Re: MTF Grade

Discussion in 'Digital Photography' started by Bowser, Aug 7, 2003.

  1. Bowser

    Bowser Guest

    MTF ratings cannot be used reliably to describe a lens' performance. Taken
    by themselves, they will not tell you what your final image will look like.
    MTF testing is done (on some popular sites) using a single sample of a lens
    at infinity. The only reliable test you can use to determine the
    capabilities of a lens is to shoot pictures, and examine those pictures.

    The old rule of thumb, that primes were far superior to zooms simply doesn't
    hold anymore, since many zooms, like the Nikon 17-35, are at least the equal
    of the corresponding primes, and in some cases better.

    "Curt" <> wrote in message
    news:...
    > Hi,
    >
    >
    > In reading alot of posts (not much to do at work this week)
    > I see people recommending lenses as being very sharp etc.
    > However when reading the mtf test result it is in some cases
    > really bad.
    > Is the mtf test something I can lean on when purchasing a lens?
    > Right now Im thinking about buying prime lenses and to use the leg-zoom.
    >
    > Regards // Curt
     
    Bowser, Aug 7, 2003
    #1
    1. Advertising

  2. Bowser

    Bowser Guest

    If you prefer a simple MTF test, have at it. But, take for example, the
    Nikon 24-120 zoom. It received a low MTF score, but gets rave reviews from
    people who have actually shot pictures using this lens. It is more than
    capable of producing professional quality images, despite the ridiculously
    low MTF score. Take a look at some of the review sites, like David
    Reuther's, and you get a different picture of the capabilities of that lens.

    FWIW, here's how I test a lens:

    1. Set up a spreadsheet that documents the combination of aperture and focal
    length for that particular shot. For a zoom, it may take all 35 shots to
    perform this test at a particular distance.

    2. Perform this test on a flat, detailed subject for three different
    distances. I usually shoot a test at infinity, about 50', and as close as
    the lens will focus.

    3. Examine the slides, usually Velvia 50, under a powerful loup.

    And, of course, the camera is mounted on a heavy tripod. I know, this is
    more than most people want to do, but those with high quality digicams won't
    have to pay for film, and their record keeping is easier, since EXIF info
    will tell them everything they need to know about focal length, aperture,
    etc.

    Lastly, I'll shoot some pix under harsh conditions, like into a bright light
    source to test for flare, a few pix of something with straight lines to test
    for distortion, and some other stuff to see what type of image this lens
    will produce under real-life conditions. MTF will tell you nothing about
    flare, distortion, color rendition, or anything else that can affect, and
    ruin your pictures. It's a one-dimensional test that tells me nearly
    nothing.

    But, if all you want to do is shoot pictures of black bars on white
    backgrounds at infinity, MTF is absolutely the way to go.

    "Bart van der Wolf" <> wrote in message
    news:3f32d731$0$49116$4all.nl...
    >
    > "Bowser" <> wrote in message
    > news:OZyYa.59839$Oz4.15593@rwcrnsc54...
    > > MTF ratings cannot be used reliably to describe a lens' performance.

    >
    > That of course depends on what performance you have in mind.
    >
    > > Taken
    > > by themselves, they will not tell you what your final image will look

    > like.
    >
    > Not any single test tells it all. But a single MTF will already tell a

    lot,
    > especially about the interaction of lens MTF and sensor MTF. It for

    instance
    > predicts the design flaw of the Sigma SD9, which was confirmed by test

    shots
    > at a much closer distance than infinity.
    >
    > > MTF testing is done (on some popular sites) using a single sample of a

    > lens
    > > at infinity.

    >
    > Yes, and if that lens was designed to be able and shoot at infinity focus,
    > it is a valid test. A good test, like the MTF, will reveal a lot. If the
    > lens already falls short of design goal at one important focus distance,

    it
    > might also fall short at other distances. What's more, if a lens varies so
    > much in its performance over the focus range, how do you instruct your
    > subjects to avoid certain distances?
    >
    > > The only reliable test you can use to determine the
    > > capabilities of a lens is to shoot pictures, and examine those pictures.

    >
    > You mean with a single sample lens? Shot at 2.63 meter focus? Pictures of
    > what? How are you going to relate that to another lens performance. On

    what
    > scale?
    >
    > > The old rule of thumb, that primes were far superior to zooms simply

    > doesn't
    > > hold anymore,

    >
    > The MTF's of some zooms indicate that they can outperform some fixed focus
    > lenses, at certain settings. The MTF will tell you exactly what to expect
    > regarding contrast resolution of subjects of different size.
    >
    > > since many zooms, like the Nikon 17-35, are at least the equal
    > > of the corresponding primes, and in some cases better.

    >
    > Was that the sample you bought, or are there other samples that performed
    > similarly? How was that conclusion reached? Based on what? Eyeballing the
    > image? How was the image prepared? What were the viewing conditions?
    >
    > Nah, I prefer a simple MTF test. It's reliable if you know what you're
    > looking at, the result is quantifyable and repeatable, and it will reveal
    > things like quality spread.
    > There must be a reason that MTF is used by lens manufacturers during
    > production, and that several ISO standards use MTF (or SFR) measurements.
    >
    > Bart
    >
    >
     
    Bowser, Aug 9, 2003
    #2
    1. Advertising

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

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. MikeWhy

    Canon Digital Rebel kit lens MTF

    MikeWhy, Dec 15, 2003, in forum: Digital Photography
    Replies:
    0
    Views:
    387
    MikeWhy
    Dec 15, 2003
  2. Dane Brickman

    MTF/Nikon?Japan???

    Dane Brickman, Jul 8, 2004, in forum: Digital Photography
    Replies:
    6
    Views:
    606
    Bart van der Wolf
    Jul 9, 2004
  3. Ilya Zakharevich

    Sharpening artefacts and MTF of monitors?

    Ilya Zakharevich, Apr 20, 2005, in forum: Digital Photography
    Replies:
    31
    Views:
    898
    David J Taylor
    Jul 7, 2005
  4. deryck  lant

    All the Nikkor MTF Charts from 14mm to 600mm

    deryck lant, Jun 17, 2005, in forum: Digital Photography
    Replies:
    2
    Views:
    501
    Michael Benveniste
    Jun 17, 2005
  5. Christopher Yeung

    MTF charts for Nikkor lenses

    Christopher Yeung, Aug 16, 2005, in forum: Digital Photography
    Replies:
    14
    Views:
    2,088
    deryck lant
    Aug 18, 2005
Loading...

Share This Page