Sony A7R vs Nikon D800E ISO Performance

In this article, I will show noise performance from the new high resolution Sony A7R mirrorless camera and compare it to its direct competitor, the Nikon D800E. Both have 36 MP sensors and both lack anti-aliasing (AA) filters, which should make it a good case for analysis at pixel level with no re-sizing/down-sampling involved. The Sony A7R is a pretty hot camera right now thanks to its compact camera body, high resolution and excellent image quality. Let’s take a look at how it fares against the older Nikon D800E.

Here is a comparison at the boosted ISO level of 50:

Sony A7R ISO 50 Nikon D800E ISO 50

The good news is, the Sony A7R does not seem to have the same exposure / brightness issues that the A7 sample that I have exhibits (as shown in this article).

You will notice that the A7R image looks softer/slightly lacks fine details when compared to the D800E image. That’s because I used the Sony 28-70mm f/3.5-5.6 kit lens from the A7, which is just not good enough at f/5.6 to take advantage of the full 36 megapixels! As I have written before, high resolution sensors need the best glass in order to shine and this particular case is a great example of that. I thought about using the 35mm f/2.8 Zeiss lens for the tests, however, I needed to stay at the focal length of 50mm (or my other tests would have been invalidated). For the upcoming Sony A7R review, I am planning to either use better glass for high ISO comparisons (if it becomes available), or use an adapter to fit Nikon lenses on the A7R.

Sony A7R ISO 100 Nikon D800E ISO 100

Sony A7R ISO 200 Nikon D800E ISO 200

Sony A7R ISO 400 Nikon D800E ISO 400

The Sony A7R seems to be a great overall performer – I do not see any differences until ISO 1600. At ISO 1600 and above, there is plenty of loss of colors and details in the shadows, as can be seen from the lower part of the ship.

Sony A7R ISO 800 Nikon D800E ISO 800

Sony A7R ISO 1600 Nikon D800E ISO 1600

The same with ISO 3200:

Sony A7R ISO 3200 Nikon D800E ISO 3200

At ISO 6400, there is a great loss of details on the Sony A7R. Now the red channel is getting mixed up with others and fake artifacts appear all over the frame.

Sony A7R ISO 6400 Nikon D800E ISO 6400

Needless to say, anything above ISO 6400 looks like garbage. And yet Nikon is still able to make cleaner images than Sony:

Sony A7R ISO 12800 Nikon D800E ISO 12800

Sony A7R ISO 25600 Nikon D800E ISO 25600

Next up: Fuji X-E2 review.

Comments

  1. 1
    ) Jon McGuffin
    December 12, 2013 at 6:34 am

    So it sounds pretty safe to say that the Sony A7/A7R despite having a FF sensors appears to have noise performance and characteristics more in line with a good crop camera such as the Nikon D7100/Canon 70d. Probably not a deal breaker considering it’s size, weight, and price advantage and yet you still get the shallow depth of field advantage/flexibility.

    • 3
      ) Stefan
      December 12, 2013 at 6:47 am

      Hi Jon,

      Well said. I have the same feeling. Although in low iso it is pretty impressive (but it was so with NEX-7).
      Here is a sample shot I took few days ago. Left is A7r with 35mm Zeiss. Right is Nikon D800 with 18-35mm at 35mm. Both at f/8.
      http://www.stefigphoto.com/images/A7r-D800.jpg
      I don’t remember the speed, but it is the same. I can clearly see the brighter exposure (Nasim talked about this for A7).

      • 21
        ) Jon McGuffin
        December 12, 2013 at 5:57 pm

        Wow, thanks for posting that. Both images are nice but the particular exposure you got on the A7R is obviously superior. Looks like you left it open for quite awhile (flags + water) so maybe there was some noise reduction type of things at play in the Nikon body that weren’t being done on the A7R or vice versa.

        Either way, I think things like this do nothing more than simply prove it’s not really the camera or gear. If you have “good stuff” the making good pictures part is all on the photographer; either camera is clearly capable of producing excellent results in the hands of the right person.

        Nice and thanks for your contribution!

        • 24
          ) Stefan
          December 12, 2013 at 6:08 pm

          Yes,
          I agree with you, Jon. The cameras are just tools.
          And my images are not even close to be used as a valid comparison.
          I think both cameras are very capable and if you know how to extract the best out of them, and also the lenses you use, the results will come.
          Let’s thank Nasim for his time and his work on these comparisons.

  2. 2
    ) Stefan
    December 12, 2013 at 6:38 am

    Hey Nasim,

    Great two comparisons.
    May be you should repeat the test with the new Zeiss 55mm 1.8 and Nikon 58mm 1.4 lenses when you have them both.
    Is it possible to post one full size image from both cameras – let’s say at ISO 3200 or 6400 – but not a test shot with books, etc. Just a real life shot.
    I still wonder if I want to keep my just arrived A7r and my D800, or sell the D800 and return the A7r getting a D800E instead.
    Thanks for your hard work on this website!

  3. 4
    ) Tim
    December 12, 2013 at 6:54 am

    I’ve got to say, I don’t really agree with your analysis here.
    You say there’s colour loss on the sony, but I think this has more to do with how it’s rendering colour in the first place, compare the iso 50 shots and you can clearly see that the reds on the sony are a little lighter/more orange and if you do a direct comparison between the iso 50 and 25k shots you will see both remain true to their original palette.
    Also the sony seems to be exposing by at least 1/3 of a stop more, this is most evident on the jolly roger boat stand and at the top of the rigging at the very top of the frame.
    The other thing worth noting is, yeah there is detail loss on the 25k shots if you compare them side by side the sony is the clear loser, but again this same thing is apparent in the iso 50 shots, the numbers at the bottom of the “bugs life” dvd case was the easiest place for me to make this comparison, all the high iso is doing is exacerbating a poor lens.
    This is a fairly worthless test until you use equivalent lenses.

    • 17
      ) SoCal Dave
      December 12, 2013 at 3:57 pm

      Doesn’t the fact that Sony doesn’t have such a lens mitigate in favor of the Nikon as a system?

      • 18
        ) Tim
        December 12, 2013 at 4:24 pm

        There’s no less than 5 full frame A mount 50mm lenses currently out and 2 official e mount to a mount adapters. So I’m not too sure what you are talking about there?

        • 20
          ) SoCal Dave
          December 12, 2013 at 5:28 pm

          That’s exactly my point! Presently there is exactly ***ONE*** native FE full frame Sony lens – and this is the one that was used in the test. ONE available with a total of four others announced and not all yet priced or even confirmed (See below).

          Compare this to Nikon’s native Full Frame lens selection of 53 Full Frame AF lenses! The Sony selection of five lenses – once available – only covers focal lengths from 24-200. Nikon has you covered from 14-800 and with a number of specialty lenses to boot – no adapter kludges – no third party manual focus – just native Full Frame lenses ready to put on your body today. That is a SYSTEM. Sony with it’s one lens and four to come doesn’t even come close.

          For me that is an ENORMOUS advantage for Nikon. Splitting hairs on whether the Sony image quality will look better with a future lens is to me irrelevant when there are so many lenses I want to use as a tool on my camera body that Sony just doesn’t have on offer. Even if the Sony is able to match the Nikon with a different lens, what do I do when Sony doesn’t have the lens I need?

          Maybe ten years from now, but presently it is not a mature system. It’s not even close.

          From the press release on DPreview:
          The 28-70mm F3.5-5.6 OSS is the basic ‘kit’ zoom for the A7, and won’t be sold separately. The 35mm F2.8 will cost $799, and be available in December 2013, while the 55mm F1.8 will cost $999 and be in stores January 2014. The premium 24-70mm F4 OSS Carl Zeiss standard zoom will cost $1199 and go in sale in February 2014; pricing and availability of the 70-200mm F4 OSS telezoom are still to be confirmed.

    • 22
      ) Jon McGuffin
      December 12, 2013 at 6:00 pm

      Worthless is harsh. In reality, tests like this are hard to do simply because you have RAW conversion process at play as well. With completely different manufactures and sensors, that’s a big deal. I think the test is what it is, Nikon is just doing a little better job at handling noise. It’s not a really big deal, looks like a stop or so once you get over ISO1600. With good noise reduction built into the RAW converters it’s probably never really going to make or break an image. If you knew you were going to be frequently above 6400, that would be one thing, but that’s just not the case for the vast majority of photographers out there (shouldn’t be at least; buy faster glass!).

      It’s just an opinion but to say the entire test is worthless is harsh.

  4. December 12, 2013 at 6:58 am

    If not wrong Sony only manufacture all Nikon sensor !!!!

    • 9
      ) Roberto
      December 12, 2013 at 7:39 am

      You wrong. Most sensors but not all. You can find on google wich one was made by sony, and wich one was made by nikon

      • 15
        ) Neil
        December 12, 2013 at 12:06 pm

        Aptina makes some of the Nikon sensors. Also the sensor is only part of the story of the imaging pipeline. All the ASICs and processors have a significant amount of influence in how the sensor data is processed which can lead to performance gain/loss.

  5. 6
    ) David Panno
    December 12, 2013 at 7:06 am

    My only take away from this article is that one should not use a second tier lens if you want to compete with the image quality of a D800 with a first tier lens.

    That said I would not be surprised if, even with equivalent lenses, the D800 came out on top. Too bad it is so big and heavy. With the introduction of the A7R it will remain the choice only for the most commited to IQ (or the Nikon system).

  6. 7
    ) oceanograful
    December 12, 2013 at 7:25 am

    off-topic: should we wait for X-E2 review before we buy an X-E1 ?

  7. 8
    ) Abhijit
    December 12, 2013 at 7:34 am

    Why not have used a Nikon 24-85 VR lens to make it a more fair comparison?

  8. December 12, 2013 at 8:29 am

    Hmm, interesting. I didn’t expect such a great difference. Perhaps the Sony due to the more compact design is more cramped, resulting in heat dissipation issues? Not to mention the EVF that also generates some heat, while the OVF doesn’t…

  9. 11
    ) Sandipan
    December 12, 2013 at 9:27 am

    It seems with respect to High-ISO and DR, Sony mirrorless has still some way to go before they reach Nikon DSLR performance. By that time Nikon will move too. But the Sony A7/A7R would be nice camera for serious travel photographers who crave for small size and small weight.
    Though ultimately mirrorless is the future.

  10. 12
    ) Sandy
    December 12, 2013 at 9:51 am

    Lens will have no impact on noise in an image, period. Probably more likely Bionz vs Exspeed. Or a little less likely maybe the sensors are not exactly the same, Nikon specced sensor maybe slightly diff than Sony. But I doubt this.

  11. 13
    ) Sly
    December 12, 2013 at 10:29 am

    “fake artefacts appear all over the frame” made me laugh I am wonder what a fake artefact can be, maybe a real non-feature :)

    • 16
      ) Sandipan
      December 12, 2013 at 12:53 pm

      How much does lens quality affect low light noise in a photo?

      • 19
        ) Sly
        December 12, 2013 at 5:21 pm

        “How much does lens quality affect low light noise in a photo?”
        That call an artefact not a “fake artefact” ! If a artefact if fake is mean that there is no artefact :)
        Anyway I found that funny.

  12. 14
    ) Sal
    December 12, 2013 at 11:49 am

    Thank you, bit would be interested if it were an apple to apple comparison – not using the same glass is invalidating, despite your diligent process.

    • 23
      ) Jon McGuffin
      December 12, 2013 at 6:03 pm

      We’re comparing a Nikon vs a Sony. How would it *EVER* be possible to have a true apples to apples? The comparison is legit enough to draw the conclusion that the Sony suffers a little in comparison to what has been somewhat set as the benchmark in ISO performance/Dynamic Range of the D800. Big deal..

      • 27
        ) Peter Harris
        December 15, 2013 at 11:29 pm

        Sony makes both sensors – they make all the Nikon sensors. I’m not sure your opinion stands up.

        • 28
          ) Jon McGuffin
          December 16, 2013 at 2:36 am

          Actually, you’re wrong in assuming Sony manufactures all Nikon sensors; the fact is they in fact do not. And sensor chip manufacturing is only one aspect as there are many other parts that go into the performance of that sensor other than those who make them.

      • December 23, 2013 at 8:30 am

        A true comparison of ISO would eliminate all variable except for the sensor. Sal is correct. To truly compare these two camera sensor you would need to put the exact same glass on both. Until Sony releases comparible glass this test only proves that a D800 with this particular lens does better than the Sony a7 with another particular lens.

  13. 25
    ) Don Tang
    December 12, 2013 at 9:04 pm

    The beauty of the E mount mirrorless system is that it can virtually fit all brands of lenses given the right adapter ring. Having said that, there is no reason why the comparisons cannot be done using the same glass when one of the camera is an E-mount.

    Since F stop is merely the ratio of the focal length relative to the pupil diameter, I would imagine that the variance in the number of elements and transmission index of the glass elements would all lend to difference in brightness, meaning that at the same F stop one lens is likely brighter or dimmer than another

  14. 26
    ) Peter Harris
    December 15, 2013 at 11:27 pm

    Since sony makes both sensors I have to think it’s a lens issue. It would also be good if you shot raw.

  15. 29
    ) bb
    December 18, 2013 at 5:51 pm

    “…or my other tests would have been invalidated”

    Really.

    I didn’t see what Nikon lens you used, but even were it the cheapest Nikkor, it is still going to be one of the sharpest 50mm lenses around. Not sure why you didn’t use the released 55mm Zeiss f/1.8 for a comparison that would be a bit more valid than the terrible a $300 kit zoom.

Comment Policy: Although our team at Photography Life encourages all readers to actively participate in discussions, we reserve the right to delete / modify any content that does not comply with our Code of Conduct, or do not meet the high editorial standards of the published material.

Leave a Comment