By Joe Kashi, for the Redoubt Reporter
Virtually unlimited photo-graphic resolution is now affordable with Olympus’ new E-M5 Mark II Micro Four-Thirds compact-system camera and its 64-megapixel, high-resolution, tripod-only mode.
That high-resolution mode found in the Mark II does provide noticeably improved image quality when used with sharp lenses. You must use a tripod, though, and only with static subjects.
Used properly under these rather restrictive conditions, the Mark II can make images files whose resolution is comparable to 64-megapixel RAW files with 9,216-pixel long-edge resolution and 40-megapixel JPEG files with 7,296-pixel long-edge resolution. That’s higher resolution than Nikon’s top-end D810 or the new Canon 50-megapixel cameras. Nikon and Canon’s high-resolution cameras theoretically can be used handheld, although a tripod is recommended here, as well, for best results.
I upgraded my E-M5 Mark II to firmware version 1.1, recommended by Olympus to reduce color noise in JPEG files. Even after that upgrade, though, I had to apply some color noise suppression to high-resolution JPEG files. Mark II RAW files processed with DXO’s newly upgraded Optics 10.4 do not exhibit any color noise or any other digital noise.
Very sharp optics are a must to make best use of all those megapixels. Sharp, interchangeable lenses at an affordable price are a major advantage of the M 4/3 system supported by Panasonic, Olympus and other camera makers. We’ll take a close look at how many M 4/3 lenses performed under both normal and demanding test circumstances.
I tested about two dozen M 4/3 interchangeable lenses for this article and reviewed quality test results for many more. During testing, several general factors became apparent in nearly all camera systems.
v With very few exceptions, single magnification “prime” lenses were significantly sharper and had better contrast and color fidelity. Prime lenses are preferred for maximum image quality. M 4/3 sensors are smaller than traditional 35-mm film, so smaller lenses with shorter focal lengths achieve a comparable optical effect. In the case of M 4/3 lenses, a 35-mm film lens would have twice the focal length of the optically equivalent M 4/3 lens. As an example, a 50-mm lens provides “normal” magnification and field of view on a 35-mm film camera. A M 4/3 camera would use a 25-mm lens for the equivalent “normal” optical effect and magnification.
Newer lenses designed for M 4/3 tend to be sharpest in the f/4 to f/5.6 range. Older lenses originally designed for film cameras should be stopped down to about f/8 for best corner-to-corner sharpness. It’s best to test every lens to determine its sharpest aperture range and whether there are any manufacturing defects that cause decreased sharpness in part of your image. I only tested one copy of each lens and noticed that, thankfully, only a few M 4/3 lenses showed mildly degraded sharpness due to manufacturing variation — less, in fact, than with most APS-C lens systems. Because modern consumer lenses tend to be assembled without a great deal of quality control, your results may be better or worse than mine. Overall, though, I’ve noticed M 4/3 lenses seem to have generally better quality control and assembly.
My tests were made of a finely detailed, completely static, woven-fabric subject with a generous quantity of very fine, hard-to-resolve dog fur. The E Mark II camera was mounted on a firm tripod and the shutter released with a remote to eliminate camera shake. Internally, the camera was set to an “anti-shock” setting to eliminate internal vibration caused by the shutter mechanism.
I used the camera at its ISO 200 base sensitivity, with all files saved in both RAW and JPEG file formats. I tested each lens at the camera’s normal 16-megapixel resolution and, to stress each lens, also at the Mark II’s 64-megapixel high-resolution setting. Any lens that was adequately sharp at the 64-megapixel resolution proved more than satisfactory in lower resolution modes.
DXO Optics Pro 10.4 now works easily and smoothly with files made by the Mark II. DXO optimally sharpens those files before transferring them to Adobe Lightroom for fine-tuning and final processing. That extra DXO processing step gives each supported lens every possible optical tweak. In my testing situation, the Mark II seemed to slightly overexpose the images, so I dialed in a -0.7 EV exposure compensation.
Some inconsistent and unexpected decrease in sharpness was noticeable when comparing several images made with the same lens. That made no sense until I realized that there’s virtually no depth of field when a very close subject is photographed at f/5. In that case, even minor misfocus or autofocus on a slightly nearer part of the subject made enough difference to result in a blurry image. As distances increase, misfocus becomes less troubling.
Although some older lenses were just as sharp as modern optics, those older lenses had to be set to a smaller lens aperture to gain good sharpness from center to edge. Older, film-era lenses optimized to resolve fine detail often seemed less sharp initially because of their lower inherent contrast. Carefully post-processing files made with good older lenses resulted in excellent final image files. Doing so required removing chromatic aberration, then increasing microcontrast “clarity” and overall contrast. These are routine corrections readily accomplished with Lightroom or DXO.
Most websites test lenses indoors at very short distances. My tests suggest that many lenses that seem to do poorly when tested with a close target are often much sharper in actual use when subjects are located at least 10 feet away. So don’t accept a poor test result at face value.
It’s difficult and expensive to make lenses that are sharp from edge to edge. A slight loss of sharpness near the edges and corners is to be expected, particularly with ultrawide-angle lenses. Major corner softness is bothersome in many instances, particularly landscape photography.
We’ll finish this discussion next week with the test results.
Local attorney Joe Kashi received degrees from MIT and his law degree from Georgetown University. He has published articles about computer technology, law practice and digital photography in national media since 1990. Many of his articles can be accessed through his website, http://www.kashilaw.com.