Canon 5D3 Tops DxOMark Sensor Performance for Canon, Still Lags Behind Nikon D800

Canon 5D Mark III

The Canon 5D Mark III turns out to be the best performing Canon DSLR to ever grace the DxO sensor testing lab; however, DxO still has the Nikon D800 markedly ahead of the 5D3.

DxO Canon Graph

With a DxOMark Score of 81 points, the Canon EOS 5D Mark III edges out the Canon 1Ds Mark III (80 points) by one point and gains two points on its predecessor, the Canon EOS 5D Mark II (79 points), putting it in 10th place overall (for all cameras) in the DxOMark rankings.

The Nikon D800 trounces the 5D Mark III in DxOMark’s tests with a Score of 95, making it the current DxOMark king.  You can dig through all the comparisons over at the DxOMark website.



  1. Alex B says

    Its shocking that TechRadar, using the DxO software and testing methodology, got a vastly different score for the MIII.

  2. says

    Dear Murat: At least DxO scores are a little peculiar.

    Look the difference between two pairs: 5D2 vs 5D3 and 5D3 vs D800 and observe the aspects: Color Depth; Dynamic Range; Low-Light ISO; etc.

    Although grosso modo (OK, “rough way”) in every pair the differences are relatively proportional, in particular about ISO, it seems there is not proportion in the final scores.

    The 5D3 just shows -at least DxO dixit- 2 points over the 5D2, but the new Canon presents an ISO almost 500 units “better” . Nevertheless, D800 shows 14 points although the supposed advantage in the ISO theme is almost 500 units too.

    It seems that DxO judges so super important the Dynamic Range that it aspect explains a great portion of that 14 points over the Canon; in this aspect, there is an supposed advantage of Nikon over Canon. I this the source of those 14 points?????????

    And of course in the last weeks many of us have been studying the files (Lab and RL) of both cameras and is very clear the advantage about noise for Canon especially in low light. This fact doesn´t allow me accept easily the version given by the mentioned laboratory.

    • Thomas says

      Dynamic range is important, and possibly the most important stat in reality. If you can’t have a huge variation in your image, then your colors wash out and the image gets flat quickly. This is the reason you shouldn’t be shooting at high ISOs if you don’t have to.

      Believe it or not, ISO and low light performance is not that important. Why? Because anything higher than ISO 6400 is grainy to the point of not usable, even on the “low light” specialists.

      Quit being so snobbish on your pet stat. It’s stupid epeen nonsense.

      • Peter Drescher says

        Dynamic Range is important. But so is context. At ISO 3200 on DXO, both the Canon 5Dmk3 and Nikon D800 have same dynamic range. After ISo 3200, the Canon improves dynamic range over the D800.

        To be honest, dynamic range, colour sensitivity and tonal range are all important factors for a shot as none of these properties can be redeemed later on, and all are desirable components to maximise before editing.

        The marking skew in DXO is only for base ISO on dynamic range, which counts for little in low-light situations. I think DXO needs to contextualise the marking system and make a DXO mark for base-ISO/outdoor/studio use and a Low-Light mark, so the message is better delivered to the two significantly different shooting conditions that predominate. Daylight/studio that enables ISO100 and low-light/mixed light that requires sensitivity and all its dynamics.

        That said, what the actual final print looks like is more important and cannot always be captured by technical reviews like DXO.

  3. says

    Now, when Canon shows cleaner ISOS, low light is not so important… And all the bla bla bla about how much important WAS low light performance when the D700 prevailed in that field???????????????? That point of view is a stupid and snobbish posture indeed.