DxO ranks Nikon # 1 & 2...but where does resolution fit in?

Discussion in 'Nikon' started by RichA, Nov 18, 2008.

  1. RichA

    RichA Guest

    1. Advertisements

  2. RichA

    Me Guest

    DxO are nuts.
    Hands-on experience with /subjective/ assessment of performance leads to
    far different conclusions than their so-called "objective" performance.
    Case in point - Pentax K20d / Samsung GX20.
    The sensor in that camera truly sucks - yet they give it a high score.
     
    Me, Nov 18, 2008
    #2
    1. Advertisements

  3. RichA

    RichA Guest

    And yet most reviewers are fixated on noise to the extent they forget
    about resolution of detail. Check out this dpreview comparison
    between the K200D (Pentax) and a few others. Look at the detail (or
    lack of) with the Nikon D60. You could clean up all the noise in the
    Pentax image, it would still show more detail. When you take raws
    from the Nikon D300 and the Pentax K20D, they are virtually the same,
    with the Pentax having slightly higher resolution and the Nikon having
    greater per pixel sharpness. I've compared the two cameras at length.
    http://www.pbase.com/andersonrm/image/105998943
     
    RichA, Nov 18, 2008
    #3
  4. RichA

    Me Guest

    All I can say is that the DxO analysis is seriously flawed.
    Take a Canon 5d and K20d and compare raw files side by side.
    The K20d has far worse noise performance, far far worse usable dynamic
    range, and much less "per pixel" sharpness than the 5d. You'd have to
    be freakin' blind not to see it. Yet DxO rate them about the same...
    According to them the Samsung sensor is /miles/ ahead of the Canon 40d
    sensor. That's so laughable I could cry.
    The problem with /wankers/ like DxO - is that /idiots/ will read that
    garbage and be inclined to believe it. Already ludicrous landscape have
    links to it - probably others.
    Their web page looks pretty good though - except it crashes Mozilla
    Firefox if the interactive pages are loaded on more than one tab.
     
    Me, Nov 18, 2008
    #4
  5. RichA

    RichA Guest

    A couple things; the only real issue with the K20D sensor is banding
    in the shadows at higher ISOs, A very annoying problem. Forget about
    noise control above 800, you'll never get it with that pixel
    density. The Canon 50D has been semi-panned by a number of
    reviewers, and lets face it: A 1.5 sensor with 15 megapixels is at the
    brick wall of performance and sensor technology has not progressed
    fast enough to get away from the fact bigger pixels simply = better
    images, at least in-terms of DR and tonality. There is only so much
    (even Nikon) can do with processing of the image.
     
    RichA, Nov 18, 2008
    #5
  6. RichA wrote:
    []
    An observation:

    The Nikon D60 shows a much sharper image on its own review page:

    http://www.dpreview.com/reviews/nikond60/page18.asp

    than on the page from which RichA copied his image:

    http://www.dpreview.com/reviews/pentaxk200d/page15.asp

    so there's something wrong, or different, somewhere, raising doubts about
    the comparison.

    David
     
    David J Taylor, Nov 18, 2008
    #6
  7. RichA

    Me Guest

    It's not just banding - there's just no recoverable detail in the
    shadows when pushing. Compared with 40d and D300, the K20d is wiped out
    in that area. There's too many wankers making obscurely quantified DR
    measurements - but who apparently never take photos where DR is needed.

    "I have shown that one can boost Nikon D300 low ISO images by 4 stops
    where as one can only boost K20D images by 2 to 2.5 stops; at high ISO's
    the Canon 40D and the Nikon D300 have about the same noise at ISO 1600
    as the K20D does at ISO 400 or the K20D has about 1.0 to 1.5 stops less
    dynamic range for images at ISO 1600, or only about 6 to 6.5 stops of
    dynamic range at ISO 1600 for medium high image quality. This is about
    what you have seen in your own tests, although of course you can't
    really compare with other cameras."
    http://daystarvisions.com/Docs/Rvws/K20D/pg3.html#dr
    And that's from a Pentax fan...

    From DxO's rating, the K20d and Canon 5d sensors are "about the same".
    Jesus wept - the K20d actually dismally fails in the one specific area
    where the Canon 5d (still - even if not compared to the D3/700) excels.

    DxO clearly haven't had the "hands-on" with the cameras to realise when
    that massive anomaly appeared in their results, that their methodology
    might need to be reassessed. A tip might be to look at photos - rather
    than run raw files through some software analysis dreamed up by a geek.

    But then again I guess a bit of controversy drags suckers to their site,
    and a few might buy their averagely performing (IMO) highly priced
    software. After all, Ken Rockwell has made millions just from
    click-throughs.
     
    Me, Nov 18, 2008
    #7
  8. Thanks for checking that, John. Perhaps I should have done so myself.
    Everyone makes mistakes - let's hope this one is soon corrected. Will you
    be reporting it, please?

    Cheers,
    David
     
    David J Taylor, Nov 18, 2008
    #8
  9. John O'Flaherty wrote:
    []
    Error reported to DP Review.

    Cheers,
    David
     
    David J Taylor, Nov 18, 2008
    #9
  10. RichA

    hooya Guest

    Yeah and he bought a K20D... But his is but one opinion - there are others
    that do not agree. Tell me, what extensive testing have you done with these
    cameras..? Or maybe you simply fuel your prejudice by cherry picking bits
    you think suit your stupid bias.
    Take a look at the images on Anandtech's site... You might be surprised.

    Try reassessing your own methodology. Oh no, of course not, you haven't got
    one! You have a single mode bias.
    You are the biggest sucker of all. Jeez what a dick!
     
    hooya, Nov 18, 2008
    #10
  11. RichA

    Me Here Guest

    Me Here, Nov 18, 2008
    #11
  12. RichA

    RichA Guest

    Well, it's like Olympus which does have reach into the shadows. It's
    there, but who would want it? Noisy and banded.
    This is wrong.

    http://www.pbase.com/andersonrm/noise_tests_pentax_k20d__nikon_d300
     
    RichA, Nov 18, 2008
    #12
  13. RichA

    RichA Guest

    Oh, Olympus scored low, as expected. Olympus's biggest claim to fame
    is having done their lenses right. Nothing much can compare to their
    Top Pro lenses and even their kit lenses are like pro lenses from the
    competition. Of course, designing them correctly for sensors and only
    having to support a 4/3rds size helped.
     
    RichA, Nov 18, 2008
    #13
  14. RichA

    Me Guest

    hooya wrote:

    Do you think that a Pentax K20d and a Canon 5d are "about the same" in
    terms of sensor performance?
     
    Me, Nov 18, 2008
    #14
  15. RichA

    Me Guest

    Something DxO seem to have missed - compared to the (old) Canon 5d, for
    which pulling detail from the shadows is remarkably good (just not
    compared to D3/700!) Yeah - I'd say so. Why would you bother showing sample photos with no
    detail and with wildly varying exposures?
     
    Me, Nov 18, 2008
    #15
  16. RichA

    RichA Guest

    The varied exposures are because the Pentax cameras produce dimmer
    (30-40% in PS about 0.5 to 0.7 stops) images than either Nikons or
    Olympus cameras, so you have to provide varying exposures to:
    1. Show what the Pentax produces when exposed exactly the same as the
    Nikon.
    2. Show what it produces when you expose to produce the same
    illumination level as the Nikon.
     
    RichA, Nov 19, 2008
    #16
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.