the 200mm tests. At this focal length, I'm able to add a third lens, the Tamron B003 18-270mm lens. This lens was discontinued in 2010 (to be replaced by another lens with the same zoom range).
As before, these are 1 second exposures at f/6.3, the widest stop available at 200mm on these lenses. Starting with center crops:
The focal lengths are not the same on all the lenses, but that's not really a problem with the lenses. They can all do 200mm; they just don't report focal length accurately enough that you can set them to precisely 200mm using the zoom ring markings or the EXIF data.
With center sharpness, all lenses are acceptable, with the Sigma being the worst and the aging Tamron managing to outperform the Nikon. Let's look at the left side:
The left side of both the Sigma and the Nikon is fairly good, although the Nikon remains ahead. The Tamron is bad. But we already know from the 300mm results that the Nikon and Sigma are not symmetrical. Let's look at the right side:
A look at the Tamron crop compared to the others suggests that superzooms have improved over the years. This Tamron was the first APS-C superzoom to reach 270mm, but the range above 160mm is only marginally usable.
The Sigma also disappoints, being considerably worse on the right side at 200mm than at 300mm. The asymmetry of the results for this lens suggest that other copies might be considerably better. In any case, this lens is going back to the store.
|Well Hello There by evancj|
from My Best Photo of the Week
|Normally Amsterdam by Rudolph Arnstein|
from Your City - Biking
|laboratory glass by bombelpl|
|57 BelAir Chevorlet by TexasGal|
from Car Shows 2018