Recently, a friend of mine who is just getting into SLR photography purchased a couple of lenses for an upcoming family trip to Yellowstone and Grand Teton National Parks. One of the first questions he asked me and another photographer friend was “how can I make sure that I don’t have bad lenses”. The other photographer friend recommended visual inspection as well as taking some test shots at various focal lengths and makes sure that the images weren’t soft.
These are the typical answers any photographer would have given when faced with the same question. Well, technology has provided us with another answer. SLRGear.com has a wonderful set of technical lens reviews available on their website. These lens reviews go way beyond the type of testing that I have seen from any other review source.
The team at SLRGear (which is part of Imaging Resource) have utilized DxO Labs’ DxO Analyzer software to mathematically characterize the optical performance of lens/camera systems. SLRGear then takes this mathematical data and plots it in a four dimensional model that lets you see the optical performance of a lens across it’s entire focal length and aperture ranges. For the amount of data that is generated through the testing, this four dimensional model is a wonderful way of seeing the performance of a specific lens…even if it does take some time to wrap your mind around viewing the data.
Below are two screen captures that I took of the graphs for the Canon 50mm f1.4/USM lens tested first on a Canon 20D (partial frame sensor) and then on a Canon 5D (full frame sensor). Note: since the 50mm lens is a prime lens, these graphs are three dimensional, not four.
So what exactly are you seeing? SLRGear provides the full details on how they test lenses. However, I’ll try to condense it down to the basic details. A picture is taken with each lens/camera combination of an extremely flat surface that has a precisely spaced grid of dots. Multiple pictures are taken at each focal length and aperture setting of the lens/camera. The DxO Analyzer software then analyzes those images to determine the Blur Measurement from 17 points across the frame. This data is then displayed in the two dimensional surface plot seen above—the rectangle is the frame and the color coding and warping indicates the amount of blur. As you adjust the aperture or focal length slider, the graph changes to show which parts of the frame are blurrier than others.
So what value is this? Have you ever bought an expensive piece of glass, taken some pictures and then were upset because they weren’t as sharp as you expected? Well that is probably because the aperture/focal length that you took the shots at is one where the lens performs poorly. If you look at the video of the 50mm F1.4 above, you will notice that the frame has a warped blur signature at f1.4 (blurry at the edges but not as much in the center). But at f2.0, the lens has low and uniform blur across the entire frame. This means to get the most out of this lens, don’t use it below f2.0!
In a nut shell, if you look at the graph of a lens and as you move through the focal length and/or aperture you see the graph “flapping”, it’s probably a poor lens that you wouldn’t want to buy. That lens has a very narrow sweet spot for sharp focus. However, if the graph stays largely still, then the lens has a wide sweet spot for sharp focus (which is probably why it costs $1000 or more…).
Using these blur surface graphs you can now determine the optimal range for lenses that you have in your bag. Or, check the optimal range of a lens before you buy it to ensure it is going to meet your need based upon your typical shooting conditions. If you don’t see your lens listed (make sure you scroll down to the bottom of the page to see all their reviews), make a request so they add it to their list. They also have a great forum for discussing real world experience from lenses.
This is a great tool that you should take the time to learn how to use. You will end up with less frustration after a shoot and more pin sharp images.
Thanks much SLRGear team, keep up the great work on reviewing our lenses!