How Lateral Chromatic Aberration Changes With Aperture
If you have a less-than-perfect lens, you might find that stopping down the lens aperture can substantially change the amount of that nasty color fringing at the edges of the frame. The MTFMapper v7.29 program has some very sophisticated features to measure that color fringing (lateral chromatic aberration), also nicknamed “CA”.
Lateral Chromatic Aberration
CA is the ugly color fringing that you see in the frame edge when you photograph, for instance, tree branches against the bright sky. You may not notice it when you switch to a really good lens or stop the aperture down enough. The color of this aberration can vary considerably, depending upon your lens.
Those color fringes are caused by the lens not bending the different colors of light equally; a ‘white’ light ray gets spread across a region of the sensor, instead of being kept as a point. This is different than axial chromatic aberration, where the different colors of light remain aligned relative to the lens axis, but get focused at different distances from the sensor plane.
Many cameras will attempt to automatically remove CA, if you shoot in jpeg mode. Most modern photo editing programs will let you reduce, if not eliminate it, even if you shoot in Raw format. This CA removal usually costs you in resolution, especially at the frame edges.
Lens reviews on the web lead you to believe that CA is a single number for a given lens, aperture, and sensor edge. False. To understand the true character of this lens defect, you can use MTFMapper, but only if you shoot in Raw format. You also need to use a suitable target. The same website that lets you download MTFMapper has files of resolution charts that you can print.
Uncorrected Lateral Chromatic Aberration from a Poor Lens, Widest Aperture.
The shot above shows how bad it can get with CA. This is example from a super-wide lens in the corner, and without any corrections applied.
Suitable chart to use for your photographs
The chart above shows you what a suitable target looks like. The bigger you can print it, the better. Wide-angle lenses particularly need large charts to get good feedback at realistic shooting distances (mine is 3 feet by 4 feet).
Nikkor 85mm f/1.4 AF-S Lateral Chromatic Aberration
In the testing that follows, I chose my 85mm f/1.4 lens. This lens was tested using a Nikon D610 that has an FX sensor with 5.95 micron pixels. The camera sensor pixel size needs to be entered into the program, or else the measurements will be incorrect. I am using the “micron measurements” program option, instead of the “pixels” or “radial offset percent” options. I use this option, so that it’s easier to directly compare measurements from other lenses or cameras.
Nikkor 85mm f/1.4 with Nikon D610 (FX)
Measurements at f/1.4 show many interesting things. First, notice how non-symmetric the readings are. Secondly, notice that the “Red versus Green” measurement range is entirely different than the “Blue versus Green” measurements.
Nikkor 85mm f/2.0 with Nikon D610 (FX)
Nikkor 85mm f/2.8 with Nikon D610 (FX)
Nikkor 85mm f/4.0 with Nikon D610 (FX)
Nikkor 85mm f/5.6 with Nikon D610 (FX)
Nikkor 85mm f/8.0 with Nikon D610 (FX)
Nikkor 85mm f/11.0 with Nikon D610 (FX)
Nikkor 85mm f/16.0 with Nikon D610 (FX)
I found it interesting that the color shifting actually changed direction with the Blue/Green pixels, going from positive to negative, after stopping down to f/2.8 or beyond. I realize that every lens design may have unique characteristics, but it had never occurred to me that CA might have a cross-over response to aperture.
This analysis is just one lens. The intent is to show you how an arbitrary lens can be evaluated to explore its limitations, at least in regards to chromatic aberration. The MTFMapper program has many additional capabilities, most notably resolution analysis in two dimensions.
The 85mm’s Red/Green CA is essentially imperceptible at any aperture and very different than the Blue/Green response. Most lenses probably favor one color pair over the other.
This is probably a good test to spot poor lens assembly or damaged lenses. Plots that aren’t reasonably balanced about the lens optical center could be easy to spot, indicating a problem such as a tilted lens element.
As I suspected, stopping down the aperture definitely helps reduce CA, at least for this 85mm lens. The whole point in using the free MTFMapper is to enable photographers to evaluate their own lens characteristics.
Since this feature has just become available in MTFmapper, I have a lot of exploring to do. Any broad conclusions I draw from studying a single lens will probably require many future revisions.
This new CA feature has definitely given me extra food for thought.