From our survey, it appears that CameraStuffReview readers think that dynamic range is the most important property of a camera. In any case, more important than resolution, color reproduction or signal-to-noise ratio. We have tested the dynamic range of 60 cameras, and compared the trend that we see in our measurements with the dynamic range measurements on DxO Mark. The measurement methods of DxO and CameraStuffReview—and thus the absolute measurement values—differ, but the trends correspond with each other. What shows up? Large pixels and large sensors do not always produce a large dynamic range. Camera age and technology differences between the camera brands, such as the lack of an anti-alias filter, are potentially more important than the sensor size and the pixel size.
Why is dynamic range so important?
Dynamic range of a camera is the ratio of the brightest light ("whitest white") to the darkest shadows ("blackest black") in a shot.
We express the dynamic range of a camera in stops, whereby each stop is a doubling of the light level. The more stops a camera is able to cover without clipping, the greater the dynamic range and thus the better the camera. The dynamic range is not only important when you photograph a subject with a lot of contrast. In many other situations, such as, for example, when you edit the brightness of the shadows or highlights, even if the dynamic range in the shot is already not that high.
The exposure meter of a camera does not always give the desired result when a small, bright subject is photographed against a dark background. The difference in light that is reflected by the black earth in the background and the white flower in the foreground is in this case more than 14 stops: therefore there is simultaneously underexposure of the shadows and overexposure of the highlights.
Overexposure will sometimes unjustly be confused with insufficient dynamic range.
Usually there is not a case of a subject with too high of a contrast; rather, the exposure meter has selected too long of a shutter time, so that a part of the highlights in the shot are overexposed. Often, making a second picture with a shorter shutter time suffices, but not if the contrast is actually too high for your camera. Further, underexposures of a scene with too great a dynamic range produces a photo with even more blocked up shadows: in the example above, a couple of flowers against an almost completely black background.
Click on the illustration for a larger version.
If the dynamic range of a shot is too low, posterization can occur during editing. The shot above on the right is a dark part of a shot that was made considerably lighter in the post-editing. The 8-bit jpg file that is stored simultaneously with the RAW file in the camera does not withstand the same editing: there are too few colors left, and the posterization is unmistakable.
Usable dynamic range
Finally, we want both the shadows and the highlights to be rendered well, without ugly noise appearing during editing of a shot. Therefore, we differentiate between the total dynamic range of a camera ("the distance between white and black") and the usable dynamic range, so that you won't even have trouble with visible noise if you make the shadows lighter during editing. The usable dynamic range is (sometimes much) smaller than the total dynamic range. A camera with a high dynamic range nearly always gives you shots without bleached-out highlights and/or blocked up shadows. If your camera also has a high usable dynamic range, then no noise appears in post-editing if you make the shadows lighter. We all want that camera.
Two shots made of a scene without high contrast. The shot on the left is made with a camera with a higher usable dynamic range than the camera that is used for the shot on the right (move your mouse over the illustration). When making the two shots lighter, the right-hand shot gets bleached-out highlights, and there is no black left in the shot. The left-hand shot still has black, and the highlights are rendered better, despite significant post-editing. But the thing that's most noticeable is probably the much greater noise in the shot with the lowest usable dynamic range.
How is dynamic range measured?
CameraStuffReview utilizes a similar method for the measurement of the dynamic range as DpReview. The dynamic range will be measured by very precisely exposing a gray scale, so that the lightest area is 14 stops lighter than the darkest area. Then (in our case), Imatest analyzes how many areas in the gray scale the camera is able to distinguish. The more areas the camera is able to distinguish, the higher the total dynamic range of a sensor. The remaining areas are completely underexposed/black or completely overexposed/white.
DxO applies no noise suppression and measures only the total dynamic range of RAW files.
CameraStuffReview measures, also just like DpReview, both the dynamic range of RAW files (for us, without any noise suppression) and of jpg files (obviously with noise suppression) that are stored in the camera. The dynamic range of a RAW file is a better scale for the performance of the camera sensor. The dynamic range of a jpg file is, thanks to noise suppression, usually higher, but strongly dependent on many parameters (image style, contrast, with or without lightening of shadows and other corrections). We measure only the dynamic range of jpg files with the standard settings/image style of the camera. Here, there are great differences between the settings by brand, and you can get better results for jpg files by tweaking. Therefore, the dynamic range of jpg files practically never counts in the final score for cameras. If a camera in our list of reviews scores higher with the RAW file than with the jpg file, then there is space to improve the dynamic range of the jpg file by choosing other camera settings for the jpg conversion.
In our reviews, the best camera scores a 9.9, and the worst camera, a 5.5. For the dynamic range, we calculate a weighted average of measurements at low ISO and high ISO settings. The performance at low ISO values weigh more heavily because most photos are made with them. We do not round off the test results. In our tests, the Canon 7D MK2 scored 0.4 stops higher than the Canon 70D (0.2 stops for DxO) for RAW files without noise suppression. The question is whether you see that difference, and the image editing also plays an important role: The difference in dynamic range for both cameras is clearly visible and measurably greater for jpg files (that are stored in the camera).
Next to the measurements, practice shots will be made under various circumstances in situations with high contrast. Sometimes, we use such a shot as an illustration for a camera review. Usually, they will only be used to check the measurements. The reproducible assessment of the dynamic range does not go as well in my opinion as with Imatest.
Dynamic range: RAW or jpg
With the naked eye, I could clearly see both the flowers in the vases in the foreground and the landscape in the background. For the camera, the contrast was too great: A good exposure of the interior produced a completely white window, and a good exposure of the landscape produced a completely black foreground. The chosen exposure is a compromise. The jpg shot above shows shadows (marked in blue) that have become completely black and at the same time a great many highlights. An edited version of the jpg file (click on the illustration above) looks terrible; the colors are unnatural due to posterization (like watercolors), the highlights are pale, and there is too much noise in the interior.
The RAW file that was simultaneously stored in the camera as the jpg shot above, approximately the same number of shadows that have become completely black are shown. But the highlights are less bleached out in the RAW shot than in the jpg: the RAW file has a greater dynamic range. Click on the picture above. The edited RAW shot shows some of the interior without too much noise, while the exterior looks more natural than in the edited jpg shot. This practice example is made with an APS-C camera that's a few years old. Modern system cameras are capable of more. You would probably improve on this result easily now with all the cameras with a 1-inch sensor that is included in our list of reviews.
Cameras with the very highest dynamic range
The absolute top in dynamic range: Nikon D810, D800E, D750, D4s, Fujifilm X-E1
The Nikon SLR cameras with a large, full-frame sensor (D810, D750) are all among the best system cameras that are currently available if you look at the impressive dynamic range that these cameras deliver. The Nikon D4s not only has a high dynamic range at the low ISO values, it is especially impressive at the high ISO values. The odd duck in the flock is the Fujifilm X-E1: a mirrorless system camera with an APS-C sensor that beats out many cameras with a full-format sensor.
Just below the absolute top when it comes to dynamic range are the Nikon and Samsung cameras with an APS-C sensor and the Sony A7r, with 36 megapixels on a full-format sensor. These cameras are also absolutely recommended as far as dynamic range is concerned. Dream cameras. One and all.
Cameras with a good dynamic range
All super (and equally good): Samsung NX30, Canon 5D mk3, Canon 6D, Nikon D3200, D3300, Sony A77, Olympus OM-D E-M5, E-M1, Panasonic GH4, GM5
When it comes to dynamic range, this group is above average. Click here for an illustration of what this group of cameras is capable: If you first underexpose a shot by 4 stops and then later make it 4 stops lighter in Photoshop, then the edited shot is nearly indistinguishable from the original, properly exposed shot. This is a group of cameras, muddled when it comes to sensor size, that all perform better than the Canon cameras with an APS-C sensor (Canon 7D, 60D, 650D, etc.) that we have reviewed. Count yourself lucky if you have one of these cameras. Even if you may sooner have to take an HDR shot than with the very best cameras from the test if you always want to have shots with beautifully rendered shadows and highlights, it won't often be necessary.
1-inch sensor surprises with dynamic range
The sensor of a camera with a full-frame sensor has a surface 8 times greater than a camera with a 1-inch sensor (the smallest sensor in our test). If there were no technological differences between the pixels on the sensor, then the large sensor under the same conditions (same light source, same aperture and shutter time) captures 3 stops more light than the small sensor. The total dynamic range of a Nikon D810 is, in our measurements at both 100 ISO and at 6400 ISO, indeed about 3 stops greater than the total dynamic range of a Nikon J3. Thanks to smart image editing (contrast, noise), you do not see that difference at the lowest ISO values in practice shots. It's amazing how good the dynamic range of a camera with a 1-inch sensor at 100 ISO or 200 ISO is in comparison with the dynamic range of a camera with a full-format sensor. Of course there is a limit to what is possible with image editing. At 6400 ISO, you see the difference in dynamic range between a Nikon D810 and a Nikon V3 immediately. Even though the difference in dynamic range between the two cameras is still 3 stops, the usable dynamic range of the camera with the small sensor is no longer adequate for being able to make shots without disturbing noise.
And the dynamic range of the rest of the 60 cameras?
In our list of camera reviews, the other results from our dynamic range measurements are listed. There you will find the dynamic range of Canon cameras with APS-C sensors, such as the Canon 7D MK2 and the Canon 70D, and various older models of other brands.
Author: Ivo Freriks
With Camera Review Stuff I hope to make a modest contribution to the pleasure that you get from photography. By testing cameras and lenses in the same way, evluating the results and weighing up the pros and cons, I hope to help you find the right camera or lens.