Interestingly enough, as spectacular as the DP2m’s lens is, it isn’t very resistant to flare in high contrast situations. When including very bright light sources in the image, it usually ends up in a fairly colourful experience, and sometimes congregating around the light source are many little magenta or green spots (seen in the image on the right). These spots can be observed on all Sigma DP Merrill cameras, and become more apparent when there are darker tones around the light source. As seen in the image below, the spots are a lot less obvious than in the image on the right.



Image Quality

Page 4

Aside from this, the Dp2m has marvellously sharp optics. It feels like the sole purpose of the aperture control is to simply increase or decrease the depth of field. The contrast and resolving power of the lens is incredibly high from it’s widest aperture, at the centre or edges of the frame. The lens’ resolving power, general lack of lateral chromatic aberration and low distortion (slight barrel) makes this lens a real joy to use.

On the right is a MTF chart found the the Sigma DP2m 2012 catalogue PDF.

Representing results at f/2.8, it shows impressive performance across the frame for both contrast and resolving power.

One should therefore try and avoid including bright light sources in the image, thus preventing the very colourful veiling flare and the occasional spots. The veiling flare is much less invasive in B&W. Often the colour flare doesn’t contribute to a big change in tonal brightness when converting to B&W, so a monochrome conversion is often better suited to these situations. This can be seen in the image-slider on the right.

The lens’ proneness to colourful flare is quite a limitation to get used to, at least for me, since I often shoot strongly backlit scenes that usually include bright light sources. However, I’m more prone to shooting B&W anyway, so perhaps it’s something that won’t bother me too often.

If, however, the strange magenta/green spots appear, there is fairly little you can do about it.

The lens design is slightly counter intuitive regarding it’s looks. One would presume, looking at the diagram above, that the light travels through the lens and into the camera from right to left. However, it’s actually the other way round, with the largest optical elements being within the camera. Sigma used two high refractive index lens elements and a glass moulded aspheric element to maximise image quality while retaining a short barrel length, and while I’m no expert at lens design, I can only confirm that the lens is incredibly sharp and offers great micro-contrast.

The image below was taken at an aperture setting of f/2.8. If you mouse-over the image, you’ll see an actual-pixels crop which will show just how sharp the lens is wide open.

As mentioned, I’m no lens design expert, but I know each lens design is more often than not, full of trade-offs and compromises. This isn’t necessarily as terrible as it sounds, and Sigma is very capable of making great lenses, given that it’s a perfectly manufactured sample.

I’ve noticed that the DP2m’s lens, shot wide open, does tend to show some chromatic aberration in the background out-of-focus areas. This can be reduced by stopping down. Mouse-over the image on the right to see the chromatic aberration in the out of focus areas.

The rendering of the background out-of-focus areas is not the best I’ve seen from a 30-ish mm lens, but it’s not terrible either. One just needs to be aware that it can get a little distracting at times, especially since one can’t throw the background out of focus very much with a 30mm f/2.8 lens on an APS sized sensor.

The DP2m’s lens shows an approximate field of view equivalent to that of a 45mm lens on a full-frame 35mm camera. While the field of view might be the same, depth of field at a given aperture is not the same, with the DP2m’s lens proving quite a bit more depth in comparison. After all, it’s a 30mm lens, and not a 45mm lens!

To end off, I want to touch on the DP2m’s lack of sensor-induced colour aliasing. Sensors with CFAs tend to produce false colour aliasing in certain situations, which are artefacts caused by the interpolation needed to create colour images from these sensors. Since the Foveon sensor doesn’t require demosaicing, the colour aliasing usually observed with this process is not present in the DP2m’s images.

On the right are two actual-pixel crops showing a B&W target, perfect for inducing aliasing when photographed. The image at the top is a straight forward SPP processed DP2m file, and below it is an ACR 8 conversion of the same target taken with a Nikon D700. Both images were taken at aperture setting f/4. Sharpening was only applied to the D700 image; a 180% 0.6px rad smart-sharpen in Photoshop. The lens used on the D700 was a Nikkor AF-D 50/1.8, a lens I can assure to be a good and sharp copy.

While this isn’t a scientific comparison by any merit, it does give the viewer an idea of the DP2m’s image clarity and lack of colour aliasing, while the D700 sample clearly shows colour aliasing. Comparatively, the use of the 50mm lens on the D700 pretty much made up for the camera’s lower resolution, and both targets were similar in size on the resulting images.