Another crummy camera from Canon

by philg

Read DxOMark and weep if you have a big collection of Canon EOS lens. This comparison page of sensor performance shows that the new 50 megapixel EOS 5DS, as with previous Canon sensors, has dynamic range performance far short of the Sony sensors inside Nikon and Sony bodies. Canon also loses on color depth and high ISO performance.

8 thoughts on “Another crummy camera from Canon”

  1. Or translated for those with eyes rather than test suites, “Canon EOS 5D looks just as good as everything else”

  2. Havard: Canon EOS 5D looks as good as the 5D II, the 5D III, and the 5DS. But even a pretty cheap Sony, such as the A6000, looks better in most real-world situations.

  3. I wonder if the A7RII with the Metabones EOS adapter would work for you. You get the Sony sensor and Canon lens support including AF, aperture control and IS. Plus a nice tripod foot on the adapter to help avoid bending the camera mount.

    I wish they made a version for Nikon lenses, or Nikon made a version of the A7RII.

  4. We tested the 5DS-R in the real world and our results are completely consistent with DxOMark’s measurements… but the real-world pictures look better than any other camera we’ve ever used.

    DxOMark doesn’t assess megapixels or sharpness as part of their overall camera scores (you’d have to look at separate lens measurements), and the 5DS-R really does show much more detail, even with less expensive zoom lenses.

    By cranking up the noise reduction, you can trade some of that detail for cleaner shadows when you need to.

    The D810 still does well and ISO 64 helps it win some specific scenarios, but owning both, we’re picking up the 5DS-R for everything nowadays.

    We did a few reviews but here’s one: https://www.youtube.com/watch?v=Gq3ISUHsfsQ

  5. Tony: Thanks for that video. Very interesting. You got great results from the 5DSR by being an expert user of Adobe tools. Did you try the opposite approach? What happens if you are lazy and use in-camera JPEGs? If you simply set the 5DSR to generate 20-25 MP JPEGs, how do they look compared to similar sized JPEGs straight from the Nikon or the old 5D III?

  6. Phil, we didn’t try it with the in-camera JPGs, nor capturing in smaller resolutions. I think the differences in dynamic range will disappear with the in-camera JPGs, being that most JPGs only show 8-9 stops of dynamic range. We’ll give it a shot, though.

    Re: the smaller size files, that’s a really interesting question, because I suspect it’s only capturing that much detail from the lens, and these 50 MB files are KILLING Lightroom.

  7. On the other hand, Nikon aficionado Thom Hogan points out the obvious:

    With the DxO Mark results out for the 5Ds camera and the many Internet posts that has triggered, one has to have a little empathy for the Canon DSLR users today. We Nikon users went through something similar when the D2h and D2x came out and just didn’t get standardized test scores as good as the Canon equivalents at the time. Indeed, the general sentiment is the same as it always is: if X doesn’t fix Y, I’ll have to consider switching. Today’s variation: “If Canon doesn’t improve the dynamic range of their sensors…”

    Funny thing is, I looked at a lot of superb photographs in the past year. An amazing number of them weren’t even made with current cameras (e.g. older Nikon’s like the D700 or D3, and older Canon’s such as the 1Ds and older 5D variations). Oh, they’d score poorly on DxO Mark today, wouldn’t they?

    While I want “better” all the time, I don’t need it. Most of whether an image I make turns out to be excellent has to do with my photography skills, not the camera in my hand. A secondary aspect has to do with my post processing skills, and again not the camera that took the shot.

    I had to go back and look at when I first wrote the following: 2013. “If you can’t get an excellent image the maximum print size of any desktop inkjet [13×19” at the time], then it isn’t the camera that’s the problem.” I stood by that statement through Nikon’s most sensor-sad times, and I’ll stand by it today with the Canon sensors today, too, though I don’t consider them “sad.”

    When I look at gallery prints or photo books or even Flickr my first question is never, “What camera did the photographer use?” That is maybe my 10th question. At some point quite a while ago synthetic DxO-style benchmarks stopped being interesting or relevant to the vast majority of people doing the vast majority of shooting.

    Personally, I’m most interested in where mirrorless cameras are going, since the size / weight reduction they offer means I’m more likely to have the camera there for the shot.

  8. It is sort of disappointing news. DxO can bring a bracing harshness to a new product launch. On one hand, Canon is moving in the right direction in some respects, although it would have been better if they had done more with the autofocus issues (anyone with a firmware patch?) as well. Even if the megapixel war beyond a certain point is relatively futile and not so beneficial to most users, not reaching for higher pixel count than the Sony and Nikon top models would signal an unwillingness by Canon to acknowledge they are falling behind their competition.

    jseliger raises an important point about mirrorless designs and portability. I have also Olympus which I have liked as a brand since the OM film days and still like for their commitment to carryable sizes and weights.

    The Sony line offers enticing options in their Zeiss premium lenses, but even there the bulk of the lens is considerable, at least for the fast 35/1.4. The ability to use Canon lenses on the alpha bodies with an adapter is tempting.

Comments are closed.