The D200 is a much faster camera with a more logical ergonomic design. With the Nikon, users can turn the camera on and snap off a shot in a single motion - an action that would require two hands and far more time with the 5D. The D200 can also shoot 5 frames a second, while the 5D can only muster 3. This combined with the D200’s more robust body (not to mention its more affordable price tag), makes it a much more formidable alternative for photojournalists or casual shooters.”
|
|
Canon EOS 5D
|
Nikon D200
|
|
Megapixels
|
12.8
|
10.2
|
|
Price
|
$3,299.00
|
$1,699.95
|
|
Dimensions (inches)
|
6 x 3 x 4.4
|
5.8 x 2.9 x 4.4
|
|
Lens mount
|
Canon EF Mount (1:1)
Canon’s line of lenses is enormous and includes many superb lenses. The EF-S series lenses won’t work with the 5D.
Many EF lenses include an ultrasonic motor, fitted around the optics. The USM increases autofocus speed and pairs it with minimal mechanism noise. Users can also utilize manual and automatic focus simultaneously with USM lenses. |
Nikon F-Mount (1.5x crop)
Nikon covers practically the same focal range as Canon, but has fewer choices within the range. Nikon’s commitment to the DX sensor format means that it has developed some very short-focal-length wide angle lenses specifically for digital DSLRs. Nikon’s image quality matches Canon – each brand has outstanding lenses, and each has inexpensive lenses that are disappointing.
|
|
Dust/moisture seals
|
The 5D is sealed as well as the 20D, according to Canon. That’s not as good as the seals on Canon’s top cameras, the 1Ds Mark II and 1D Mark II n, but considerably better than the Rebel XT. It’s also not as good as the Nikon D200.
|
The D200 features the same level of environmental seals as the Nikon D2X and D2Hs, the top cameras in the line.
|
|
Camera controls
|
Canon tries to pare down the number of buttons on the camera, for a cleaner look and perhaps a less confusing interface. Many controls do double duty.
|
Nikon gives many key functions dedicated buttons, increasing the number of controls and filling up the surface of the camera more than the Canon. The dedicated buttons may be simpler and quicker to use, at least for some users.
|
|
Menus
|
Canon’s menus appear in a long, scrolling list, with a jump feature that speeds up navigation. The entries are color-coded into three major categories: shooting, playback and setup. The Custom item in setup brings up 21 more items for changing the interface.
tyle="font-size:100%;">
|
Nikon’s menus appear in a tabbed interface, which is quick to navigate. Color coding and icons improve the organization. The tabs divide the menu entries into shooting, playback, setup and custom settings.
 |
|
Flash
|
No built-in flash or wireless command without dedicated Speedlite. E-TTL II autoflash compatible with all EX-series Speedlites. Assuming current equipment, the wireless flash automation capabilities of the 5D and the D200 look similar on paper.
|
Pop-up flash with wireless commander built-in. i-TTL compatible only with current Nikon flashes, the SB-600, SB-800 and R200.
|
SPEED / TIMING
Speed and timing tests on the Canon EOS 5D and the Nikon D200 were conducted using a 1GB SanDisk Extreme III CompactFlash card and fully charged batteries.
Start-up to First Shot (Advantage: Nikon D200)
Both the 5D and the D200 start up very quickly – they got their first shots off within 1/5 of a second of the camera being turned on. The 5D took 0.19 seconds and the D200 made an exposure after 0.16 seconds.
Shot to Shot (Advantage: Nikon D200)
The D200 has a clear advantage in this category, with a 5 frame-per-second burst mode. The 5D manages only 3 frames per second. However, the 5D managed 71 JPEGs in a burst, while the D200 petered out after 30 JPEGs. The 5D finished writing those 71 images 21.62 seconds after the last one was shot, while the D200 took 25.16 seconds to write its 30 files.
The Canon’s writing performance is impressive and a real testament to the power of the DIGIC II processor. Considering that the 5D’s files have almost 30 percent more pixels than D200’s files, it’s clear that the 5D has the more impressive architecture – it’s just a shame that the capture rate isn’t a bit faster.
Still, there are many circumstances where shooting 30 frames in 6 seconds will get a shot that shooting 71 frames in 24 seconds will not – in sports and action photography, speed is the thing.
Shutter to Shot (Advantage: Canon EOS 5D)
We measured very little shutter lag in either the 5D or the D200. The 5D rated a 0.01-second delay and the D200 waited 0.03 seconds. Neither delay ought to be significant in practice.
Resolution (Advantage: Canon EOS 5D)
On our single camera reviews, we traditionally test each camera’s resolution by collecting a sequence of images of an ISO 12233 resolution chart at various aperture settings and focal lengths and importing the results into Imatest imaging software. Imatest reads the uploaded images and reports sharpness results in line widths per picture height (lw/ph). LW/PH is similar to lp/ph, the conventional unit of measuring resolution; however, lw/ph takes the size of the recording medium into account.
Having conducted full individual reviews on both the Nikon D200 and Canon EOS 5D, our typical full-chart resolution results can be found here: Nikon D200 / Canon EOS 5D.
For this Head-to-Head review, we evaluated the sharpness of the two DSLRs with standard 50mm f/1.4 prime lenses on both models.
A common argument in favor of APS-sized sensors over full-frame chips is that they have the potential to retain better edge-to-edge sharpness because they read from a cropped segment in the center of the chip rather than the full space of the sensor. To test the validity of this claim as it relates to the 12.8 megapixel 5D and 10.2 megapixel D200, we fitted each model with a standard 50mm f/1.4 prime lens and took our resolution readings from the lower left and upper right portions of the recorded frames. The shots were taken without any in-camera sharpening, shooting the D200 in Normal and the Canon EOS 5D in Neutral Parameters. Both cameras were tested at ISO 100 with the lenses fully open to f/1.4, as well as with them stopped down to f/8. Unfortunately, with the D200’s 1.5X magnification conversion, it was necessary to move the tripod slightly to retain equal compositions.
As you can see from the graph above, the additional 2.6 million pixels spread across the Canon 5D’s full-frame CMOS sensor did yield better resolution results. However, the results are quite close with both lenses open all the way. At f/1.4, the additional pixels on the 5D gave the camera a slight advantage, although comparatively, it is far less of an edge than you might expect from the sensor differential. It’s interesting to note that the 5D performs in the same manner as most digital cameras, with stronger horizontal resolution, while the D200 by contrast exhibits greater resolution in the vertical direction with the standard 50mm lens wide open.
When both prime lenses were stopped down to f/8, the 5D’s performance edge became more pronounced and for the most part, resolution was more consistent on both cameras both vertically and horizontally. At f/8, sharpness was much more in line with expectations, relative to the cameras’ respective resolutions. Corner-to-corner sharpness on the Canon EOS 5D supported the claims of full-frame advocates and at least in this instance, refuted the argument for utilizing the “sweet spot†of the sensor on an APS DSLR.
The 5D does in fact offer more edge-to-edge resolution than the D200, and the difference is not just academic. While the results plotted in the charts above may appear as an insignificant difference between the 10.2 megapixel D200 and the 12.8 megapixel EOS 5D, when observing 8 x 10-inch prints made from each camera – held roughly a foot away – the difference is clearly noticeable. The 5D produces sharper images with more observable detail, even at 8 x 10. The results obviously become more pronounced as the print size is increased.
With the prime lenses, we also took a few shots of the ISO resolution chart, fully framed to its proper 3:2 markings, using the cameras’ out-of-the-box defaults. The overall resolution results here are much more imbalanced. As you can see from the Imatest results below, the Canon EOS 5D is much sharper in both directions than the D200. Although much of this di
screpancy can be attributed to stronger in-camera sharpening on the Canon, neither image holds up to much sharpening post-capture.
Click on the charts to view the full resolution version
Canon EOS 5D - ISO 100, 50mm, f/8
Nikon D200 - ISO 100, 50mm, f/8
Color Fringing: Canon and Nikon 50mm f/1.4 Lenses (Advantage: Canon EOS 5D)
We tested the Canon EOS 5D and the Nikon D200 with a couple of lenses each. We had access to the Canon 24-105mm f/4L IS USM, and the Nikon AF-S VR 24-120mm f/3.5-5.6, which are both high-quality, versatile zooms. For a closer comparison, we looked at both manufacturers’ 50mm f/1.4 prime lenses.
Both companies have been making that lens for decades. And they have made thousands upon thousands of them: 50mm f/1.4 used to be a common kit lens with film SLRs. We expected the lenses to be sharp, and they were. Distortion wasn’t a problem, either. We were surprised by the extent to which they both exhibited color fringing, though. Even at f/8.0, our standard aperture for dynamic range testing, both lenses showed obvious fringing. Here are 200% crops of the sides – not the corners, but closer in, on the sides, from both lenses. These are shots from our dynamic range test, which conceivably creates a situation that is most vulnerable to fringing.
EOS 5D
D200
The shots taken with the D200 show a more distracting degree of fringing, however, the artifacts were visible with both cameras. The fringing apparent in the shots recorded with the 5D and the 50mm lens showed fringing skewed to one side of the shapes, while the numbers on the chart recorded by the D200 were overpowered by discoloration on both sides of the figures.
It’s disappointing to see the color problem in lenses that cost hundreds of dollars. We did not have an opportunity to print these images ourselves; however, well-made 8 x 10-inch prints would likely show the fringing effects. The f/1.4 lenses are appropriate in low light, but for ideal image quality, they aren’t nearly as good as Nikon and Canon’s macro lenses.
Color (Advantage: Draw)
Our typical color tests for individual camera reviews are conducted in controlled studio conditions using a GretagMacbeth color chart and Lowell Softboxes with tungsten bulbs. (If you wish to view individual color results for either camera, click here: Nikon D200 / Canon 5D.) We attempted to replicate our studio tests outdoors, using daylight illumination to examine the cameras’ exterior color capabilities. Both cameras were tested for saturation and color accuracy at each available ISO setting using their most natural color modes. For the EOS 5D, we shot the daylight sequence in its Neutral parameter, while the D200 was shot in its default, Color Mode I.
The graph below is a visual representation of our results, with data for the Nikon D200 in shades of red and data for the Canon 5D in shades of blue. The red and blue bars represent the saturation percentage points away from 100 percent for the D200 and 5D, respectively, and the plotted points represent the color error score. The error scores are not in percentage points, but we place these plots atop this graph to let you see all the data at once. The closer the saturation percentage to 0 (or 100), the more accurate the saturation; the closer the color error score to zero, the more accurate the color.
We found the D200’s colors to be much more saturated straight out of the camera. These slightly embellished tones should produce a more “consumer-friendly†appearance without any post-processing. The 5D’s colors, on the other hand, were a bit under-saturated when using the Neutral parameter (although that jumped to roughly 6-8% over-saturation when using the Standard parameter) yet maintained a greater degree of accuracy throughout the ISO range. The tones produced by the D200 had a greater mean color error and displayed more variance at the extremes of the sensitivity range.
However, when we switched the 5D into its Standard parameter, we noticed the camera tended to blow out single channels of color, while the D200 retained detail, albeit with contaminated color.
We shot a still life of some flowers and opened the images in Photoshop, then broke them down into their individual channels to see the color information in each one.
Canon EOS 5D
Nikon D200
Here’s a crop of the 5D image in RGB:
Next, just the blue channel of the 5D shot:
Next, the D200’s RGB and blue channel:
Next, the D200’s blue channel:
Next, just the red channel from the 5D:
Finally, here’s the D200’s red channel from a similar crop:
ht="300" width="277" />
The switch from Neutral (or Faithful) on the 5D will yield bolder colors straight out of the camera, although it comes at the expense of information. The D200, in both Color Mode I and Color Mode II, offers more detail in the individual channels and is much more editable. However, you’ll notice in the image above, the D200 shifts the purple hue in the violet flower over to a more bluish tone (more cyan than purple). The D200 had a lot of trouble rendering this particular shade. The 5D’s reproduction of the violet flower, on the other hand, is much closer to the flower’s natural color.
To utilize all the detail the EOS 5D is capable of capturing, the camera must be set to RAW and colors must be realized in post-processing. The D200, by contrast, processes JPEGs more favorably with more careful attention paid to the individual channels.
Noise / High ISO Performance (Advantage: Canon EOS 5D)
The D200 performed admirably at low ISO settings, staying with the Canon 5D in terms of noise, color, and dynamic range. However, at higher ISO settings – ISO 400 and beyond – the 5D began to pull away and justify its much steeper price tag.
To test the cameras’ performance at high sensitivities, we shot a succession of exterior images, gradually increasing in ISO. We utilized the noise results attained in our outdoor color tests (garnered with a GretagMacbeth color chart and Imatest imaging software) and plotted those results in the graph below.
As much of our other test results concluded, the D200 remains close to the 5D at ISO 400, with only moderate noise resulting from each camera. By ISO 500 and 640, the 5D’s slight advantage increases significantly and the benefits of the full frame CMOS sensor become apparent. As you can see in the chart above, the 5D’s individual noise progression increases steadily, with the only drastic leap occurring from ISO 1600 to 3200, which is also a jump into the camera’s “ISO Expansion.†This is unfortunately not the case with the D200. The Nikon’s noise graph takes on more of a stair-step-like pattern, displaying its first jump in noise from ISO 500 to ISO 640 and increasing almost exponentially at each step beyond. At ISO 1000, the cameras are in different classes. Images from the D200 are flat and lack the detail recorded in shots at half the sensitivity. The 5D also shows a slight drop in image quality, but still retains the quality characteristics that make the camera’s images desirable.
Shots at higher ISO settings with the 5D contain less noise and more detail, but also show much smoother tonal transitions, which result in far more depth in the images. The 5D is also able to retain color accuracy while ISO is increased. This is unusual for most DSLRs, as shots at high ISOs generally lose some saturation and begin to appear somewhat washed, as is the case with the D200.
To help validate these results, along with illustrate the impact of noise reduction on each of the cameras, we’ve included a series of night scenes below.
Long Exposures / Noise Reduction (Advantage: Canon EOS 5D)
Electronic sensors build up noise during long exposures, and both the Nikon D200 and the Canon EOS 5D offer separate long exposure noise reduction settings. We tested noise reduction throughout each camera’s ISO range and with exposures of up to 30 seconds. In general, we found that the D200 images were noisier. The Nikon noise reduction was more likely to make a visible difference in the image, but the result was still inferior to images shot at similar exposures with the 5D. The 5D’s images started out less noisy and more detailed than the D200’s – even with noise reduction, the D200 couldn’t catch up.
All shots below are 200% crops. Click on any of the images to view the full resolution version.
Images shot with the D200’s long exposure NR mechanism engaged show a greater loss of detail, along with the more visible decrease in noise.
EOS 5D - ISO800, NR-ON
<
/em>
Both cameras built up noise most severely in the blue channel, which is typical of digital cameras, and it is easiest to see the effects of noise reduction in the blue channel as well. We have included a few samples to show what we saw: the D200 at ISO 3200, with and without sharpening, shows very prominent noise. Without noise reduction, the noise shows a “pinpoint†effect, with individual pixels apparently jumping and dropping significantly; with noise reduction on, the noise is more diffuse, decreasing detail and lowering contrast.
The effect is harder to see in images from the 5D. Looking at blue-channel samples from the ISO 1000 shot, it would be hard to guess which is noise-reduced and which is not – they aren’t as noisy as the D200 images, but noise reduction didn’t seem to do much good.
We also ran some tests in our lab to provide some quantitative data on the cameras’ performance during prolonged exposures. The low light shots were captured in a studio setup, with the cameras set to ISO 1600 and with long exposure noise reduction engaged. The D200 was set to Normal Long Exposure NR, while the 5D’s Long Exposure NR was turned “On.â€
The chart below shows our results. The horizontal axis indicates the exposure duration, while the vertical axis displays the corresponding noise at each exposure.
Although the D200 starts out far noisier than the 5D at 1.5 seconds, noise also increases more drastically on the D200 as exposures are extended. The 5D maintains a relatively constant line from 2 to 30 seconds; more noise enters the images at about 20 seconds, but it is not a drastic increase. We unfortunately did not have an opportunity to make any extreme long exposures (30 minutes or longer), though we suspect at some point the compounded heat in the 5D would lead to a more substantial increase in noise. However, at 30 seconds or less, the 5D manages noise much more effectively than the D200.
We have also included full shots from each camera, shot at ISO 1600 with noise reduction OFF. These images are not magnified.
Canon 5D: ISO 1600, 82mm, f/10, 1.6 sec. exp, NR OFF
Nikon D200: ISO 1600, 52mm, f/10, 1.3 sec. exp, NR OFF
Dynamic Range (Advantage: Canon EOS 5D)
We tested dynamic range on both cameras using Imatest software, the same testing suite we use for color and resolution. To conduct the test, we shot a Stouffer 4110 transmissive step chart on a light box. The chart is a strip of silver-based film, with a row of 40 progressively darker rectangles. The darkest rectangle is more than 13 stops darker than the lightest. We shoot the chart such that the brightest box is a pure white.
We feed the cameras’ images into Imatest software, which measures the number of rectangles that can be distinguished in each image. That’s a little more complicated than it sounds, because noise levels rise significantly in the dark rectangles. Imatest can detect steps of dynamic range even when the noise level would overwhelm picture information. To generate results that are relevant to real-world photography, Imatest provides separate ratings for overall dynamic range, and for dynamic range within certain noise ranges. For instance, at ISO 100, the Nikon D200 detected 12.2 stops of dynamic range, but recorded only 7.79 stops at Imatest’s High Quality level. High Quality indicates that image noise was under 1/10 of a stop. The total range includes areas with more than 1 full stop of noise. At Low Quality, with up to one stop of noise, the D200 distinguished nearly 11 stops of dynamic range at ISO 100 – a higher score than the 5D, which came in around 10.5 stops. The low-quality rating is less important than high quality, because a full stop of noise pretty much overwhelms visible data.
The test concluded that the Canon 5D delivers better dynamic range than the Nikon D200, which is not entirely surprising considering the 5D’s full-frame sensor has larger photo sites. This should yield better noise results and more dynamic range anyway. What’s remarkable is how marginal the Canon’s advantage is at low ISO settings – for High Quality, the two cameras’ results are virtually the same at ISO 100. However, as the ISO is pushed, the cameras begin to part ways. Canon does a bit better than Nikon at ISO 200, delivering 0.58 more stops. At 400, the 5D scores 0.86 stops better than the D200, and at 800, the 5D’s advantage jumps to 1.77 stops. At 1600, the Nikon’s noise does it in, and the D200 delivers under 4 stops, giving the 5D a 2.32-stop advantage. At the highest settings, noise jumps for the Canon as well, though it maintains a nearly-o
ne-stop margin.
There is an important point to remember about these figures: we present them as a means of comparing cameras, not as an absolute guide to the range that users will achieve in normal shooting. The numbers we’re presenting represent an absolute best-case scenario, and are much wider than users should expect when making a natural-looking print of a typical scene.
Therefore, we shot some still life settings with both cameras as well to look at how they handle highlights, very saturated color and shadow areas. We wanted to see how dynamic range performance plays out when the cameras aren’t pointed at step wedges.
We put some flowers and a linen tester on a sheet of typing paper and shot a series of exposures, with the goal of getting the paper to look clean and white, while keeping color and detail in the flowers and the pale shadows they cast. There are a couple of challenges in this scene: first, the background starts looking very dingy when it’s underexposed, but the shadows get ragged if it’s too bright; and second, the bright red and purple flowers can saturate single channels.
We bracketed by 1/3 stops, and found something disappointing – the D200 jumped from a bit too dark to too light in 1/3 of a stop. The 5D’s highlights didn’t blow out as much in a single step. To illustrate the difference, we’ve imitated the highlight warning feature on many cameras, and turned the blown out areas black on our sample pictures. The white areas in the 5D images maintain more texture than the D200’s images
D200 darker:
D200 lighter:
EOS 5D darker:
EOS 5D lighter:
The prints we made of the lighter 5D image had much more pleasing whites than the D200 images – there is texture even in the lightest parts, while the lighter D200 image is blown out. The faint texture in the 5D’s whites really doesn’t make the image look darker, but it helps the camera deliver smooth transitions from the pale shadows to the whites. It’s very hard to move from any color to pure white, and in our D200 shots, the edges of the shadows are ragged. We saw the problem in both ISO 100 and 400 shots.
White Balance (Advantage: Nikon D200)
The D200 and the 5D handle white balance about as flexibly as any camera out there. Both offer several presets, plus custom settings and direct Kelvin input. Furthermore, both cameras offer fine-tuning, which is the feature that most distinguishes their white balance systems. The 5D has Canon’s unique two-axis fine-tune mechanism, which allows color shifts on both a blue-amber axis and a green-magenta axis. The D200 adjusts only along the Kelvin scale – a “warm to cool†axis that is the industry standard. Plotting the Kelvin axis on Canon’s Blue-amber/Green-magenta chart would be simple: Kelvin runs diagonally between Canon’s axes.
Canon’s custom white balance picks up white readings from regular saved images, rather than from dedicated readings, as the D200 does. In practice, there isn’t much difference between the two systems, but a user could theoretically save many, many images on the 5D from which to select white balance readings, while the D200 saves just four custom white balance settings.
Both the D200 and 5D allow white balance bracketing. Both allow the user to set the bracket increment, and the 5D allows the user to set the axis on which it brackets.
The difference here again lies in the degree of control afforded by one camera and not the other. While both cameras offer significant customization of color calibration in-camera, the Canon 5D goes a bit beyond and adds a degree of control not present on the D200. While many shooters may disregard the flexible four-directional white balance fine-tune on the 5D - preferring to render these adjustments post-capture - the ability to store an unlimited number of white presets (determined by the capacity of the memory card) may aid the efficiency of those shooters working in multi-light studio setups.
While the Canon offers more flexibility in terms of color calibration, 5D users will unfortunately need it. In practice, the 5D’s Auto and preset white balance settings did not prove to be nearly as accurate as the D200’s. We tested the color balance of both cameras under three types of lighting: daylight, tungsten, and fluorescent. Daylight tests were conducted using the cameras’ Auto WB setting, as well as their Daylight preset. Tungsten and Fluorescent white balance tests were administered using the cameras’ designated preset settings.
We shot the GretagMacbeth color chart under the three types of illumination and ran the results through Imatest Imaging Software. The software exaggerates the white balance error to help observe the color bias. As you can see in the charts below, the D200 and 5D both shift colors similarly, however, in all three setups the variances in the 5D images were much more pronounced.
Daylight test:
Canon EOS 5D - Auto WB
Nikon D200 - Auto WB
Canon EOS 5D - Daylight WB Preset
Nikon D200 - Daylight WB Preset
Fluorescent Light Test:
Canon EOS 5D - Fluorescent Light WB Preset
Nikon D200 - Fluorescent Light WB Preset
Tungsten Light Test:
Canon EOS 5D - Tungsten Light WB Preset
Nikon D200 - Tungsten WB Preset
Portrait (Advantage: Canon EOS 5D)
Images of people are a mainstay of digital photography, for both professional and amateur photographers, so we think it’s useful to directly compare how the 5D and D200 stack up taking portraits. We shot portraits in a number of situations with the D200 and 5D, but to create the particular images specifically discussed in this section, we photographed one of our staff by window light with each camera on a tripod. We took custom white balance readings with each camera and bracketed exposure by 1/3-stops. Because both cameras offer Portrait image presets, we shot the cameras in Portrait at ISO 100 and in their default parameters at both 100 and 400. For Nikon, that’s Normal mode, and for Canon it’s Standard. Because Nikon’s “Normal†and Canon’s “Standard†differ considerably, we also shot the Canon in “Neutral,†which we find closer to Nikon’s Normal, and more useful for general photography. All images were shot as JPEGs.
The D200’s images look best at exposures about 1/3 EV above the metered reading. The 5D looks best at its metered value. For what it’s worth, the cameras’ meters agree.
Our preferred image from the Canon 5D was the one shot on Neutral at ISO 100, without exposure compensation. The transitions from midtones to highlights are smooth, and the highlights retain detail. The color is natural from highlight to shadow. The detail drops off smoothly in the shadows – though it drops off sooner than in the D200 images.
Click on any of the portraits below to view the full resolution image.
The best shot from the Nikon D200 was shot at Normal, ISO 100, 1/3 stop above the metered reading. Seen by itself, the Nikon image looks good, but it suffers in comparison with the Canon. Color gradients are less smooth: while the Canon shot shows steady transitions of tone and hue across the subject’s expansive forehead, the Nikon shot looks more patchy as one’s eye moves from the highlights on the brow and the center of the forehead, to the pinker area above the left eye, to the darker area of his temple. The highlight area itself turns cool gray in the Nikon image, while its color remains warmer and more skin-like in the Canon shot. The Nikon handled the darkest tones very well – it shows more detail in the black sweater than the 5D does.
Consistent with our other testing, we found that the D200’s performance deteriorated much more at ISO 400 than the 5D. At ISO 400, the D200’s colors were less saturated and the gradients were rougher. The 5D’s performance held up much better – though for portraits, we’d shoot each camera at the lowest practical ISO.
The reasons to not use Canon’s Standard setting are nowhere more apparent than in a portrait. Standard boosts contrast and sharpens the image more than is kind in human portraiture. The creases and textures one observes in the Neutral shot turn unambiguously into wrinkles and blemishes in Standard.
It’s more puzzling to contemplate the cameras’ specific Portrait modes. We don’t recommend either of them. Nikon’s is more straightforward: it decreases contrast and saturation, and opens up the shadow detail. The changes don’t improve highlight detail or transitions, though, and left the image flat. Canon’s Portrait mode increases contrast in our test and decreases yellow in the skin tones. The result is awful – magenta-pink cheeks with highlights the cold white of skim milk. We definitely found the results unappealing. The best way to get the benefits that a Portrait mode promises is through post-processing RAW files. We don’t present manipulated RAW files in this review because that would be a test of image editing rather than the cameras, but for perfectly smooth color, rosy cheeks and open, detailed shadows, we’d suggest shooting in the mode that preserves as much image data as is practical and editing the images on a computer. Each shooting situation is unique, and a single Portrait mode can’t address the range of variables a photographer is likely to see.
Image Parameters (Advantage: Canon EOS 5D)
Canon and Nikon have taken different approaches to image parameters, which Canon calls “Picture Styles†and Nikon calls image optimization presets. Nikon’s presets have a lot in common with the typical ones on compact digital cameras. The names would be recognizable to most novices: Normal, Portrait, Vivid, Vivid Plus and Softer. The effects themselves make big changes in images. Vivid and Vivid Plus punch up the colors so much that images shot on those settings really can’t hold up to post-processing, while Portrait and Softer tone down colors a similar amount. The D200 accepts tone curves downloaded from Nikon Capture 4, PC software that Nikon magnanimously sells separately from its cameras.
The 5D’s Standard default parameter boosts saturation and sharpens the image more than Nikon’s Normal setting. Canon’s Landscape, like Nikon’s Vivid and Vivid Plus, can over-saturate bold colors in blue skies, flowers and foliage. However, the Neutral and Faithful picture styles are more practical. Neutral images don’t show the saturation boost or sharpening that Standard images do. In this mo
de the 5D produces images that match the output of the Canon EOS 1D Mark II and Mark II n, and the company suggests that photographers shooting with both a 1D Mark II and 5D can get matched color across bodies. This offers a practical advantage for those shooters who pack both an EOS 5D and 1D Mk II n in their bag – the 5D for resolution and the 1D Mk II n for speed. The 5D’s Faithful parameter is essentially the same as its Neutral setting, although it’s calibrated for accurate results under 5200 Kelvin lighting – an assistance to studio photographers. Both Neutral and Faithful parameters produce images that can handle post-processing.
EOS 5D
D200
D200
D200
However, in terms of default parameters, Nikon’s Normal setting is much less hyped up than Canon’s Standard. The D200’s Normal images look less sharpened and maintain more detail in all three color channels, even in areas of bold color, than the 5D’s Standard. Most users, and particularly ones who adjust their JPEGs, will keep the D200 in Normal. On the 5D, we expect that Neutral or Faithful settings will be more useful than Standard, especially as a starting point for image-processing.
Nikon D200 - Default
Canon EOS 5D - Default
EOS 5D
Of course, the 5D and the D200 are cameras meant for delivering top-quality prints. For many photographers, that means shooting RAW, which makes parameters pretty much irrelevant – the presents might as well be on the user’s computer in the RAW converter, rather than in the camera, because the RAW file preserves the original shooting data and the shooting parameters can be reversed or adjusted in conversion.
Banding – Nikon D200
The D200 can produce image artifacts when significant areas of the frame are substantially overexposed. The banding appears as a repeating pattern of vertical lines and is most pronounced at ISO 400, though it remains visible at other sensitivity settings as well. Nikon acknowledges the banding problem, but the company says it does not affect every D200. D200 owners who see lines in their files can send a sample image to Nikon Technical Support. Once Nikon technicians establish that a given camera has the problem, they authorize the owner to send it in for a free adjustment. For further information, go to: http://support.nikontech.com/cgi-bin/nikonusa.cfg/php/enduser/std_adp.php?p_faqid=13872
Our sample camera produces a visible effect from ISO 160 to 800. When we view the image in Photoshop at 100% magnification, we see vertical stripes in the normally-exposed areas above or below any large blown-out area. The stripes are two pixels wide, and in our samples, slightly lighter than the unaffected areas, but don’t show any color cast. The effect is not impacted by image quality settings.
We printed an image with the problem on an Epson Stylus Photo 2200. When we set the image to fit on letter-size paper (about 80% reproduction), the effect was visible only with a magnifying glass. At 100%, we could see it when we viewed the print from 7 or 8 inches – closer than normal viewing distance. At 160%, or twice the “fit-on-page†magnification, the effect was obvious at a normal viewing distance.
How big a deal is this? Consider what’s required to make it happen: the shot must contain a good-sized blotch of blown-out pixels – exposed at least two stops more than would show any detail at all. Blown-out areas usually make for ugly prints even without fine vertical lines anyway.
On the other hand, many valid and interesting pictures have been taken backlit, or with light sources in the frame. Adding digital artifacts won’t improve them and surely will impact the ability to sell them. We wouldn’t completely downplay the existence of the artifacts as some have attempted. Both of these cameras are high-priced, high-performance models that users rely on for strong images. This is clearly an issue worth taking a hard look at when deciding whether to purchase the camera.
We don’t know what causes the problem, and won’t speculate whether it could be corrected with a firmware upgrade designed to address it. We have not seen the problem in any other camera – we shot the 5D in the same setup we used for the D200, and could not replicate the banding.
Leave Your Comments Below