Featured DXO Mark just rated the E-M1 II - Impressive results!

Discussion in 'Olympus Cameras' started by Jonathan F/2, Jan 20, 2017.

  1. Jonathan F/2

    Jonathan F/2 Mu-43 Hall of Famer

    Aug 10, 2011
    Los Angeles, USA
    • Like Like x 4
    • Agree Agree x 2
    • Informative Informative x 1
  2. Repp

    Repp Mu-43 Top Veteran Subscribing Member

    947
    Jan 27, 2011
    Oak Harbor, WA
    The ISO performance in particular saw the largest increase. I'm really curious to see how the GH5's sensor will score now. Exciting times.
     
    • Agree Agree x 1
  3. Jonathan F/2

    Jonathan F/2 Mu-43 Hall of Famer

    Aug 10, 2011
    Los Angeles, USA
    This is why I can't get into the APS-C mirrorless cameras. You don't gain much on image quality over M43 and it's not much smaller compared to FF.

    Though I wonder if Olympus is using the same 20mp sensor from the Pen-F/GX8 or if it's an improved sensor or the same sensor with improved image processing?
     
    Last edited: Jan 20, 2017
  4. Turbofrog

    Turbofrog Mu-43 Hall of Famer

    Mar 21, 2014
    I really wish I could believe it, but to me this is just another example of highly dubious data from DXO. This is pretty much a direct contradiction from the visual data.

    Studio shot comparison: Digital Photography Review

    hb6SKKh.

    DXO claims there's a 0.7 stop difference between the E-M1 II and GX8, and just slightly less than that for the PEN F. The DPReview doesn't let you adjust 1/3 stops, but a 0.7 stop difference is closer to 1 stop than it's to nothing, and at ISO 1600 the GX8 and PEN F are very much cleaner than the E-M1 II is at 3200.

    Lies, damn lies, and statistics...
     
    • Agree Agree x 4
    • Like Like x 1
  5. Repp

    Repp Mu-43 Top Veteran Subscribing Member

    947
    Jan 27, 2011
    Oak Harbor, WA
    I know they removed the AA filter the GX8 still has, not sure on the Pen-F. Probably reworked the processing algorithms for it as well. There is apparently a drastic change in the readout mechanics/software for the GH5's sensor... but only time will tell since it hasn't even launched yet.
     
  6. Brian Beezley

    Brian Beezley Mu-43 All-Pro

    DXO uses measured ISO while DPReview and everyone else use indicated ISO. If you look at the DXO ISO sensitivity curve, there's quite a difference between the two. ISO sensitivity also differs between the E-M1 I and II. Perhaps this accounts for what you're seeing.

    Brian
     
    • Like Like x 2
  7. Jonathan F/2

    Jonathan F/2 Mu-43 Hall of Famer

    Aug 10, 2011
    Los Angeles, USA
    I went through the shot comparison and while the Pen-F and E-M1 II files appear similar, it looks like the E-M1 II is retaining more color information in the raws. Detail is similar, but in real world use that extra bit of DR at high ISO, might be what's giving the E-M1 II such a good high ISO rating from DXO Mark. Either that or Olympus slipped them a briefcase full of cash! ;)
     
    Last edited: Jan 20, 2017
    • Agree Agree x 1
  8. alex g

    alex g Mu-43 All-Pro Subscribing Member

    Mar 30, 2016
    New York / Bath
    It's an entirely different sensor. The Pen-F sensor doesn't have any PDAF points.
     
    • Agree Agree x 1
    • Useful Useful x 1
  9. coffeecat

    coffeecat Mu-43 Top Veteran Subscribing Member

    995
    Aug 4, 2012
    SW England
    Rob
    Being an engineer, I went to the DXO website, and added my E-PM2 to their "compare" tool and looked at the graphs.
    And I thought - "yeah, for what I paid, I'm a 'bang-for-buck' winner!"
    That should keep new-camera GAS away for a while...


    ;-)

    Seriously though, the EM1-2 results are impressive, if you're in the market for a $2k body.
     
    • Agree Agree x 2
    • Funny Funny x 1
    • Winner Winner x 1
  10. Machi

    Machi Mu-43 Veteran

    317
    May 23, 2015
    Clearly new sensor combines the best of both worlds:
    Great full well of older m43 Sony's sensors from E-M5, E-M10,..., PEN-F (good SNR and better preservation of highlights)
    with low readout noise of older Panasonic's sensors from GX7 or E-M1 (good DR and better preservation of shadows).
    Nevertheless 1EV better DR was marketing BS.
     
  11. AussiePhil

    AussiePhil Mu-43 All-Pro

    Jun 1, 2014
    Canberra, ACT, Aust
    Phil
    The fact that two "highly respected" sites are now seemingly contradicting themselves brings into question both sites. Whilst I personally disregard DXO nearly entirely as it's a pure measurebator wet dream, in this case I'll take DXO over DPR as I truly believe the DPR studio scene is comprised for ALL DSLR/ILC cameras due to the lens camera combination and all the quality variations the combinations create
     
  12. Amin Sabet

    Amin Sabet Administrator

    Apr 10, 2009
    Boston, MA (USA)
    Very impressive. I wish my Pen F had that sensor!

    PS - I trust DxOmark a lot more than I trust DPReview. Many times I've noticed odd things about DPR methods (example here), whereas only rarely have I seen mistakes in DxOmark data.
     
    Last edited: Jan 21, 2017
    • Agree Agree x 3
    • Funny Funny x 1
  13. Drdave944

    Drdave944 Mu-43 All-Pro Subscribing Member

    Feb 2, 2012
    Just got mine out of the box. IS is superb.Focuses on what you think it should,not background,so far. untitled (15 of 16). untitled (15 of 16)-2.
     
    • Like Like x 4
  14. Machi

    Machi Mu-43 Veteran

    317
    May 23, 2015
    Sometimes DXO measurements are at least partially flawed.
    Curve of sensitivity vs DR for most cameras has mostly predictable shape which isn't often true for DXO data.
    This is probably the case of GX8 where after ISO1600 there is a drop in DR which doesn't look right (the same is true for drop at ISO400 for PEN-F).
     
  15. b_rubenstein

    b_rubenstein Mu-43 All-Pro

    Mar 20, 2012
    Melbourne, FL
    I have the original E-M1 and a GX8, so I examined the curves to see how they compare to the Mark II. Depending on the parameter and ISO the difference ranged from nothing to less than a stop. Want a clearly visible difference between sensors, go to a FF sensor.

    Thinking that there should be a significant visible difference in the noise of two different pictures from different cameras assumes that a high correlation exists between a specific type of measurement and what is visible, and that is not a valid assumption.
     
    • Agree Agree x 1
    • Winner Winner x 1
  16. alex g

    alex g Mu-43 All-Pro Subscribing Member

    Mar 30, 2016
    New York / Bath
    Agreed. As I think I've mentioned before in another thread, I've been pleasantly surprised by the subjective improvement in noise performance between the E-M1.2 and the E-M1. While the various test results point to there being only a marginal improvement, in practice, there's a considerable one, in my opinion. The original 16Mp sensor has an aggressive luminance noise pattern that is present even at base ISO — the luminance noise from the new sensor has a more diffuse pattern and is a lot less intrusive. Added to which, it seems possible to push the ISO further before the colours start falling apart. In short, I'd say that the improvement is greater than the figures suggest.
     
    • Like Like x 1
    • Agree Agree x 1
  17. It's kind of funny though. Bad score - retest, retest, retest! Must be something wrong! I never see that. Good score - well done DXO, I always liked you.
     
    • Like Like x 1
    • Funny Funny x 1
  18. pdk42

    pdk42 One of the "Eh?" team

    Jan 11, 2013
    Leamington Spa, UK
    At least human nature is predictable...
     
    • Like Like x 1
  19. pdk42

    pdk42 One of the "Eh?" team

    Jan 11, 2013
    Leamington Spa, UK
    Like most of us, I'm not completely convinced by DxO's measures, but I have to concede that at least they try. If that figure of 80 is correct, then there should be quite a jump in performance from the mark 1 or the Pen F. I'd really like to get my hands on one for a few hours of comparitive shooting. I wonder when the Try BC and Buy scheme will include the mark II?
     
  20. dpswbab

    dpswbab Mu-43 Regular

    157
    May 30, 2014
    The biggest difference I see with my mk ii over previous OMD bodies is the ability to recover high brightness areas. I took some sunset shots of a bridge with the sun behind it, and was happily surprised that I could pull the highlights from the sun/clouds near it to values that were not clipped. I'm pretty sure my E-M5 ii would not have behaved nearly as well.