Loading ...
Sorry, an error occurred while loading the content.

Trying to set up NN5 with Canon 20D & 10-22mm lens

Expand Messages
  • Robert Slade
    Hello, I m new to panoramic photography and have just joined this list today. I m using a Nodal Ninja 5 panoramic tripod head, and am trying to set it up for
    Message 1 of 6 , Oct 13, 2008
    • 0 Attachment
      Hello,

      I'm new to panoramic photography and have just joined this list today.
      I'm using a Nodal Ninja 5 panoramic tripod head, and am trying to set it
      up for use with a Canon 20D and 10-22mm lens.

      Unfortunately, this camera/lens combination is not listed on the NN5 site:
      http://www.nodalninja.com/html/canon_settings.html

      Can anyone please help me with the NPP for this lens?

      Also, the above table gives the lower rail setting (using plate) as
      52.5mm which is within 0.5mm of where I measured it using a T-square
      aligned with the red dot on the lens. But when I look through the
      viewfinder with the camera in the 90 degree nadir position, the NN
      dartboard logo is displaced significantly upwards. To make it central
      requires a setting of 47.5mm, that's 5mm away from the setting I have
      been using.

      Should the logo be exactly centred in the nadir position?

      Thanks for any help,
      Rob
      --
    • erik leeman
      Welcome to the group Rob! Please note that the centre of the viewfinder might not be exactly lined up with the centre of the sensor (with my 5D it isn t).
      Message 2 of 6 , Oct 13, 2008
      • 0 Attachment
        Welcome to the group Rob!

        Please note that the centre of the viewfinder might not be exactly
        lined up with the centre of the sensor (with my 5D it isn't).
        Therefore you should take a photo looking straight down and use your
        camera's display zoomed in to the max. to judge nadir alignment, not
        the viewfinder.

        Regards,

        Erik Leeman

        (http://www.erikleeman.com http://www.flickr.com/photos/erik-nl/)

        --- In PanoToolsNG@yahoogroups.com, Robert Slade <robert@...> wrote:

        > But when I look through the
        > viewfinder with the camera in the 90 degree nadir position, the NN
        > dartboard logo is displaced significantly upwards. To make it central
        > requires a setting of 47.5mm, that's 5mm away from the setting I have
        > been using.
        >
        > Should the logo be exactly centred in the nadir position?
        >
        > Thanks for any help,
        > Rob
      • John Houghton
        ... At 10mm focal length setting, the no parallax position is probably around 94mm. It s not difficult to measure the exact setting yourself. See
        Message 3 of 6 , Oct 13, 2008
        • 0 Attachment
          --- In PanoToolsNG@yahoogroups.com, Robert Slade <robert@...> wrote:
          >
          > Can anyone please help me with the NPP for this lens?

          At 10mm focal length setting, the no parallax position is probably
          around 94mm. It's not difficult to measure the exact setting
          yourself. See http://www.johnhpanos.com/epcalib.htm .

          John
        • Robert Slade
          Thanks Erik, It does seem that the viewfinder is not precisely aligned with the centre of the sensor. Using your method, I have found the best setting to be
          Message 4 of 6 , Oct 14, 2008
          • 0 Attachment
            Thanks Erik,

            It does seem that the viewfinder is not precisely aligned with the
            centre of the sensor. Using your method, I have found the best setting
            to be 50mm which agrees exactly with the following table (allowing 10mm
            for the camera plate):

            http://wiki.panotools.org/Entrance_Pupil_Database

            Thanks again,
            Robert
            --

            erik leeman wrote:
            >
            >
            > Welcome to the group Rob!
            >
            > Please note that the centre of the viewfinder might not be exactly
            > lined up with the centre of the sensor (with my 5D it isn't).
            > Therefore you should take a photo looking straight down and use your
            > camera's display zoomed in to the max. to judge nadir alignment, not
            > the viewfinder.
            >
            > Regards,
            >
            > Erik Leeman
          • Robert Slade
            Hello John, Thanks for your email. I have measured the NPP using a light stand with trial and error and found it to be about 105mm. I ll try your exit pupil
            Message 5 of 6 , Oct 14, 2008
            • 0 Attachment
              Hello John,

              Thanks for your email. I have measured the NPP using a light stand with
              trial and error and found it to be about 105mm. I'll try your exit pupil
              tutorial method which seems more consistent. I was hoping to confirm my
              measurements with someone else using the same camera/lens combination,
              but the NN5 table did not include 20D/10-22mm data.

              The panotools wiki gives 36.5mm + 66mm = 102.5mm, so maybe a little
              adrift there.

              I am using PTGui, and typically getting the control points optimized as
              "very good" with a maximum error of between 10 and 30 after deleting the
              worst control points. I read on this list yesterday that some people
              were seeing errors of 1-2 pixels, so should I be looking to improve my
              optimization errors?

              Regards,
              Robert Slade
              --

              John Houghton wrote:
              >
              >
              > --- In PanoToolsNG@yahoogroups.com
              > <mailto:PanoToolsNG%40yahoogroups.com>, Robert Slade <robert@...> wrote:
              > >
              > > Can anyone please help me with the NPP for this lens?
              >
              > At 10mm focal length setting, the no parallax position is probably
              > around 94mm. It's not difficult to measure the exact setting
              > yourself. See http://www.johnhpanos.com/epcalib.htm
              > <http://www.johnhpanos.com/epcalib.htm> .
              >
              > John
            • John Houghton
              ... Robert, Yes, you could do better. A maximum error of 2-3 pixels is usually obtainable. Check that you only have control points between neighboring images
              Message 6 of 6 , Oct 14, 2008
              • 0 Attachment
                --- In PanoToolsNG@yahoogroups.com, Robert Slade <robert@...> wrote:
                >
                > I am using PTGui, and typically getting the control points
                > optimized as "very good" with a maximum error of between 10 and 30
                > after deleting the worst control points. I read on this list
                > yesterday that some people were seeing errors of 1-2 pixels, so
                > should I be looking to improve my optimization errors?

                Robert, Yes, you could do better. A maximum error of 2-3 pixels is
                usually obtainable. Check that you only have control points between
                neighboring images (i.e. between 0-1 and 1-2 but not 0-2) and include
                the lens shift parameters d & e in the optimization. Also check that
                if all the images are rotated into the vertical position for input to
                PTGui, they have all been rotated by -90 degrees. (Sometimes the nadir
                and/or zenith images can be inadvertently rotated by +90 degrees,
                because of bogus data from the auto rotation sensor).

                John
              Your message has been successfully submitted and would be delivered to recipients shortly.