Sony a99 DxOMark score: not as good as the Nikon D600


DxOMark published their test results for the Sony a99 full frame camera. The a99 scored below the Nikon D600 and D800 cameras. It seems that Nikon still has the "secret sauce" that made the same 24MP Sony sensor to perform better in the D600. For reference, here are the test scores of the Canon 6D and 5D MkIII:


This entry was posted in Sony and tagged . Bookmark the permalink. Trackbacks are closed, but you can post a comment.
  • Fuzz

    I would have thought this was obvious, dynamic range and colour depth are almost identical. The A99 suffers on high ISO because some light is being directed away from the sensor up towards the AF module. That means that the sensor has to work harder to hit the ISO figures.

    • But something like a 1/3rd of a stop of less light wouldn’t result in halving of the ISO result…

      DxO also shows different scores for the D800 and the D800E… go figure.

      • Žele

        DxO shows almost identically scores for D800 and D800E whith difference to sensor sample variation. Measurements from DxO are accurate but you must know what they measure and how they calculate final score.

        • Honestly, have you seen sample variations between two of the same type of camera ever? I’ve only seen different defective pixels and AF calibration… apart from that the sensors are identical. (Lenses of course have greater variation).

          So if DxO’s results show such variations on sensors, those scores aren’t significant to all the digits shown. I.e. they’re _not_ accurate. And they’re not scientific… because any scientist worth his/her salt knows how to state the limitations of the system and not over-claim. DxO does the opposite.

          Also some of DxO’s tests themselves are totally meaningless. Their “dynamic range” is only partly defined and it can be totally faked by any manufacturer. But to understand that you need to have dealt with some scientific data.

          • Žele

            If you look at graph measurment at DxO site you will see that for D800 and D800E theey are almost identical nothing but minor sample variation.

            Their dynamic range is good, sorry for Canons outdated ADC.

          • “nothing but minor sample variation”

            lol… it’s called an observational error. Not sample variation. Their observations are wrong.

          • Žele

            So you say that’s impossible to have a any sensor sample variation? Ofcourse it’s possible, same is for all electronics.

          • Before I explain any further, can I ask, do you have any experience in performing a scientific experiment? I ask this because my explanation will have to change depending on your answer to this question.

          • nicon

            They didnt lose on the dynamic range. They lost on the ISO! Thats light and software.

          • That’s not the point… their DR measurements are silly wrong.

  • Bare

    Nikon isn’t a SLT and that’s all of mystery.

    • 800mm f/2.8 DX VR

      No it isn’t. That half mirror only takes away a third of the light (2/3 of a stop).

      • ashwins

        ..and that’s all that is needed!

  • So? How exactly does it translate to:
    a) real world performance
    b) sales

    Well, it does’t. It’s a synthetic benchmark that only matters for photography nerds and fanboys battling for their favourite brand.

    • I resent that… I’m a proper photography nerd and I know better than to trust DxO.

    • Opie

      Well said Michal. THREE downvotes? Damn, there are some angry “photographers” around here, aren’t there?

      Nobody who actually takes pictures could possibly care less about the results of these tests. They’re the pinnacle of fanboyish pedantry, and as a photographer, I find it a little embarrassing how much attention they’re paid in the blogosphere.

      Among all of the photographer friends I have, the only ones who even know what DxO is are the ones whose work leaves a lot to be desired. Funny how that works…

      • Mike Svitek

        I happily shoot with Pentax and think its the system for me. The high ISO performance on my K-5 was rated higher than the D7000 but I say: WHO CARES!? I don’t give a damn about figures and numbers, its why I took up Photography and not math. What people have to remember is that this is an art, not a science fair.

    • rkas

      No one has ever said that DxOMark score is all that you need to look at when you buy a new camera. But its a VERY good source to look at when you want to compare sensor performance. But I guess you’re one of those that goes to dpreview and compare those 100% high-iso pictures for sensor comparisons…

  • The secret is in the software. That said when is the Hasselblad Lunar is released and tested, we will see if it lives up to its name by taking the Nex7 sensor to a new level or if its just an overpriced Nex7 with a fancy skin.

    • The software can’t magically improve things. If it does NR then detail will be lost.. with that said, I did notice that the D600’s detail is worth only around 12MP or less at around ISO25600.

  • David.

    Differences fully explained by the SLR tech. No secret sauce required.

    • “SLR tech”?

      • People forget one thing, it is all about the end result. Compare a picture or video of the D600 with the A99. At my Samsung plasma 64d8000 ( last year most realistic screen in colour in the world) you eassily can see the difference. And this is big. Sony is more analogue like. Image is more dynamic and you feel more emotion. This year I went from Nikon ( used it over6 years) to Sony and it was the best thing I ever did in this hobby. I want to see emotion, for me Sony does what I want!

        • Photo Assassin

          with all due respect, “emotion” has nothing to do with camera brand.

  • jake

    actually, I have the A99v and D800 , I have to say now I lost all my faith in DXO(crappy test), I know how they compute etc for each image quality dept that they test but I guess no one knows how they calcurate the final DXO mark score.

    and in a lab resampling all images at 8mp size , maybe the D800E is so much better than the P65 or IQ180 as they say, but in real life ,the Phase always wins.

    so, to me now DXOmark score is completely useless, almost irrelevant to my camera choice procedure. I no longer wait it before proceeding to actually order any upcoming camera, I ‘d rather test them myself in my own studio.

    • It doesn’t take a genius to know that a tiny 35mm 36MP doesn’t come close to a 645 full frame 65MP or 80MP DMF back. But sadly, it’s too advanced for a lot of camera users lol

      • jake

        exactly, that’s what I wanted to say in short! thank you sir, now I know at least one logically sane person here.

        • Jim

          Would bet a dollar neither of you own, or ever used either of the latter 2 either.

  • Kernel

    For gods sake. There is no “secret sauce” with Nikon. Tranclucent mirror makes weaker S/N-ratio with Sony. Just wait for the RX1-results and you will see.

  • Warpig

    It’s so funny to read genotypewriter panic for Canon’s low scores. Chillout man.

  • Chuck Wilson

    Who cares??!! At that level the differences are virtually undetectable outside a lab … except at high ISO. The differences will be more detectable between the Sony/Nikon and the Canon sensors, but again who cares???? To paraphrase Michael Reichmann … No one ever looked an image in a gallery or magazine and said “That camera needed a higher DXO score”.

  • Back to top