The Certainty Of Sharpness

How to test your lens’ sharpness in the digital era
This Article Features Photo Zoom

An example of light falloff when shooting into the sky. The edges are darker than the center. Canon PowerShot G10 at 1⁄1,000 second, ƒ/5 and ISO 80.

On a recent photo safari, a colleague asked me to review the images he had captured with his brand-new, top-of-the-line DSLR and matching high-end lens. He had obtained the camera the day before he boarded a plane to Africa. We had paid a prime fee for a hot-air balloon trip over the Namib Desert that morning, and every one of his captures from the shoot was out of focus. He suspected the new camera was the culprit. So we improvised some tests of the camera using the fold of a map as the focus point. We determined that the camera’s actual focus was significantly behind the area selected in the autofocus sensors; his aerial views of features on the landscape were actually sharp somewhere underground!

Lesson learned—again. Nature photographers need to be sure their cameras and lenses are working at optimum levels before they head into the field—or the air. If you’re going to invest time, effort and money on any subject, you need to have complete confidence in the camera and lenses in your bag. For years I conducted complex, painstaking comparative lens tests for my journals, The Natural Image and The Digital Image. Now, in the era of highly capable, sophisticated DSLRs, you can quickly and painlessly test the performance of your equipment, make adjustments, if necessary, and work with assurance that your camera/lens combination is at the top of its game.

The LensAlign system set up for focus testing.

Common Lens Problems
What causes a photographer to question his or her camera or lenses? Sharpness is usually the first concern, and the complaint I hear most about. To assess a sharpness problem, eliminate the biggest variable—the photographer—first. If you get even one sharp image from a camera/lens combination, the odds are that any problems with sharpness after that result from improper technique.

Beyond operator error, lens sharpness can be compromised in a number of ways. The optics can be poor or faulty from the outset. Dust, dirt, water spots or smudges on the front or rear elements have greater deleterious impact than you might imagine, or the lens could have sustained damage in transit or during use in the field. The camera body plays a big part in perceptions of lens sharpness if the autofocus functions are faulty. These problems manifest themselves in images that aren’t properly defined, low in contrast, or have the focus either behind the intended subject or in front of it.

Other vexing lens problems include fringing (unwanted color around the edges of details); distortion (horizontal and vertical lines are bent); flare and lack of contrast (resulting in flat and uneven lighting); and vignetting (where the light falls off to the edges of the image). So how do we determine if the optics we’re using are giving us the image quality we expect? Before you experiment, check out the manufacturers’ specifications and the reports of independent lens testers to see how equipment like yours has performed in other assessments. An example of the latter is the Photozone Lens Testing section at The information available at this site is very thorough and well presented, but keep in mind that the data presented is specific to the individual lens tested, and yours may vary.

This Article Features Photo Zoom

A very large blowup of the edge of a wide-angle image showing a green fringe. This is easily corrected with imaging software.

Performing Your Own Tests
Let’s look at each lens problem—what it means to your photography, how to recognize and test for it, and what can be done to solve or minimize it. In the days of film, the medium was the limiting factor on sharpness. Today, it’s your lens. It will be sharpest in the center and less sharp out to the edges. As you stop down your lens, it will get sharper to about ƒ/8 and ƒ/11; if you test it at ƒ/16 and ƒ/22, you’ll find that the sharpness falls off. If you need the depth of field offered by the smaller apertures, you may find the compromise on sharpness is acceptable. Wide-angle lenses tend to have the most light falloff (vignetting) at the edges. As you stop the lens down, the light will be more consistent. And it bears repeating: If you don’t use good technique, even the best of lenses won’t give sharp results. These are the givens that must be kept in mind as you evaluate your equipment.

Resolution And Sharpness. A trip or photo session can be completely ruined when you bring back unsharp results. If this is happening with a new lens or camera body, the culprit may be faulty equipment that needs replacement. It’s better to know this before you invest time in an important subject, so I always “field-test” new equipment before I head out on a shoot. For a simple test, set up a tripod and photograph a highly detailed subject across the whole frame and in the same plane. Use a low ISO for the most detail possible. Be sure the tripod is sturdy, a reasonably short shutter speed is employed, and shoot with several ƒ-stops from wide-open to stopped down (ƒ/16). Using a cable release can’t hurt, and lock up the camera mirror if possible. On the computer, inspect the images at 1:1 (100%) to evaluate sharpness and contrast. When testing a new body, pick a trusted lens for the test. If there’s a problem, it will be revealed under these conditions.

You can undertake more formal tests with a resolution chart. The same charts we used in the film days work fine now—sharp is sharp, no matter the medium. The test targets have patterns of fine detail in the center and at each edge. You photograph them at different ƒ-stops in good light from a tripod, using a low ISO and a cable release to eliminate any external degrading factors. Examine the images on your computer monitor at magnifications up to 1:1, then compare these findings with the results shown on a website like Photozone.

A test result using LensAlign showing my EF 500mm lens attached to the Canon EOS 5D Mark II at ƒ/4. It shows that with the camera’s micro adjustment set to -20, it still front-focuses slightly. The same lens attached to the EOS-1D Mark IV was dead-on without any adjustments. The test has been rendered using Photoshop Emboss to make it more readable.

You can conduct extremely detailed testing by using the methods found at Their charts run around $250, and a basic testing kit with evaluation software (Windows only) can be had for $99; their high-end testing systems run from $200 to thousands. For self-testing, charts such as the ISO 12233 are available online from You can also make your own testing chart using the USAF 1951 chart with components found at Use a high-resolution printer and gloss or semi-gloss paper to print your chart.

The bottom line is that for any test, you want to photograph very small, verifiable details that can be enlarged and evaluated on your computer screen. You’ll need to consider how your lens resolves these details both in the center and out to the edges of the capture.

Autofocus Combinations. As was the situation with my friend’s DSLR in Africa, unsharp images result when the autofocus function of the lens/body combination is inaccurate. This is especially noticeable with long telephotos that are often used wide-open. When you have a very narrow depth of field, you want it on the animal’s eye and not somewhere at the back of the head or on the nose. Narrow depth of field also poses challenges with medium telephotos with wide apertures, such as an 85mm ƒ/1.2, and macro lenses. If this is your problem, you’ll get sharp images, but the sharpness will consistently lie in front of or behind the intended subject. This inaccuracy can be resolved in many new DSLRs that offer micro adjustments for autofocus that you can accomplish yourself, even to the point that you can program the camera to automatically make different adjustments for each lens you use.

A way to check your autofocus is with LensAlign, an ingenious system for testing the critical focus of your lens. Set up your camera/lens combination on a tripod located 25 times the focal length of the lens (8 feet per 100mm) away from the calibration system, then autofocus on the LensAlign target. If your camera body has Live View capability, you can view the calibration ruler and make camera adjustments for each lens directly. Otherwise, view the image on your monitor and then make the necessary micro adjustments at the camera. Get more information at

This Article Features Photo Zoom

Examples of two resolution charts that can be used by photographers for lens testing. These can be printed on a high-res printer or purchased online for extreme high-res versions that are fairly pricey. The larger chart is the ISO 12233 and the inset is chart USAF 1951, which needs to be printed out multiple times and arranged on stiff board to make a full lens chart.

Color Fringing. Chromatic aberration is another term for color fringing. This problem occurs when the colors captured by the lens don’t focus at the same distance on the sensor, causing magenta or green edging. If this is minor, you probably won’t even see it. If the fringing is extreme, or if you plan to greatly enlarge the image, then you must deal with the problem. While the technology and glass used in today’s lenses minimize fringing, when it does occur, it easily can be corrected in your imaging software.

Adobe Camera Raw (ACR) has a menu under Lens Corrections that allows the photographer to move sliders to Fix Red/Cyan Fringe and Fix Blue/Yellow Fringe. Enlarge the image in ACR. Lightroom 2 and the new Lightroom 3 beta version also address the fringe by using the Develop module for Chromatic Aberration.

Distortion. This aberration is readily apparent if you notice bowing or convergence of sets of parallel horizontal or vertical lines. The problem is more prominent with wide-angle lenses vs. telephoto optics. There’s some improvement as you stop down the lens, but you should check the Photozone website to see what’s expected with specific optics. Some corrections of distortion are possible with imaging software such as Photoshop and DxO Optics Pro, which not only corrects distortion, but also tackles color fringing.

Flare And Lack Of Contrast. If the results from your lens are very flat, with little contrast, you may have a defective lens, but most likely it’s a result of poor technique. Keep the front and rear lens elements clean, and always use a lens hood. Flare is a result of direct sunlight entering the front of the lens and then bouncing around inside the lens. Today’s glass and special coatings do a lot to prevent these situations.

Light Falloff. Many lens designs have a minor flaw where the brightness is greater in the center than out to the edges. Usually, you won’t notice this phenomenon when vegetation fills the frame, but it’s obvious in a prominent sky, and more likely with wide-angle lenses than longer focal lengths. Photograph a multiple-image panorama with sky included and any problem with light falloff will be evident; the composited image will show alternating bands of lightness and darkness in the sky. Help has arrived in newer cameras with a built-in feature that eliminates the light falloff with each lens. If you don’t have one of these latest DSLRs, you can correct the falloff using imaging software like ACR, Lightroom and DxO Optics Pro, to name a few.

Be Vigilant
Like most technical aspects of photography, you can become so concerned about your lens quality that you lose sight of your creative purpose, so you need to achieve some balance. Of all the problems related to lens performance, sharpness is by far the most critical. A standard part of my editing process for every shoot is evaluating the results in the context of my expectations for the equipment I used under the conditions and environment of the capture. If I suspect a problem, I’ll do a field test and check the resulting images on my computer monitor; then I’ll make my own adjustments or send the equipment for repair, if needed. But more often than not, the cause of unsharp captures is technique—in the field or even when conducting camera/lens tests. So while you should be ever vigilant about keeping your equipment in top-notch shape and evaluating its performance, you should apply even greater effort to finding great subjects, satisfying your creative vision and perfecting your technique in the field.


    You mention, ” And it bears repeating:If you don’t use good technique,even the best lenses won’t give sharp results.” .

    This was a well researched and intirely informative article well written. Pitty no one wants to elaborate on this ” practice good technique “. Perhaps a future article on said quote is due. Quit eluding to, reveal all to those who crave this “GOOD TECHNIQUE”.

    To James G.; I believe the gentleman has addressed “practicing good technique” when he suggests that you use tripod, cable release and even mirror lock-up along with the correct f-stop/shutter speed/ISO settings to get the most sharpness for your efforts.

    That’s without considering the various tests mentioned in the article.

    To DJ: If your camera has mirror lock-up, go ahead and lock it up (simulating no mirror as your hopefully rhetorical question poses) and then take a peek through the view finder. How does it look? Being able to see the image through the view finder is why we have mirrors in our DSLRs.

    When you take a picture, the mirror first has to move out of the way. The slap of that mirror moving away can add unwanted vibrations to the camera, softening the focus as a result. Locking up the mirror before taking the shot allows those vibrations to dampen before the exposure is made.

    To DC- Thanks DC, it’s make sense now. I’ve never used mirror lock-up before so I didn’t know what happens when it does.
    I should probably try things first before I open my mouth next time!

    It may not be the lens/dslr combo. It may be the software. The hardware may in tact but the algorithm may be dysfunctional. Remember, a $10,000 lens may be dressed up as a fancy lens which it is, but it’s not the Hubble telescope which itself has had issues over the years. The algorithm is trying to mimic what the lens sees, but it’s rendering may be the issue.

Leave a Reply

Main Menu