23.22.17 SONY A7 III CAMERA REVIEW | |
Sony’s a7 III features a new 24-megapixel back-illuminated Exmor R CMOS sensor with up to 15 stops of dynamic range. It has a new front-end LSI and updated BIONZ X processor to improve and speed along image processing. Thanks to this computation punch, you’ll enjoy a native ISO range of 100-51,200 (expandable to 50-204,800). As for autofocusing, 93 percent of the frame is covered by AF points, and there are a lot of them: 425 contrast AF points and 693 focal plane phase detect AF points. According to Sony, both low light focusing speed and tracking speed have been doubled compared to the prior model. You’ll enjoy burst shooting speeds up to 10 fps with AF-C using either a mechanical or electronic shutter. You can store up to 177 JPEGs, 89 compressed RAW or up to 40 uncompressed RAW images during burst shooting. Sony says that while images are writing to buffer memory several camera functions, such as playback and the menu, will remain accessible. There’s now a joystick on the camera body to help you select AF points or you can use the 3-inch touch screen to select your focus point. Like the other models in the a7 line, the a7 III features 5-axis in-body image stabilization good for up to five stops of correction. It has dual card slots, with one slot supporting speedy UHS-II SD cards. Rounding out the a7 III’s features:
The a7 III ships with Sony’s Imaging Edge software for RAW processing and tethered shooting. The company says the newest version of the software (1.1) improves data transfer speed and responsiveness during RAW editing. The a7 III sells for $2,000 for the body or for $2,200 in a kit with the FE 28‑70 mm F3.5‑5.6 kit lens. NOTES FROM THE TIPA TEST BENCHPDN is a member of the Technical Image Press Association which has contracted with Image Engineering to perform detailed lab tests of digital cameras. See here for a full methodological rundown of how Image Engineering puts cameras through their paces. Full res files of every visual in this review are available to download for your pixel-peeping pleasure here. Resolution
This graph shows the loss of contrast (y-axis) as a function of the spatial frequency in line pairs per picture height (x-axis) for different ISO-sensitivities (colored lines). The further to the right a curve stretches before descending, the better the resolution at that ISO. The limiting resolution for each ISO can be found by identifying to the highest spatial frequency which results in a contrast of 0.1, or where the ISO curve crosses the thicker horizontal thicker black line marking 0.1. The vertical pink line is a reference representing half the number of pixels in the sensor height (the Nyquist frequency). Texture loss
An artifact is an alteration in a digital image due to technology or technique of processing. Artifacts stem from noise, compression, and sharpening. This graph plots the calculated difference in digital signal between two methods (DeadLeavesCross & DeadLeavesDirect). The colored lines represent response at different ISOs and in reference to a high-contrast target and a low-contrast target. Values plotted are the Dead Leaves SFR difference against the spatial frequency. The larger the area under the curve, the more artifacts are present. Edge contrast / sharpening
This graph shows the degree of sharpening in the image by representing an over- and undershoot along contrasted edges. The colored lines represent measurements at different ISOs and in high- and low-contrast situations. The size of the dip before the edge (in both depth and breadth) indicates the degree of undershoot; similarly, the amount overshoot is indicated by the height and breadth of the peak. Thus, larger dips and/or peaks indicate that a sharpening effect is visible. OECF VN / visual noise
This chart shows the noise behavior at various ISO-sensitivities (colored lines) as a function of the brightness of the target image, which is indicated by the relative darkness of the circle on the outer edge of the diagram (noise in shadowed areas are above, and in highlights below). The larger the area inside a curve, the stronger the noise. The degree to which noise disturbs the appreciation of an image, depends on the image size and the viewing condition. The right-hand side of the chart shows the visibility of the noise in an image that is displayed 100% on a monitor (VN1). The left-hand half shows the visibility of noise in a 40-cm tall print (VN3).
This chart shows the noise behavior at various ISO-sensitivities (colored lines) as a function of the brightness of the target image. The perception of noise is represented by the area that is encircled by the curve. The larger the area, the stronger the noise. How much the noise disturbs the viewing of an image, depends on the image size and the viewing distance. This chart shows the noise visibility for an image that is displayed 100% on a monitor (VN1).
This chart shows the noise behavior at various ISO-sensitivities (colored lines) as a function of the brightness of the target image. The perception of noise is represented by the area that is encircled by the curve. The larger the area, the stronger the noise. How much the noise disturbs the viewing of an image, depends on the image size and the viewing distance. The chart shows the noise visibility for an image that is about postcard size (scaled to a height of 10cm) viewed at a distance of 25cm. Dynamic Range
Color Reproduction
Color reproduction is shown here in two ways. The upper figure is a chart comparing a reference color (right-hand half of each color patch) directly with the color reproduced by the camera (left-hand half of the color patch). Below is a table that lists the DeltaE of each color patch. Red cells indicated strong color deviations, light green cells represent colors with noticeable deviations, and a dark green field represents a moderate deviation. Automatic white balance
Video
This graph shows the sharpening in the image due to an over- and undershoot along edges. Depending on the size (based on width and height) of the additional emerging area, a lower (shallower additional area) or stronger (higher and narrower additional area) sharpening effect is visible.
This graph shows the loss of contrast (y-axis) as a function of the spatial frequency in line pairs per picture height (x-axis) for two ISO-sensitivities in video mode (colored lines). The further to the right a curve stretches before descending, the better the resolution at that ISO. The limiting resolution for each ISO can be found by identifying to the highest spatial frequency which results in a contrast of 0.1, or where the ISO curve crosses the thicker horizontal thicker black line marking 0.1. The vertical pink line is a reference representing half the number of pixels in the sensor height (the Nyquist frequency).
This chart shows the noise behavior at two ISO-sensitivities (ISO100 and ISO1600) as a function of the brightness of the target image. The amount of noise perceived is reflected in the size of the area encircled by the curves. The larger the area, the stronger the noise and its perception. The degree to which the noise disturbs the viewer, depends on the image size and the viewing distance. This chart shows the noise visibility for a video frame that is displayed 100% on a monitor (VN1).
Start-up time
Continuous shooting
Autofocus (300lx) and Autofocus (30lx)
Autofocus (300lx) Live View
Autofocus (30lx) Live View Autofocus in low light 0.31 seconds for a total shooting time of 0.35 seconds.
The speed of autofocus (orange) in high and low light conditions. The green portion of each bar represents the shutter time lag, which is the time (in seconds) between depression of the release button and the start of the exposure with a camera that is already focused on the target. It excludes the time required to focus on the subject. The height of each bar represents the total shooting release time lag: autofocus and shutter time lag together. Image Quality In Depth The Sony Alpha 7 Mark III is a mirrorless camera with a 24-megapixel full-format back-illuminated sensor. At ISO100, the Alpha 7 Mark III produces images with a resolution of 1964 line pairs per picture height (LP/PH; 98 percent of the theoretical maximum). This very good resolution is consistent throughout the lower ISO range (up to ISO1600: 1860 LP/PH or 93 percent of the theoretical maximum). In comparison the predecessor camera, the Alpha 7 Mark II produced images with slightly poorer resolution: 1869 LP/PH at ISO100, falling to 1625 LP/PH (81 percent of theoretical maximum) at ISO12800. The recently launched sister model, the Alpha 7R Mark III, had a sensor with 42 megapixels and so better resolution (2723 LP/PH at ISO100, 103 percent; and 2528 LP/PH at ISO1600 (95 percent of theoretical maximum)). The Mark III has an ISO range which extends to ISO51200, higher than the top ISO of its predecessor which was ISO25600. Even at the higher end of its native ISO range, resolution in images made by the Alpha 7 Mark III is very good: for example, 1744 LP/PH at ISO 12800 (87 percent of the theoretical maximum), and 1698 LP/PH at ISO25600 (85 percent). At the highest native ISO of 51200, the resolution is 1655 LP/PH (83 percent). However, the extended high ISOs show much lower resolution: for example, 1535 LP/PH (77 percent of the theoretical maximum) at Hi1. Texture reproduction is fairly good, without much evidence of loss of detail at low ISOs. At ISO100, texture reproduction tests measure an MTF50 of 1409 LP/PH in high-contrast areas and 1424 LP/PH in low contrast. The proportion of artifacts in images recorded at ISO100, is 22.0 percent (high contrast) and 18.6 percent in low contrast. This shows slightly improved performance over the Mark II, which showed an MTF50 of 1274 LP/PH in high-contrast parts of the scene at ISO100, with 23.4 percent artifacts, and 1223 LP/PH (15.3 percent artifacts) in low-contrast areas. MTF50 at ISO100 in images captured by the related Alpha 7R Mark III, was 1732 LP/PH in high-contrast areas (14.6 percent artifacts), and 1637 LP/PH in low-contrast scenes (with 22.1 percent artifacts). Like resolution, texture reproduction remains fairly good at ISOs up to ISO1600. At ISO1600, tests measure1296 LP/PH with 23.4 percent artifacts in high-contrast areas, and 1153 LP/PH and 25.1 percent artifacts in low-contrast areas of the scene. At higher ISOs, ISO3200 and above, texture loss is more obvious. For example, at ISO12800, only 747 LP/PH are recorded at MTF50, with 41.4 percent artifacts in high-contrast areas, and 504 LP/PH and 59.3 percent artifacts in low-contrast areas. Texture loss is most apparent in the low-contrast parts of the scene: at ISO3200, 902LP/PH and 35.9 percent artifacts were measured in the low-contrast areas, compared to 1160 LP/PH (24.0 percent artifacts) in high-contrast areas. This is still an improvement over the performance of the Mark II, especially in low-contrast parts of the scene: at ISO3200, this older model produced an MTF50 of 535 LP/PH (15.2 percent artifacts). In the high-contrast areas, the Mark II did about as well as the Mark III does in the more challenging low-contrast areas: 922 LP/PH with 25.2 percent artifacts. Sharpening produced by the Alpha 7 Mark III is mild: at ISO100, overshoot along high-contrast edges is 3.0 percent and undershoot 7.6 percent; along low-contrast edges, the overshoot is 9.0 percent with an undershoot of 15.1 percent. The predecessor Alpha 7 Mark II showed similar sharpening along high-contrast edges at ISO100, with an overshoot of 9.3 percent paired with 8.7 percent undershoot. Along low-contrast edges, the Mark II showed slightly stronger sharpening at ISO100: 13.3 percent overshoot and 16.3 percent undershoot. The Mark III produces a consistent degree of sharpening at ISOs up to ISO12800, at which point the overshoot falls to 2.5 percent and undershoot is measured at 6.0 percent in low-contrast parts of the scene. In high-contrast scenes, the undershoot drops at ISO12800 to 5.0 percent, although the overshoot declines only slightly to 2.1 percent. As texture loss is stronger and the proportion of artifacts higher also at ISO12800, the decrease in sharpening helps avoid creating even more artifacts in the image. Nevertheless, despite decent resolution, the sharpening and texture loss together combine to make image shot at ISOs of 12800 and higher relatively poor compared to images captured by the camera at lower ISOs. ‘Pixel peepers’ are likely to be pleased with the Sony Alpha 7 Mark III: visual noise is not noticeable even at 100 percent enlargement for images shot at lower ISOs up to ISO1600 (where the score is 1.0). This is better performance than the Mark II, the tests of which showed that images shot at ISO100 or 400 would be the only photos without observable levels of visual noise (scores 0.8). In the Mark III, noise would be noticeable but not disturbing even for images shot at ISO3200 (1.1) when viewed at 100 percent. In images captured by the Mark III at the higher ISO of 12800, visual noise would be obvious (1.9), and noise becomes very disruptive in images shot at the highest ISOs. However, this is still an improvement: images produced by the Mark II at ISO12800, for example, showed very noticeable visual noise (score 2.4). In Viewing Condition 2, simulating viewing a postcard-sized print or mobile screen, visual noise would not be noticeable in images captured by the Mark III at most ISOs, although the two highest ISO settings would display just-observable visual noise (ISO25600, score 1.0; ISO51200, score 1.2). This is slightly better than the Mark II, which produced noticeable noise in images shot at ISO12800 and ISO25600 (the Mark II’s top native ISO was ISO25600). In Viewing Condition 3, simulating a large print, visual noise would not be noticeable, except in images shot at ISO25600 (score 1.1) and above. Images captured by the Mark II showed noticeable levels of visual noise in Viewing Condition 3, at ISO6400 (1.1) and above. The dynamic range exhibited by the Alpha 7 Mark III is very good: over 10.1 f-stops (up to 10.3) at all ISOs tested between ISO100 to ISO800. Dynamic range remains high as ISO increases (9.8 f-stops at ISO1600) up to ISO6400 (9.2 f-stops). However, once again, ISO12800 shows poorer results, with a drop of nearly a full f-stop to 8.3. The extended ISOs are even worse: for example, 6.9 f-stops at Hi1. Dynamic range was also excellent in images produced by the Mark II: 10.0 f-stops at ISOs ranging from 100 to 1600, and declining only slightly with higher ISOs until ISO12800, where the dynamic range was 8.0 f-stops. Color reproduction by the Sony Alpha 7 Mark III is accurate, with only one color showing a strong deviation from the original. ∆E ranges from 8.9 at ISO100, to a small range between 9.7 and 10.0 at ISO400 to ISO 1600. The automatic white balance implemented by the Alpha 7 Mark III is excellent, with values between 0.0 and 0.2 at all ISOs measured between ISO100 and ISO25600. Automatic white balance was also excellent in the Mark II (0.1 or 0.2 at ISOs ranging from 100 to 1600, rising only slightly to a maximum of 0.8 (ISO25600) at higher ISOs. The Alpha 7 Mark III is faster than its predecessor. It shoots 10.0 frames per second in JPEG format, capturing up to 172 images in a row before the camera slows down. Burst shooting in RAW format is almost as fast: 9.7 frames per second up to a total of 40 images captured before the camera slows down. In comparison, the Mark II could capture 5.0 JPEGs per second for a total of 70 shots before the camera slowed down, and the same speed of RAWs (5.0) for a total of 28. The Alpha 7 Mark III starts up quickly, in 0.9 seconds. It is slightly faster than its predecessor in autofocus: Autofocus time in bright light takes 0.26 seconds, for a total shooting time of 0.3 seconds. In comparison, the Alpha 7 Mark II autofocus took 0.29 seconds in bright light. These are both very similar results to those measured from the Alpha 7R Mark III. Shutter lag was measured for the Mark II as 0.04 seconds. The Mark III is nearly as fast in low light as in bright conditions: autofocus takes 0.31 seconds for a total shooting time of 0.35 seconds. Video It is not a surprise that the Alpha 7 Mark III continues the Sony tradition of fabulous video quality in a digital still camera. Test results show excellent resolution in frames grabbed from videos shot at low ISO (ISO100): 1183 LP/PH, 110 percent of the theoretical maximum. At high ISO (ISO1600), resolution is nearly the same (1164 LP/PH, 108 percent). Video resolution is a lot better than that shown by the Mark II, which reached only 76 percent of the theoretical maximum of its sensor (412 and 411 LP/PH) at both low and high ISO. Sharpening in video is moderate at both low and high ISO: 10.3 percent and 10.1 percent overshoot, respectively, along high-contrast edges, with 13.2 percent and 12.0 percent undershoot. Along low-contrast edges, video stills show overshoots of 12.2 percent (low ISO) and 12.1 percent (high ISO), and 17.0 percent and 13.8 percent undershoot, respectively. Visual noise would be barely noticeable in video (score 1.0) at 100 percent (Viewing Condition 1) in frames shot at low ISO, but noise would be very noticeable (2.3) at high ISO. Similar trends are visible in other viewing conditions, with low ISO frames displaying visual noise below the threshold of being noticeable (score 0.7 in both viewing conditions), and high ISO frames showing some observable noise (1.2 in Viewing Condition 2 and 1.4 in Viewing Condition 3). Dynamic range is good: 9.0 f-stops at low ISO and 8.0 at high ISO. Automatic white balance is not excellent, with scores of 1.5 (low ISO) and 1.4 (high ISO). Handling – Assessment The Alpha 7 III body feels comfortable in the hand and robust without being heavy. The grip is a good size for medium-size hands, and the textured surface adds to the feeling of a secure grip on a well-made camera. However, at least with some lenses tested, the lens change was extremely stiff. The camera is customizable for those who take the time, with two customizable drive modes for still photography and ten picture profiles (for video). The custom buttons can be easily programmed – as long as the user knows that Sony calls them ‘custom key’. These are set via the long menu, which also has a customizable ‘my menu’, offering a choice of thirty entries. The long menu is fairly typical for Sony. It has five tabs, the first of which has 14 pages, which must be scrolled through forwards (one cannot reach the last page directly by scrolling in reverse from the first page). Each tab is designed to show the current page number and the total pages in each tab, which helps alleviate the fact that some menu pages have only one item (and therefore give an impression of being the last page). The menu can be navigated by using the joystick, the front and rear wheels, or the directional tabs: the touchscreen functionality does not work for menu navigation. As is also typical for Sony, the menu uses unusual abbreviations. The meaning is neither intuitive nor always clear. For example, what the menu item ‘Disp. cont. AF area’ might do exactly, is not entirely obvious to this user. Does ‘Disp.’ mean Display? (if so, why not use ‘show’?) Disperse? Disparage? Does ‘cont.’ mean continuous? Contiguous? Controversial? Similarly, the error message which appears to explain why certain functions are not available, is not always as informative as the user might want. The 10-item quick menu is accessed via the Fn button, and turning the front wheel (or, for some items but not all, the rear wheel) permits setting of elements in the quick menu, a rapid and easy way to change settings. However, not all items displayed can be changed via the quick menu: the drive is displayed but selecting this item sparks the display of a message: ‘P/A/S/M You can change it by turning the mode dial’. Some settings can be accessed via different routes. Thus, for example, in addition to the obvious mechanical exposure compensation dial on the top of the camera body, it is possible to set exposure compensation via the quick menu or the long menu. Having software settings means that exposure compensation can also be set remotely, such as through a mobile phone app. However, having software setting of exposure compensation could also lead to confusion as the mechanical dial remains at 0 even when exposure compensation active. A twist of the mechanical dial will override the menu setting, facilitating re-setting of the exposure compensation – at least for the user who is familiar with the exposure compensation options and the order of prioritization of hardware and software. The monitor is a good size, with a diagonal of approximately 7cm. It does not swivel on a vertical axis but does tilt upward 90 degrees so it can be used horizontally. The monitor also tilts about 45 degrees downward, allowing the camera to be used over one’s head. The display has five options for various amount of information, from an info screen to image preview with very little distracting detail. The electronic viewfinder is bright although the resolution is a tiny bit on the coarse side. The viewfinder does not show a rainbow effect, although some moiré is visible as a colored shimmer when viewing dense patterns such as text. Touchscreen and touchpad (monitor navigation while using the viewfinder) functions can be set for only one of the two to be functional, or both: thus, the focus area can be moved on the touchpad for both still photographs and video. When using the monitor, touching the screen can activate the autofocus mechanism on that spot. One popular feature of some Sony cameras is eye-start autofocus, which autofocusses when the user puts the camera to his or her eye – but unfortunately for the Alpha 7 Mark III, eye-start does not work with ‘native’ E-mount lenses but only an A-mount lens attached with lens adaptor LA-EA2 or LA-EA4. Thus, in order to use this function, the camera owner would have to acquire a lens with a different mount plus an adaptor. The adaptor for the A‑mount lenses permits phase-detection autofocus. The eye-start AF is often confused with eye detection auto-focus. Eye detection is not automatic (even when face detection is turned on) but rather is activated via a custom button. This function is advertised by Sony as functioning even when the subject’s face is not entirely visible and also when the subject is looking down. Informal trials found, to the contrary, that eye detection did not always work when the subject’s faces were turned partially away or with only one eye visible. Profile, back-light, and even normal face-on compositions did not always successfully activate the eye-detection focus, even when the camera has already identified the face as a face. It is possible to magnify the preview while using autofocus, which might help with creating images focused precisely where the focus is wanted. Focus peaking is available, although not focus bracketing. One can bracket with white balance. There are a number of options for in-camera set-up: 14-bit RAW, 16:9 format for still pictures, and eight types of picture effects including pop color, various monochromes, posterization. The high and low boundaries of the auto ISO range can be determined in advance, as can a general rule of how slow the minimum shutter speed may be when using auto ISO. Burst shooting has three speeds: at ‘fast’, exposure may lag behind, resulting in some very slightly underexposed images in the series. At ‘low’ burst speed, exposure is adjusted in advance of the shutter being released. The Sony Alpha 7 III has two card slots for SDHD or SDXC cards; one slot also accepts a Memory Stick Duo. The Alpha 7 Mark III had good connectivity, including a USB micro terminal, a USB Type-C (3.1 Gen 1 compatible), a HDMI micro connection, and jacks for external microphone and headphones. These jacks are under two separate doors which, being on tethers, won’t get lost easily. Further connectivity is ensured through NFC, Bluetooth, and WiFi. The camera can be used when directly connected to power but only if the battery is inserted. The battery can be charged in-camera although only when the camera is off (it does not charge while the camera is on). Video The Sony Alpha 7 III is well-designed for video with many options. It can record in 4K and Full HD, as well as a format which the Help Guide describes converting 4K into MP4 using the MPEG-4 AVC/H.264 codec. The HDMI connection can be used to output 4K movies, including recording on an external device directly while filming, which could alleviate limitations of recording speed of some SD cards. Options also include slow-motion and time-lapse (quick motion) options. HDR video is possible using S-Log3 and S-Log2 gamma curves, which permit a broader range of tweaking with regards to color in post-production. Another option is HLG (Hybrid Log-Gamma), which Sony suggests creates videos with realistic colors. Up to ten different picture profiles can be saved, or the preset profiles can be used. The picture profiles in video comprise elements such as black level, gamma, black gamma, knee, color mode, saturation, color phase and color depth for each of R, G, B, C, M, and Y. Other features for videographers include time codes and marker display options. However, the camera suffers from focus breathing even with the prime lens used for testing, which detracts from the smooth appearance of a video recording. Video can be recorded using either the obvious red-dot video button or the shutter release button (after setting the not-entirely-obvious menu item ‘movie with shutter’ to ‘on’). A very nice option of ‘auto slow shutter’ permits the shutter speed to be automatically slowed to compensate for low light during video recording. The Sony Alpha 7 Mark III performs very well and is in nearly every way improved over the Mark II. This good all-round camera produces excellent images with low noise. Potential purchasers should be aware, however, of some issues beyond the menu quirks of Sony, such as focus breathing in video, and the need for A-mount lenses and an adaptor in order to use the eye-start autofocus. Handling – Bullet Points Body & Buttons
Display and Viewfinder
Menu structure
Functionality
Special features
Connectivity
Video
| |
|
| |
| Total comments: 0 | |