28mm Landscape Lens Comparison test

Pentax K 28mm f3.5 vs. Pentax M 28mm f3.5 vs. Carl Zeiss Distagon 28mm f2.8 AE

I set out to demonstrate the differences between these three legacy lenses when stepped down for landscapes, but ended up demonstrating their similarities.

Setup

I used two cameras for these tests, a full-frame Sony A7s for the side and corner crops, and a Sony a5100 for the centre-crops (since the pixel density is higher on the a5100). All images are Jpegs straight from the camera with sharpening turned down as far as it will go. In truth, there is probably more variation due to the changing light between shots, than between the three lenses! They were, of course, shot on a tripod with remote control. All lenses were focused at infinity.

Conclusions and Summary

All three of these lenses are excellent for landscape, once stepped down, and offer incredibly similar results as the images above demonstrate. However, they all have their own strengths and weaknesses:

The Carl Zeiss Distagon 28mm f2.8 AE gets very soft in the extreme corners and suffers from the most chromatic aberration. On the plus side, it is also the most contrasty. I had thought that it offered the most centre sharpness, however this test doesn’t seem to confirm that.

The Pentax-M is very sharp and has the least chromatic aberration, it also has the worst barrel distortion of the three lenses tested here. It probably has the worst bokeh of these lenses, but, frankly, none of them have particularly good bokeh anyway.

The Pentax K has the least distortion and is sharp across the frame, but it does show some field curvature and more chromatic aberration than the Pentax-M.

I’m selling the Pentax-M on Ebay, since I can’t justify hanging on to all three 28mm lenses. I may well regret selling it, and I would probably keep it if I had a Full-frame camera with more megapixels than the Sony A7s. With just 12 megapixels, every pixel is precious and I don’t want to lose the extra resolution by correcting for the barrel distortion in post.