This article discusses a method of lens selection to assist you as a photographer in capturing a higher image “keeper percentage”. I will discuss the analysis method through the medium of street photography, however the technique discussed is applicable to all genres of photography, and therefore can be adjusted to your photographic areas of interest.
It is a simple in-home method of semi-quantitively assessing and comparing the image characteristics lenses produce, and where these characteristics start to break down. The procedure does not require any special measurement devices or complicated techniques, and only takes a few hours to collect and analyze data from the lenses that you already have, or can borrow or rent.
Since it relies on the data you collect, it is not subject to inconsistencies associated with comparing different reviews, measurements and assessments from multiple sources. Of course, this is only one of many figures of merit to consider when selecting a lens for a specific application. However, I find it a convenient way of understanding what I can expect out of the lenses I already have, and therefore better achieve the desired image results.
Working in Japan, I have the pleasure of living in a very camera-friendly environment. Many folks in Tokyo always carry a camera or two on weekends, ever ready for a photograph.
I have become interested in street photography or what I like to term, “spontaneous portraiture”. Therefore, I have decided to analyze the lenses I have in order to select lenses that render the look I am trying to achieve, and accentuate to the variables I want to emphasize.
Some basic criteria for my lens selection-
- A medium telephoto lens for standoff, but not too large as to make subjects uncomfortable.
- A lens with a reasonable f-stop that caters to low light situations, subject isolation and a high enough shutter speed to stop the action as street photography is really a form of action photography. Therefore, a fast lens of f2.8 or better is important to me.
- I am looking for critically sharp subject focus with fine details from 3 to 10 meters. Therefore, a fast autofocus is helpful, but not a must. However, a high contrast image rendering is important to me.
Here is an example of the kind of spontaneous portrait I am attempting to achieve.
Armed with these basic criteria I have selected several lenses that I think are potentially suitable. All are faster primes, but two are manual focus lenses, (I do not want to discriminate against older glass, and like many of us, I grew up using manual focus lenses and emulsion films). Note – If you are a landscape or architectural photographer, you may want to adjust your criteria to emphasize wide(er) angle lenses, critical focus and minimal distortion across the image from 10 meters to infinity, and de-emphasize a fast lens.
The set-up I used consists of a controlled environment to minimize variables that may convolute comparison of lens performance, (the tatami room in my apartment). Of course, real world street results count, but to get there I want to first test, understand comparatively, and select lenses that best meet my requirements. I have used the following:
- All tests performed on the same sturdy tripod with a 3 second shutter delay to reduce vibration.
- All test using the same camera and same CIS sensor I plan to use for street photography, (Nikon D7200).
- All tests done using manual mode with the independent variables being the fixed ISO I usually shoot with for street photography and modulation of the aperture extremes I use, while the dependent variable that is adjusted is the shutter speed in order to roughly meet the same TTL exposure meter reading.
- All photographs recorded in RAW and no post-processing was conducted. Raw images were used for comparison.
- A variety of resolution targets that can be found on the internet, printed out on a high-resolution printer, and taped to the wall.
- A room with controlled lighting that is similar in illumination to what I may find on the street during moderately lit conditions.
- Manual focusing using live view magnified to maximum image size possible.
- VR / OS / IS off on appropriate lenses.
The method consists of examining the enlarged center and edge areas of images taken through the chosen lenses at different apertures. The first set of data was taken at f-stops of f/8.0 and f/2.8, followed by an additional test at f/2.0 for selected faster lenses. You can pick any combination of ISO, aperture and shutter speed you want for this test. The above apertures were selected because while photographing street scenes, I find it beneficial to stay away from the absolute widest lens apertures due to issues with too thin a DOF, too weak lens contrast and stronger edge artifacts as well as vignetting for most lenses in the f/1.4 to f/1.8 range.
There are great lenses that perform well at these apertures, but they are expensive. Note - If shooting at night or in extremely low light situations where bright lights would look more pleasing with nice round bokeh spheres, I will shoot wide open and correct as best as possible in post processing. There is one more method that I decided to use on this test, and there may be differing views on this. I adjusted the distance from the camera to the resolution target in order to maintain approximately the same image size on the sensor. This was done to remove differences in focal length from the assessment. In real life street photography there is an image capture size zone that is desirable to use to minimize post process cropping to maintain as high a resolution, (as many pixels) as possible in the final photograph.
Finally, keep in mind that as I do not possess multiple copies of these lenses, the lenses used may be better or worse than the basic lens mean performance. Like most folks, when I purchase lenses I test them to make sure they are on par or above par as based on subjective literature on the web.
The line-up of lenses was as follows:
- Sigma 50mm f/1.4 DG HSM
- Nikon 60mm f2.8D Micro-Nikkor (macro)
- Nikon 85mm f/1.8D
- Nikon 85mm f/1.8G
- Carl Zeiss Jena 80mm f/2.8 Biometer MC (45mm for FF equivalent)
- Sigma 135mm f/1.8 Art
- Sigma 150mm f2.8 DG Macro
- Mamiya 150mm f/2.8 645 (93mm for FF equivalent)
The targets and full image looked as below, (note that each sheet of paper is A4 – close to letter size):
Below are the results and commentary –
The lenses at f8.0 aperture
No real surprises at f/8.0. All lenses looked acceptable across the field of view. The Nikon 85D, Sigma 150 and Mamiya 150 645 were standouts, while I was a bit surprised at the performance of the Sigma 50 and the Sigma 135 lenses, as I know the Sigma 135 is a highly corrected, strong contrast lens. Note that it can be observed that the Nikon 85D and the Sigma 150 were slightly overexposed in the test. One needs to be careful of this when performing this sort of test to ensure the shutter speed is set properly to achieve uniform exposure from all lenses as ISO and aperture are fixed.
More articles that might interest you