Previous page Next page

Studio Tests - Samsung NX 18-55mm F3.5-5.6 OIS lens

The NX10 is available in kits with two lenses - the image-stabilized 18-55mm F3.5-5.6 OIS zoom, and the fast, compact 30mm F2 'pancake'. On this page we're going to see how the zoom measures up in our studio tests.

The NX 18-55mm F3.5-5.6 OIS is fairly typical for a kit lens, producing reasonably acceptable results at most settings. The corners are very soft towards the wide end at larger apertures, and distortion is quite high, but this is balanced by notably low lateral chromatic aberration. Overall it's pretty well on a par with typical kit lenses such as the Nikon 18-55mm F3.6-5.6 VR or the Olympus M ZD 14-42mm F3.5-5.6. Only Panasonic's Lumix G 14-45mm F3.5-5.6 OIS really does any better, but this is substantially due to the integration of software correction for distortion and chromatic aberration into the overall system design.

At the wideangle end, central sharpness is very high wide open, but the corners are distinctly soft. They sharpen up progressively on stopping down, with the best results obtained around F8-F11. At longer focal lengths sharpness becomes more consistent across the frame, and the lens performs just as well wide open as stopped down. As usual on APS-C, apertures smaller than F11 give noticeable softening due to diffraction.
Chromatic Aberration
Lateral chromatic aberration is extremely low, especially in the visually more disturbing red channel. Indeed it's much the lowest of all kit zooms we've yet tested, and this shouldn't be overlooked as it results in cleaner-looking images.
We consider falloff to become perceptible when the corner illumination falls to more than 1 stop less than the center. There's really nothing to see here at all.
Distortion is distinctly high for an 18-55mm kit zoom, ranging from 2.2% barrel at wideangle, through neutral at 24mm, to -1.9% pincushion at the tele end. Indeed it's very unusual to see such high levels of distortion outside of a superzoom design.

Macro Focus

Macro performance is reasonable but not great - in common with many internal focusing lenses. Our measured magnification is 0.26x, and this is is achieved at a focus distance of about 25cm, giving a working distance of about 13cm between the subject and the front of the lens.

Central sharpness is very good, but at this close distance the corners of our test chart shot are rather soft, although they progressively sharpen up on stopping down. However distortion is low, and there's no visible chromatic aberration.
Macro - 91 x 61 mm coverage
Distortion: very slight pincushion
Corner softness: Moderate
Focal length: 55mm (82.5 mm equiv)

Software Correction of Lens Aberrations

Software correction of lens aberrations has become increasingly common recently - most notably it's fully integrated into the image processing of the NX10's most obvious competitors, the Micro Four Thirds system cameras from Olympus and Panasonic. We were therefore interested in finding out whether there was any evidence for Samsung adopting a similar approach with its NX system.

We've looked into this by comparing the NX10's JPEG output with the corresponding raw files, converted using Samsung's own software (Raw Converter 3), Adobe Camera Raw, and dcraw via conversion to DNG. This allows us to draw conclusions on which corrections Samsung is using. Our conclusion is that the NX system behaves much like conventional DSLRs; neither distortion nor falloff (vignetting) is being corrected, and while colour fringing due to chromatic aberration is suppressed in the camera's JPEGs, it isn't (necessarily) in RAW.

Chromatic aberration suppression in JPEG

Here's a real-world example showing how the camera suppresses chromatic aberration in its JPEG output, in comparison to the raw file converted using Samsung Raw Converter 3 (SRC 3) and Adobe Camera Raw 5.7. The JPEG processing appears to detect and desaturate edge fringing, which works well in the case of high contrast edges, but is less effective in regions where CA is less obvious. In the examples below showing two different crops from the same image, the camera has removed fringing very effectively from the edges of the white bars (upper right), but less so from the tree trunk (bottom left corner). Samsung Raw Converter 3 and Adobe Camera Raw both show fringing (to a greater or lesser extent) in both crops.

Samsung NX10, 18-55mm @ 18mm F7.1  
100% crops, camera JPEG
100% crops, raw + SRC 3
100% crops, raw + ACR

Optical Image Stabilization

The 18-55mm F3.5-5.6 OIS incorporates in-lens Optical Image Stabilization. Like Panasonic's similarly-named technology, there are two modes, but they behave subtly differently. Mode 1 runs the stabilizer when the shutter button is half-pressed, while Mode 2 runs it all the time (and therefore probably has a negative impact on battery life). There's no automatic panning detection, so OIS should probably be turned off for panning shots. When activated the OIS unit is near-silent in operation, but when the lens is taken off the camera it rattles around audibly inside the lens (again like Panasonic).

We've generally found the optical stabilization units in SLR lenses to be pretty effective in real-world use, and to quantify this, we subjected the 18-55mm to our studio image stabilization test, using the long end of the zoom (55mm). 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 perfectly usable for all but the most critical applications.

OIS OFF OIS Mode 1 OIS Mode 2

It's clear that Samsung's OIS is providing some real benefit here: about two stops in Mode 1, and one stop in Mode 2. This is certainly worth having, and more or less on a par with the image stabilized 18-55mm DSLR kit lenses from Canon and Nikon (although it's some way off the current best for in-lens systems, which can provide up to four stops of stabilization). Given that Mode 1 is both more efficient, and likely more battery-friendly, it seems the logical choice (for the 18-55m at least).

Previous page Next page