Canon EOS 5Ds R Review

If you want the most extreme image resolution available in a DSLR camera and you want the sharpest images available at that resolution, the Canon EOS 5Ds R is your camera.

With one significant exception, the 5Ds R is identical to the Canon EOS 5Ds sans "R". In respect of your time, I'm not going to present a second 15,000 word camera review just to glean the one essential difference between these two cameras, so I'm suggesting that you to read through the Canon EOS 5Ds review if you have not done so already.

On this page, I present a comparison of the 5Ds and 5Ds R, showing their differences along with the advantages and disadvantages of each. The primary focus of this review is image sharpness, moiré and false color as these are the primary image quality differences photographers must understand to make an informed decision between these two camera models.

What are the Differences Between the 5Ds and 5Ds R?

Back to Top

There is only one difference between the 5Ds and 5Ds R and that is the low pass filter. The 5Ds R features an optical low pass filter effect cancellation while the 5Ds gets a conventional low pass filter. A conventional low pass filter creates a slight image blur, reducing the potential for moiré and false color. The advantage created by the 5Ds R's low pass filter effect cancellation is sharper image quality.

Chuck Westfall (Canon USA) explains the difference between the 5Ds and 5Ds R: "The EOS 5DS uses a conventional low pass filter design in which a single point of image data entering the filter is first separated into two points and ultimately into four points by the time the data reaches the image sensor. By comparison, the EOS 5DS R uses a different low pass filter design in which the single point entering the filter is first separated into two points and then the two points are merged back into a single point by the time the data reaches the image sensor."

Using the optical cancellation technique (vs. complete filter removal) greatly reduces development and implementation costs. The "R" delivers sharper images, but moiré and aliasing are potential side effects, notably in scenes that include patterns repeating at a specific frequency. Fortunately, Chuck expected this phenomenon to rarely occur and indicated that "If it does, it can usually be mitigated during post processing."

Here is CPN's overview of the technology:

"Traditional digital camera designs – including the rest of the EOS range – use a low-pass filter to prevent against digital artifacts appearing in images, such as moiré and false colour. Such filters work by slightly softening the image so that repeating patterns in the subject do not coincide with the grid-like pixel pattern on the sensor. However, this effect involves a small sacrifice in sharpness."

"With its low-pass cancellation filter the EOS 5DS R can use its massive 50.6 Megapixel resolution to the [fullest]; delivering that extra bit of critical sharpness when it really matters. This is done at the expense of an increased risk of repeat-pattern artifacts, meaning the EOS 5DS R is best suited to users working in the studio, where conditions can be carefully controlled, or for those photographers shooting landscape and nature images that do not feature such regular patterns."

"Rather than physically removing the low-pass filter from the camera in the EOS 5DS R, a cancellation filter is used to negate its effect. This ensures the distance from the lens mount to the sensor is exactly the same for both the EOS 5DS and the EOS 5DS R, allowing the cameras to share the same AF sensor mechanism and mirror box assembly."

At review time, the 5Ds R is priced $200 USD higher than the 5Ds for an about 5% increase. For that difference, I do not recommend using price as the deciding factor between these two cameras.

One more 5Ds and 5Ds R difference that is barely worth mentioning is name badging on the camera (and neck strap):

Load Comparison Images

Aside from the results of the optical low pass filter effect cancellation, there are no differences in the physical and optical performances of these cameras including high ISO noise level performances.

The Difference in Sharpness

Back to Top

How big is the difference in sharpness? Sharpness is the 5Ds R's biggest advantage, so this a key question to ask. The difference is noticeable, but seeing the difference with your own eyes is the easiest way to answer that question.

My first suggestion is to review the resolution chart test results comparison between these two cameras. The difference seen here is only slight.

I'm finding the real world differences to vary modestly, but the difference in fine details is generally very noticeable. Here are three outdoor comparisons:

Load Comparison Images

The examples shown above were captured with a Canon EF 11-24mm f/4L USM Lens set to 11mm and f/5.6. This aperture is wide enough to show no diffraction, but f/5.6 along with the 11mm focal length will show significant depth of field as can be seen in the near and far subjects included in some of these images. The 11mm focal length also allowed the subject details to be made small while keeping the distance relatively close, better avoiding the photographic effects of heat waves. The test lens was removed and mounted on the next camera with carefulness to insure that no adjustments to focal length or focus distance were made, though slight camera-mounting-caused tripod movement can be seen. RAW files were processed in DPP using the Standard Picture Style and sharpness set to "1" (very low).

Like to see needles on your spruce trees? The 50.6 megapixel resolution has no problem with that. The first two comparisons (those with spruce trees) are from opposite sides of the same frames. While the sky was not quite completely clear, the sun had no clouds near it during these captures and the identical exposures created the same brightness at a sampled white point in the frame.

The sky was more cloud-filled in the building sample set (bottom row) and the 5Ds R image was brightened in Photoshop by .1. While waiting for a clear sky was an option, the weather forecast predicted nothing close to that (mostly rain) for at least a week out. Sorry, I didn't have that much patience. I wanted to see the differences now, so ... I chose to deal with slight cloud effects. The results still have a significant meaning, but avoid comparing shadow contrast in this set.

The primary comparisons are based on a sharpness setting of "1". That doesn't mean that this setting must be used and images can be tuned as desired. Some may prefer to adjust 5Ds R images down to "0".

In the bottom right example links, we see that increasing the 5Ds results to a setting of "2" produces sharpness similar to the 5Ds R at "1" and the 5Ds R images at "0" resemble the 5Ds image at "1". But, the 5Ds cannot make detail from what doesn't exist in the RAW data. For example, look at the vinyl siding on the press box in the 5Ds S=2 result and compare it to the base 5Ds R result. The lines in the siding remain clear even in the 5Ds R S=0 image, but the 5Ds does not clearly delineate these lines even at S=2. That data does not exist in the 5Ds RAW file.

The clear takeaway is that the 5Ds R delivers a noticeably sharper image and captures more detail. I included the 5D Mark III results in this comparison to show the increased resolution in the 5Ds models. Both 5Ds models have significantly more resolution than the 5D Mark III.

At this point, you are probably thinking that the 5Ds R has your name on it. That might be the right choice, but ... you need to be aware of the potential moiré issue, its avoidance and its mitigation.

What is Moiré and What Does Moiré Look Like?

Back to Top

The Wikipedia definition of moiré is: "In mathematics, physics, and art, a moiré pattern (/mw?r'e?/; French: [mwa'?e]) is a secondary and visually evident superimposed pattern created, for example, when two identical (usually transparent) patterns on a flat or curved surface (such as closely spaced straight lines drawn radiating from a point or taking the form of a grid) are overlaid while displaced or rotated a small amount from one another."

When a subject pattern nearly matches the pixel pattern on the sensor, moiré can be the result. The 5Ds' standard low pass filter is responsible for its slightly lower resolution and slightly softer image quality, but this filter is also responsible for the 5Ds better-avoiding moiré issues than the 5Ds R.

Moiré often shows as waves of false color. Here are a pair of moiré comparison examples, clearly showing the differences between these models (100% crops shown).

Load Comparison Images

The first challenge to creating a moiré comparison is finding a situation that will create a moiré pattern. This was not an easy task. A long exploration determined that I had to use a 14mm focal length at about 300' from this field house to make the brick pattern mesh closely with the sensor's pixel pattern, creating the moiré effect I had been hunting. No, that brick wall is not rippling in the wind and there is no earthquake occurring.

Changing the sharpness setting down to "0" in the 5Ds R image slightly improves the look of the brick wall, but the waves of color remain.

I knew that the vinyl siding on the press box could also be made to create a moiré pattern if I could make the details small enough in the frame. I was eventually successful, but ended up shooting in the back of the parking lot using a 14mm focal length to pull off this quest. Again, reducing the 5Ds R image sharpness does not improve the moiré pattern.

While you may look at these images and say "Yuck!", I knew the troublesome subjects to look for and spent many hours fine tuning the distance and focal length parameters to create these samples. However, I should note that, once the brick wall/long distance/ultra-wide angle focal length situation was found, the moiré pattern proved sticky and hard to eliminate.

Most of today's DSLR cameras employ a low-pass filter to very lightly soften image details to avoid moiré, but ... if the fine detail pattern is just right, moiré can be seen in images from any of these cameras. It is important to note that, as seen in the examples on this page, the 5Ds non-R is not moiré-free (nor is the 5D Mark III), but the 5Ds produces a more-subtle moiré effect because of the slight blur implemented by its low pass filter.

What Common Photo Subjects are Likely Candidates for Moiré Effect?

Back to Top

Moiré is most likely to be seen in sharp/in-focus subjects containing strongly-contrasting lines that create a tight pattern relative to the rows of pixels on the sensor. Subjects creating that situation are moiré pattern candidates.

Causes of moiré patterns are most often manmade and are seldom found in nature.

As I've discovered, moiré is not easy to find, especially among the manmade items in my house and studio. I spent hours photographing various fabrics including clothes in the closets, draperies and upholstery with no significant effects found. I could produce a small amount of false color when the shingles on the house roof were photographed at a hard angle, but ... the effects were barely noticeable. I'm not saying that moiré does not exist here, but ... it has not been easy to find.

Portrait and fashion photographers, especially those shooting in-studio or under strong lighting, may encounter troublesome fabrics. Architectural photographers may encounter moiré in brick walls, siding (vinyl/aluminum/wood/metal), some fencing and large HVAC vent grills when these items are very small in the frame. Another common source of moiré is an LCD panel. Take a picture of a grid of LCD pixels and waves of moiré are often seen in the results.

Again, moiré is seldom found in photographs of nature, but bird feathers are a natural subject that I suspected could be made to show a moiré pattern. Creating an example of this required multiple attempts with multiple samples, but I was able to pull it off. While the moiré is stronger in the 5Ds R, it again shows presence in the 5Ds and even in the 5D Mark III at the same focal length.

Load Comparison Images

You want feather detail? The 5Ds models can deliver it in a big way. The feathers seem to have feathers on them.

How Can Moiré be Avoided?

Back to Top

To reduce or eliminate moiré from images, the sharp, in-focus, high contrast line patterns causing the moiré effect must be softened or their frequency relative to the pixels on the sensor must be changed.

One way to accomplish this is to very slightly soften (blur) the subject that is causing moiré. If the lines in the problem pattern are not sharp, moiré will disappear.

One softening option is to stop down the lens being used to an aperture that shows some diffraction effects. In the case of the 5Ds R, f/11 is very effective for eliminating moiré.

As noted, buildings are a common cause of moiré. Those shooting architecture, cityscapes and similar are among those likely to encounter moiré on occasion. Photographers targeting these subjects are typically looking for lots of depth of field and can easily opt for a slightly-diffraction-softened f/11 aperture (if that was not their first choice aperture). The diffraction acts like a low pass filter. Here is an example:

Load Comparison Images

If the f/11 aperture solution works for you, the 5Ds R is probably your better camera option over the 5Ds. The 5Ds R f/11 images are similar in sharpness to 5Ds f/8 images and can be sharpened in post or in-camera if desired.

Those shooting compositions with moiré-causing patterns caused by a non-primary subject in the frame have another option: use depth of field to make the moiré effect-causing subject go at least slightly out of focus. Focus can be adjusted slightly forward/rearward and/or a wider aperture can be used to narrow the depth of field. If the lens being used is not too sharp at its widest aperture, the wider aperture solution can be used even if the subject (a building, clothing a model is wearing, etc.) is the cause of the moiré pattern.

Another primary way to eliminate moiré is to change the frequency of the lines in the moiré-causing pattern in respect to the pixel frequency of the sensor. Moving closer or farther away, moving up or down and/or moving to the side can be effective at reducing or eliminating the moiré effect on closer subjects. The distance change required is relative to the distance of the subject. If the problem subject is far away (such as the brick field house), more distance and/or angle change will be required for the pattern frequency to be adequately changed (and this distance can sometimes become problematic).

Changing focal length can effectively change the moiré problem-causing pattern frequency. In my brick field house example captured at 14mm, only a 2mm focal length change (to 12mm) was needed.

Load Comparison Images

Changing shooting position and changing focal length both effect the final image. The first affects the perspective and the second affects final image framing. If you are not happy with those changes, shoot a slightly wider focal length and crop the image to the desired framing. Cropping reduces overall image resolution, but ... the cropping headroom provided by the 5Ds and 5Ds R is huge.

Even rotating the camera slightly has the potential to reduce moiré. Not all images work rotated and rotation correction is at least somewhat destructive, but again, the amount of pixels these cameras make available to work with is huge.

Avoiding moiré is complicated by the fact that moiré cannot be seen in an optical viewfinder. Optical lenses including the viewfinder pentaprism do not have the secondary grid needed for the moiré pattern to be created. While an LCD viewfinder could show moiré, it would not have the same grid density that the sensor has, so seeing moiré in an LCD viewfinder would be irrelevant. This means that creating moiré examples requires time consuming trial and error attempts with the image review at 100% on the camera LCD revealing success or failure.

This also means that moiré will often not be noticed in the field. Most photographers are not going to thoroughly review every image at 100% on the LCD review, so moiré will most often be discovered when reviewing images on a computer. A little false color (as seen in the resolution chart results for the 5Ds R) is not going to be noticeable in most images, but the waves of color shown in the above examples are more troublesome. This means that moiré removal during post processing may sometimes be necessary. Fortunately, Adobe Lightroom has an easy-to-use moiré correction slider in the Develop module.

To correct moiré in Lightroom:

• Open the image in Lightroom
• Press "D" (Develop Module)
• Press "K" (Adjustment Brush)
• Check the "Auto Mask" box
• Press "O" to show mask
• Paint the mask
• Press "O" to hide mask
• Adjust the moiré slider as desired

To demonstrate this correction, I took the feather image above into Lightroom for a very brief correction session with the following outcome:

Load Comparison Images

Yes, with some more time spent, I could have done a better job with that correction, but ... I wanted to see what a quick correction attempt would produce. The color problem is mostly gone and some pattern remaining is unavoidable using this technique.

Summary

Back to Top

The 5Ds vs. 5Ds R decision is not an all or nothing one in regards to moiré. It is a more or less decision and the same regarding image sharpness. If you are shooting images of subjects that are likely to induce the moiré pattern and do not have time to work on mitigation in the field (such as when shooting a wedding), the 5Ds is the safe option. That camera delivers very sharp images and will not leave you wanting.

If you want to wring every bit of resolution and sharpness out of the camera and can deal with a little moiré here and there when certain, mostly man-made lines and patterns are in the frame, go for the 5Ds R.

When the 5Ds and 5Ds R were announced, there was no question that one or both of these cameras were going to become my primary-use models. The bigger question was "Which one?" My initial expectation (and preordering plan) was that a pair of 5Ds R models were going to be serving my needs and my final decision was the same. I purchased (retail) a pair of 5Ds R bodies. Aside from some sports action, these will be my workhorse bodies going forward.