By Jenny Neyman
Image sharpness is always a legitimate concern for photographers. It’s our basic starting point when making pictures.
Not every photograph needs to be tack sharp. Some images work best when they’re not sharp everywhere. Consistently making effective photographs requires the ability to achieve sharpness or lack of sharpness when and as desired, rather than casually relying on unpredictable results. Generally, though, even apparently abstract images look better when at least some areas of the image are crisply sharp. With that in mind, this week we’ll look at two of the most important factors affecting sharpness and overall image quality.
Foremost among sharpness-killers is blurring due to camera shake or a subject’s fast motion. Camera shake is best controlled using image stabilization hardware, a sufficiently fast shutter speed, or a steady support, like a tripod, depending on the particular circumstance.
Previously, the usual rule of thumb when handholding a 35-mm film camera was that a shutter speed equal to one focal length was sufficient. As an example, when using a standard, 50-mm normal lens, a shutter speed of 1/60 second or faster was considered to be adequate for handheld use. That rule of thumb assumed that chemical prints no larger than 8-by-10 or 11-by-14 would be made from the relatively large, 35-mm film negative.
This assumption is no longer true for sub-35-mm digital sensors. Images made with these are frequently enlarged to far greater sizes and magnifications. That also magnifies the blurring effect of camera shake that would have been inconsequential during the film era, to the point that previously imperceptible blurring becomes objectionable.
Blurring depends on the effective focal length of the lens. Higher-magnification telephoto lenses show more obvious shake and blur. Because lenses have a higher effective magnification when used with sub-35-mm sensors, the old one focal length rule needs to be modified to take that higher magnification into account. Now, a 1/2x shutter speed is generally recommended when handholding unstabilized, sub-35-mm camera/lens combinations. That means you’ll want to use a 1/100 second or faster shutter speed when handholding a 50-mm lens mounted on an unstabilized APS-C or Micro Four-Thirds camera.
Image-stabilization hardware helps. Some brands, such as Sony, Pentax and Olympus, build image-stabilization hardware directly into the camera body, moving the imaging sensor itself in order to counteract camera shake. Until recently, so-called in-body image stabilization was economical but not considered quite as effective as adding stabilization hardware separately into each lens.
Recent improvements have placed IBIS and lens-based stabilization on par with each other. In fact, Olympus’ unique magnetically suspended sensor built into its OM-D (E-M5) body may be the most effective image-stabilization yet. A good IBIS-based system has several advantages — it’s less expensive, more compact, effective with every lens that fits the camera, and potentially superior optically.
Regardless of the type of image-stabilization hardware you use, you’ll gain some benefit under most circumstances. At a minimum, you should be able to use a shutter speed that’s at least two “stops” slower with static subjects. That means using 1/25 second rather than 1/100 second. This allows you to use a smaller lens aperture and/or lower ISO sensitivity, as needed. Some camera and lens combinations do even better, potentially allowing a shutter speed that’s three or even four “stops” slower than what’s suggested for unstabilized cameras.
You’ll need to experiment a bit to determine the slowest shutter speed that works for you. When used carefully, I’ve been able to get decent results at shutter speeds as slow as 1/4 to 1/6 second using an Olympus OM-D.
Blurring due to fast subject motion, on the other hand, requires a sufficiently fast shutter speed. There’s no other way. Most modern, upper-end cameras feature shutter speeds of 1/2000 second or faster. That’s fast enough to stop nearly any action assuming correct exposure.
Using that fast-enough shutter speed requires either bright light, such as full outdoor sunshine, a very-wide aperture lens, and/or a high ISO sensitivity setting, such as ISO 1600 to ISO 3200. Many modern, large-sensor cameras are now capable of producing good quality images at these previously unusable ISO sensitivities, which allow you to use a faster shutter speed to counteract both camera shake and subject motion blurring.
Generally, larger sensors provide better high-ISO performance, but the relationship between sensor size and high ISO performance is not as clear-cut as it was a few years ago. That’s due to rapid improvements in both sensors and in the image processing chips that take sensor data and convert that data to an image file saved to your memory card.
Previously, there was a noticeable improvement in high ISO image quality whenever a person traded up to the next-larger sensor. For example, there was a noticeable improvement in image quality when comparing high-ISO images made with a Canon G9 premium compact camera (1/1.8-inch sensor) and a Canon 40D using a larger APS-C sensor. In the same way, a full-frame Canon 5D was more usable at high-ISO settings than the sub-35-mm sensor in the Canon 40D.
Over the past two years or so, digital imaging electronics have improved to the point that there’s less benefit when moving to a camera system incorporating the next-larger sensor size. Those improvements can result in cost savings.
As a rough approximation, there are seven major sensor sizes:
v Cellphone cameras, which use the smallest possible sensors and lenses.
v 1/2.3-inch, the smallest regular camera sensor, now used in most consumer-grade, long-zoom and point-and-shoot cameras.
v 1/1.6- to 1/1.8-inch sensors, the next larger size, now used primarily in premium compact cameras, such as the Panasonic LX-7, Canon G15 and S110, and Olympus XZ-2.
v 2/3- to 1-inch sensors, about twice as large as 1/1.6-inch sensors. These are used in top-end compact cameras, like the Fujifilm X20, Sony RX100 and Nikon 1 Series.
v Micro Four-Thirds compact system cameras, made primarily by Olympus and Panasonic, the next-larger common sensor, capable of professional-grade image quality.
v APS-C cameras, typically digital SLR cameras like the Canon Rebels, Nikon D series and Pentax K series. These are about twice as large as those used in Micro Four-Thirds systems and about half the size of full-frame sensors. APS-C cameras are used by a high proportion of serious photographers.
v Full-frame sensor cameras, such as the Canon 5D Mark III and Nikon D800, are at the top end of equipment needed, and/or affordable by anyone other than a full-time working professional photographer.
Most people won’t notice enough improvement to justify discarding a recently purchased camera to jump up one sensor size. That’s actually a positive development. Upper-end modern cameras are now good enough that they’re quite usable without apology for years.
Assuming that you now use a standard, 1/2.3-inch consumer camera, there’s not quite enough image quality benefit to justify buying a 1/1.6-inch premium compact camera, like a Canon S110. Rather than buying a single step up, if you’re intent upon better image quality, then you’re more likely to get noticeable improvement and better value by purchasing a somewhat more expensive camera using a two-thirds- to 1-inch sensor, like the Sony RX100.
Similarly, Micro Four-Thirds cameras, especially those using Olympus’ new 16-megapixel sensor, produce images whose quality is virtually indistinguishable from good APS-C digital SLR cameras, and even better than cameras using 1-inch sensors. Similarly, if you already own a top-end APS-C dSLR camera, like the Pentax K-5 series, then you may not see enough improvement to justify the high cost of a full-frame system.
Of course, if you’re purchasing a high-quality camera in any event, then the calculation’s a bit different. You’ll likely want to compare premium compacts, like the S110 or LX-7, 2/3- and 1-inch cameras, like the Sony RX100, and entry-level Micro Four-Thirds models like Olympus’ E-PM2.
There’s not a huge cost differential among these models. The large-sensor E-PM2 usually sells for about $200 less than the Sony RX100, even though the RX100 uses a smaller sensor, has somewhat lower image quality and cannot accept interchangeable lenses. The real difference is size — cameras using smaller sensors also use physically smaller lenses and are thus smaller, in some cases pocketable.
Local attorney Joe Kashi received his bachelor’s and master’s degrees from MIT and his law degree from Georgetown University. He has published many articles about computer technology, law practice and digital photography in national media since 1990. Many of his technology and photography articles can be accessed through his website, http://www.kashilaw.com.