The new pellicle mirror system in the Sony SLT-A55 is fundamentally different from previous pellicle designs.
Pro DSLRs provide a lot of vital features, many of which have been around since film days: quick, accurate autofocusing; multi-area metering (plus centerweighted and spot, when desired); rapid shooting rates; rugged, well-sealed bodies; excellent image quality, even at higher ISO settings; 100% optical viewfinders; and 3-inch, 920,000-dot LCD monitors, sensor dust removers, handy live-view operation and more.
But there are features that haven’t yet come to the pro DSLR—or that have come only recently—of which you should be aware. In this article, we look at some technologies that have been filtering their way into consumer-level cameras, but that we think will—or should—soon make their way into higher-end pro-level DSLRs. At a glance, you might dismiss some of these features as gimmicks, but look closer, and you’ll start to see serious potential.
SLRs are wonderful, showing you the image formed by the lens, regardless of focal length or shooting distance, and enabling you to see when the image is sharply focused. But along with these benefits come some drawbacks. It takes time—not a lot, but measurable—for the mirror to flip up out of the light path and then back down into viewing position once the exposure has been made. During this “blackout” period, you can’t see anything in the finder, and the camera’s phase-detection AF system can’t function—so the camera must focus, then shoot, then reacquire the subject and focus, then shoot. The mirror’s movement causes vibrations that can reduce image sharpness. And the mirror box, focusing screen and pentaprism add bulk and complexity (and cost) to the camera body.
By using a semitransparent pellicle mirror, SLR designers can provide SLR viewing without having to move the mirror. This eliminates mirror vibration and mirror blackout, as well as the bulk and cost of a moving mirror assembly and associated mechanisms, and can speed up shooting as well. The main drawback is that the pellicle mirror doesn’t transmit all the light from the lens to the image plane, so exposure must be increased to compensate, and the viewfinder image is dimmer than with a conventional SLR.
Use of a pellicle mirror in an SLR isn’t new. Canon offered the Pellix back in 1965, with a fixed pellicle mirror and the selling point of no viewfinder blackout while shooting (as well as no mirror vibration, less weight and minus the cost of a moving mirror mechanism). Of course, that was a 35mm film camera, with no motordrive. In 1972, Canon introduced the pro F-1 High Speed Motor Drive Camera, with a fixed pellicle mirror that enabled a top shooting rate of 9 fps—you could blow through a 36-exposure roll of film in four seconds! That was followed by a model in 1984, which could do 14 fps. In 1989, Canon introduced the EOS RT with a fixed pellicle mirror that reduced shutter-release lag time to just 0.008 seconds—the first AF pellicle-mirror SLR.
Fast-forward to today: Sony’s new SLT-A55 and SLT-A33 are the first digital SLRs to employ pellicle mirrors. That is, they look like DSLRs, but they do away with the pellicle-mirror, dim-viewfinder problem (and pentaprism/pentamirror finder bulk) by employing an eye-level electronic viewfinder instead of an optical SLR finder. Because the mirror doesn’t have to move, the camera can focus and shoot simultaneously for super-quick operation; in Continuous Priority AE mode, the A55 can shoot its 16.2-megapixel images at 10 fps— pro-DSLR speed. Another advantage is that you have convenient eye-level viewing when shooting videos, something the moving-mirror DSLRs don’t provide.
Will pellicle mirrors find their way into pro cameras? Could well be. Current technology can provide speed with a moving mirror: Canon’s EOS-1D Mark IV can shoot 16.1-megapixel images at 10 per second, and Nikon’s D3S can do full-frame, 12.1-megapixel images at 9 fps and 5.1-megapixel, DX (APS-C)-format images at 11 fps—with a conventional moving SLR mirror and with continuous auto focusing for each shot. But these are $5,000 cameras, and neither could be considered compact. The Sony SLT models could herald a new sort of quick, compact pro DSLR sometime in the future.
Low-resolution video had been available in compact digital still cameras for years. The technology finally came to the DSLR, in high definition, with the Nikon D90, and soon thereafter, the Canon EOS 5D Mark II late in 2008. Video in DSLRs has caught on so well that almost all new DSLR models offer it.HD video can be 1920 pixels wide by 1080 lines deep, or 1280 pixels wide by 720 lines deep. (By comparison, standard-definition digital video is 640 pixels wide by 480 lines deep.) Canon’s HD DSLRs, Nikon’s D7000 and D3100, Pentax’s K-5 and Sony’s SLT-A55 and SLT-A33 record 1080p video. Other HD DSLRs record 720p (Nikon’s D3S, D300S, D90 and D5000; Olympus’ E-5 and Pentax’s K-7, K-r and K-x). HD DSLRs also offer different video frame rates. Some feel that 24 fps, which is the same rate as feature films, produces a more cinematic look, others feel that 30 fps, which is the video standard, produces a smoother look. Some cameras will shoot at 60 fps, which provides an even smoother look with action subjects. The Pentax K-5 and K-r do 25 fps, a standard rate for European PAL video, but unusual for the U.S.’s NTSC video.
Video in DSLRs allows you to record motion and sound, as well as pro-quality still images. The image sensors in DSLRs are much larger than those in most HD camcorders, providing better low-light and high-ISO images and a cinematic reduced depth of field that allows the user to create effective selective-focus shots. Based on prices of new models with video compared to their video-less predecessors, the HD feature is almost free, so that makes it a great deal.
That said, there’s a difference between HD video from a DSLR and HD video from a pro HD camcorder. The camcorder generally has a sensor (or sensors, in tricolor cameras) with the resolution of the video: 1920×1080 for full HD camcorders, 1280×720 for HD camcorders. HD DSLRs all have sensors with far higher pixel counts than that, so the images have to be downsampled to 1920×1080 or 1280×720 pixels for video. Image quality suffers in the process, due to the artifacts of downsampling. The CMOS still-camera image sensors also seem vulnerable to the “jello” effect in panned shots—because the images are scanned from top to bottom, a panned subject will waver in the resulting images. Newer DSLRs handle this better than early video DSLRs, but it still can occur and needs to be worked around.
Another consideration is that DSLRs were designed to still photography and aren’t as ergonomically suited for video work, nor do they generally have such camcorder features as eye-level viewfinders (the DSLR finder blacks out during live-view/video shooting, so you have to use the LCD monitor on the back of the camera to compose).
Of course, the real proof is in the pudding, and the quality of video from DSLRs can be outstanding. The problems are most visible with subjects containing fine detail. Shots of people tend not to show problematic artifacts as readily. Professional cinematographers do use HD DSLRs for pro video—you’ve probably heard that the 2009-2010 season finale of the TV series House was shot with Canon EOS 5D Mark II cameras—and commercials and even portions of feature films are also being shot with DSLRs, taking advantage of the good high ISO performance, shallow depth of field, compact camera size (compared to pro HD camcorders), wide range of interchangeable lenses and greatly reduced cost.
Ultimately, we expect that HD video in DSLRs is here to stay, and new high-end models that come along are likely to have the functionality built in. On balance, it’s a useful feature, and as a professional photographer, you can decide how much or how little you want to delve into video.
AF Advancements In Video Mode
One drawback to video with DSLRs has been the slow to nonexistent autofocusing during video recording. DSLRs use phase-detection AF for still photos because it’s quicker and more accurate than contrast-based AF, and is very effective with action subjects. The problem is that the SLR mirror has to be in the down (viewing) position for the phase-detection AF sensor to function, while the mirror has to be in the up (recording) position in the video-recording mode. The upshot is that phase-detection AF hasn’t been available while you’re shooting video with a DSLR.
So DSLR makers have offered two AF options in Live View mode: phase-detection, with the drawback that the live view is momentarily disrupted while the camera focuses; and contrast-based, which reads contrast at the image sensor and doesn’t affect the live view, but which is slower than phase-detection AF. Obviously, phase-detection AF can’t be used during video recording—the video would be disrupted each time the camera focused.
Contrast-based AF has improved a lot in recent years, and Panasonic’s version (in all their Micro Four Thirds System mirrorless cameras) is surprisingly quick. And Nikon’s new D3100 and D7000 DSLRs provide AF-S full-time, contrast-based servo autofocusing during video recording. These AF capabilities aren’t on par with those of pro HD camcorders (and much pro video work is done with manual focusing), but it’s progress since the first video DSLRs couldn’t autofocus at all during recording. Look for better video autofocusing capabilities in future professional-level HD DSLRs.
Sony’s new SLT-A55 and SLT-A33, with their fixed pellicle mirrors and electronic viewfinders, can use the camera’s phase-detection AF for both still and video shooting, with no disruption, even with moving subjects. And you get convenient eye-level viewing via the EVF.
Panasonic states that the Light Speed contrast-based AF system in its new Lumix DMC-GH2 mirrorless camera is faster than high-end DSLR phase detection. We haven’t seen that camera yet, but are eagerly awaiting its arrival.
HDR (high dynamic range) digital photography has become popular over the last few years, thanks to its ability to render remarkable detail from darkest to brightest areas, even in high-contrast scenes, and its special-effects capabilities. The idea is to shoot a series of frames at different exposures, then, using HDR software, combine the best detail from each frame into a single image with a wide dynamic range.
A number of pros have been enamored of the hyperreal look you can create with HDR, and now some cameras provide automatic HDR in-camera. When set up for it, the camera shoots two or three bracketed exposures and processes them into an HDR image automatically. The results aren’t as extensive or controllable as doing it from many frames in-computer with special HDR software, but the convenience can’t be beat—and the results are good. We think this will be a popular feature in pro DSLRs of the future because it allows the photographer to create a “quick and dirty” effect with minimal effort to give a client a sense of the more precise shot that you could make if they want to pay for it. Think of it like having the camera make a mock-up of the effect.
You’re familiar with the process of creating a panorama. These high-impact photographs always generate buzz because they’re unusual and they stand out. The downside is that they can be time-consuming to shoot and to combine in the computer.Sony has offered in-camera Sweep Panorama for a while in some of its compact digital cameras, and the features is now available in the SLT-A55 and SLT-A33. Activate the mode, press the shutter button and sweep the camera across the scene horizontally (or vertically, if desired), and the camera will align the images and stitch them into a wide (or tall) panoramic image.
Now, there’s 3D Sweep Panorama: As you sweep the camera across the scene, it records separate right-eye and left-eye images that are then viewed in 3D on 3D TVs. At Photokina 2010, Panasonic revealed the Lumix G 12.5mm ƒ/12 3D lens for its Lumix DMC-G2 and GH2 mirrorless cameras (it’s a Micro Four Thirds System lens, but only these two cameras can use its 3D capabilities). The compact lens contains left and right optic systems (4 elements/3 groups each), and the resulting processed images appear in 3D on a 3D TV set like Panasonic’s 3D Viera. The lens can be used for moving subjects and static subjects as close as two feet away. The Lumix G2 and GH2 aren’t actually DSLRs (they’re mirrorless interchangeable-lens cameras), but they’re capable of delivering pro performance and image quality, and this is a feature that will interest some pro photographers, especially as more 3D TV sets are sold.
We think Sweep Panorama and 3D Sweep Panorama should be on pro models for the simple fact that they enable high-end photographers to create a specialized look easily and quickly. As a pro, you’re always looking for a way to show a client a high-impact shot that will help get a gig—something that sets you apart.
Tilt/Swivel LCD Monitors
The ability to tilt and/or swivel the LCD monitor during live-view or video operation is a real asset, allowing for easier high-, low- and odd-angle shooting, and more comfortable hand-held video operation. While Live View mode has become a standard feature in almost every DSLR produced today, only a handful have articulated monitors, and only the Olympus E-3 and new E-5 provide one on a pro model. If you want to use live view for handheld work, or do videos with your DSLR, you’ll appreciate a tilting/swiveling LCD monitor, and we may well see them in more higher-end DSLRs in the future. Canon put a vari-angle monitor in its most recent DSLR, the mid-range EOS 60D—the first on an EOS camera. We expect to see the trend continue in higher-end DSLRs.
Electronic viewfinders have had their faults—unusable in dim light or for fast action, low resolution—and even the very latest ones aren’t perfect. But the EVF has come a long way, and replacing the DSLR’s bulky pentaprism, focusing screen and mirror box with one has led to a whole new class of compact interchangeable-lens cameras with DSLR image quality and performance, exemplified by DSLR look-alike mirrorless models that are hitting the consumer market.
While the best EVFs still don’t match the image quality of an SLR’s optical finder, as EVF resolution and frame rate improve, these finders become more and more appealing. In theory, cameras with high-performance EVFs could give pros faster autofocus and higher shooting speeds. An example of the direction we’re talking about is found in the new Sony SLT-A55 and SLT-A33 cameras. These models use the fixed, pellicle mirror that allows light to reach the image sensor and the phase-detection AF sensor simultaneously for quicker performance (the A55 can shoot and focus at 10 fps, the A33 at 7 fps).
Panasonic’s Lumix DMC-G2 mirrorless camera features a touch-screen LCD monitor—basically, you can control many features and functions merely by touching the appropriate icon. With the new DMC-GH2, you even can use Touch AF during video recording—just touch the subject and the camera will focus there.
Pro photographers are well-practiced in using their cameras’ features, of course, but touch-screen capability can speed things up—always a good thing, especially in fast-breaking situations. And it can simplify less familiar features like HD video recording. Plus, touch-screens are just cool to use.