DxOMark tests Canon 5Ds, performs less than D810 or a7R

[Jan Anne]

DxOMarks tested the Canon 5Ds packing a whopping 50MP sensor but found it lacking compared to those in the Nikon D810 and Sony a7R, two cams midway or at the end of their lifecycle.

At NG we respect all gear choices so this is not intended as an A versus B bashing, this doesn't mean we can't discuss the sensor technologies used by the different brands however. In this case I'm genuinely feeling sorry for our Canon friends as their favourite brand apparently doesn't stack up to the competition



Read the full review here:
http://www.dxomark.com/Reviews/Canon-5DS-5DS-R-Review-New-top-ranking-Canon-EOS-sensor/Canon-5DS-5DS-R-Comparison-3-EOS-5DS-vs-Nikon-D810-vs-Sony-A7R

[Jan Anne]

DPReview also took a look at the DxO data:
http://www.dpreview.com/articles/3673531883/dxomark-eos-5ds-r-sensor-is-highest-ranked-canon-sensor-yet

They are talking about a whopping 2.5EV DR advantage for the D810, insane.

[stenrasmussen]

Although there is more to photography than DR it is still an area of interest to landscapers for instance. What Canon does really well, AF and Live View to name two, it does better than most. What I would line to see from both of the biggies is more invetion/progress. But I guess the old "never change a winning formula" still rules in the inner decision rooms of both makers...or maybe they got them blinders on 😊

[Akira]

When I saw some sample images of elephants and others shot with 5DS (or SR), I immediately felt, though as expected, that the DR suffered, which is really a shame because Canon excels Nikon in other aspects as Sten mentioned.  I also prefer the external user interface of Canon, although the Nordic people would say that Canon's buttons are too small to operate with the gloves on.

[pluton]

Canon makes good equipment.  I always felt that the finder of the 5D2 was nicer for MF than the Nikon I had at the time.
I have read that Canon is stuck at a certain level of micro circuit size in their sensor designs and fab.  The original Canon strategy of developing and making their own sensors may not be helping them compete with the Sony sensor shop.

[elsa hoffmann]

Thanks for posting this

I was sent a link by a friend who just purchased the 5Ds  http://blog.mingthein.com/2015/07/02/canon-5dsr-review-part-1/

I white paper on a body is just that. And we don't shoot white papers - we shoot pictures

Having said that - give me a proper review and test anytime over a review by some photographers.

[PedroS]

Well, regarding bodies I don't envy what Canon does.
But regarding TCs, oh man, when Nikon will learn to make them...

[Bjørn J]

Lloyd Chambers at diglloyd.com has started his review of the 5Dsr, and so far he likes the incredible details in the photos, especially when using Zeiss Otus lenses. But he also sees the lack of dynamic range, and he says the files don't have the same cleanliness as those from the D810, and the shadows can not be lifted in the same way without inducing noise. This makes it difficult to expose for the highlights and then lift the shadows afterwards - a technique I use all the time on the D810.
It seems the perfect camera still has not been made, and of course it never will be

[Jan Anne]

This makes it difficult to expose for the highlights and then lift the shadows afterwards - a technique I use all the time on the D810.

As did I with the D800E and it always amazed me how much I could pull from just one properly exposed 14 bits NEF, stopped using HDR all together with that camera.

My Sony can't do this either btw, at least not with the software I'm using at the moment, partly because they don't use 14 bits for the RAW files (but they're working on that)

But Ming is on to something with the colours, during sunsets here in the NL there's a lot of purple in the sky which the Canon captures ever so nicely where Nikon renders it as plain blue, not capturing what I saw (and I'm not the kind of guy to use digital color filters to get it back).

[Ilkka Nissilä]

Color photography is always approximate and requires the artist's interpretation of the scene and operations in post-processing which implement the interpretation into an image that others can view. However, there is no telling what other people will see and think when they see it.

Physical differences in the cameras which affect colour reproduction are mainly the differences in the sensitivity spectra of the R, G and B sensor elements. The  reproduced colors can vary greatly depending on how they are calculated from the raw data, but the different wavelengths of the original scene cannot be completely decoupled from each other since the cameras only use three different filters. More could be used and on the printing side inkjets can use many more inks, but the benefits of "true color" reproduction on the camera side are unlikely to be worth the extra complexity unless the purpose is the reproduction of art, for example. In the end the subjective aspects of colour dominate preferences and the artist adjusts the image until satisfied. 

A tradeoff may exist between color specificity and signal-to-noise ratio. A system which records many narrow spectral bands to achieve accurate color reproduction likely loses in sensitivity because the passband transmission of narrow band filters (or ones with sharp cutoffs) is lower than the passband transmission of filters that have gradual rolloff and are less specific.

[Frank Fremerey]

Interesting tool. I practical terms I already found that my D3 is much worse than my D600 in nearly any respect. Which does not make the D3 a bad camera at all. A pity that it is worth near to nothing in Euro Value...

[Jan Anne]

Interesting tool. I practical terms I already found that my D3 is much worse than my D600 in nearly any respect.

Well there's no denying the fact that the latest batch of high res Nikon sensors bring more to the table than just more pixels, when I went from the D3s to the D800E the difference was like night and day in DR and color depth department, as mentioned before it's simply insane how much you can pull from a well exposed 14 bits NEF.

[HCS]

Well there's no denying the fact that the latest batch of high res Nikon sensors bring more to the table than just more pixels, when I went from the D3s to the D800E the difference was like night and day in DR and color depth department, as mentioned before it's simply insane how much you can pull from a well exposed 14 bits NEF.

And there i was, so convinced that what the D3s did was pretty much out of this world  But then again, i was coming from D300/D2h.

I still need to learn this agility in the D810 files. I really need to find me some high DR scenes on my upcoming holiday to try this.

[Jan Anne]

And there i was, so convinced that what the D3s did was pretty much out of this world 

And it still is IMO, but it's strengths are speed and high ISO where the D810 is perfect for slow low ISO stuff like landscapes and such.

The two together are a winning team able to capture nearly all disciplines out there

[HCS]

...
The two together are a winning team able to capture nearly all disciplines out there

So right, only ... i had to sell the D3s to get the D810. But, i find the D810 isn't too shabby in the high ISO area. FPS wise, i hardly ever used my D3s above the 5FPS (Cl) i kinda had it glued to, so not much lost there for me (fortunately).