Because of the short flange distance of the M-series Leicas, there have been difficulties with situations where the light from the lens intercepts the sensor at angles greatly different from ninety degrees. This occurs at the edges (normally mostly the short edge) and corners of the image, and can be pronounced for short lenses. The effect can be seen as magenta or green color shifts in those regions, or occasionally, the “Italian flag” effect, where one side gets a green tinge and the other a magenta one. Leica has been on top of this since the beginning, and in the M9, for lenses that the camera recognized, the firmware did a credible job of producing a corrected raw file.
Has there been any backsliding with the M240? It appears not. The 18mm Super-Elmar is about as difficult a lens as Leica makes – the 16/18/21 Tri-Elmar is easier, even at 16mm, because it is a retrofocus design – so I was interested in seeing if there were any problems.
Here’s a challenging image, with the contrast kicked up to emphasize any problems:
Bar Harbor Sunrise. 18mm SE, f/8 @ 1/350, ISO 200
There’s very little corner color shift. There’s quite a bit of light falloff, but that would occur even with film. It’s pretty easy to fix with a curves layer (I like to leave some to draw the eye to the center of the image):
Good job, Leica. I haven’t done much testing on other wide-angle lenses, but I would expect comparable results. Note: the Canadian images were made with firmware version 184.108.40.206. There’s a new firmware image out: 220.127.116.11. Among other things, it’s supposed to have better lens profiles. I’ve updated my camera, but have done no testing since.
Leave a Reply