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

Focus and depth of field of fisheye lenses

Expand Messages
  • michel thoby
    Hi all, This is a recurrent subject of discussion here and here. I could not make my mind clear about the actual shape of the region in space -when shooting
    Message 1 of 7 , Sep 8, 2009
    • 0 Attachment
      Hi all,

      This is a recurrent subject of discussion here and here.
      I could not make my mind clear about the actual shape of the region
      in space -when shooting with the widest aperture stop set- where the
      object can be predicted to be viewed eventually sharp (i.e. in-focus)
      on the image.
      Almost all responses that I got from searching the Web is how huge is
      the DOF with fisheye lenses. This is obviously misleading, at least
      with some lenses and at close to moderate shooting range. So I have
      attempted to perform a study by myself and to report it:
      http://michel.thoby.free.fr/Fisheyes_Focus/Focus_mapping.html
      or http://tinyurl.com/nn3e7y

      Regards,

      Michel Thoby
    • Trausti Hraunfjord
      Wow... that was enlightening. Me being in the fringe society of Zuiko 8mm users, would very much like to know how that lens behaves... with your setup, I
      Message 2 of 7 , Sep 8, 2009
      • 0 Attachment
        Wow... that was enlightening. Me being in the "fringe society" of Zuiko 8mm
        users, would very much like to know how that lens behaves... with your
        setup, I can replicate the test (when I take the time), and figure out how
        it would look.

        Thank you for sharing this with us!

        Trausti

        On Tue, Sep 8, 2009 at 4:54 AM, michel thoby <thobymichel@...> wrote:

        >
        >
        > Hi all,
        >
        > This is a recurrent subject of discussion here and here.
        > I could not make my mind clear about the actual shape of the region
        > in space -when shooting with the widest aperture stop set- where the
        > object can be predicted to be viewed eventually sharp (i.e. in-focus)
        > on the image.
        > Almost all responses that I got from searching the Web is how huge is
        > the DOF with fisheye lenses. This is obviously misleading, at least
        > with some lenses and at close to moderate shooting range. So I have
        > attempted to perform a study by myself and to report it:
        > http://michel.thoby.free.fr/Fisheyes_Focus/Focus_mapping.html
        > or http://tinyurl.com/nn3e7y
        >
        > Regards,
        >
        > Michel Thoby
        >


        [Non-text portions of this message have been removed]
      • panoramicsdk
        ... The problem will not be as much visible in the same way with an 8mm. Only at full aperture but if you stop down to f8 you will have a DOF from 0.23 to
        Message 3 of 7 , Sep 8, 2009
        • 0 Attachment
          --- In PanoToolsNG@yahoogroups.com, Trausti Hraunfjord <trausti.hraunfjord@...> wrote:
          >
          > Wow... that was enlightening. Me being in the "fringe society" of Zuiko 8mm
          > users, would very much like to know how that lens behaves... with your
          > setup, I can replicate the test (when I take the time), and figure out how
          > it would look.
          >
          > Thank you for sharing this with us!

          The problem will not be as much visible in the same way with an 8mm.
          Only at full aperture but if you stop down to f8 you will have a DOF from 0.23 to infinity with focus set to 0.5m.
          That will in practice eliminate the problem.

          However with the Tokina set to 12mm on a 5D you will see it even stopped down to f11.
          The best way is to shoot zenith and shoot the 4 around at -10 degrees .

          Hans


          >
          > Trausti
          >
          > On Tue, Sep 8, 2009 at 4:54 AM, michel thoby <thobymichel@...> wrote:
          >
          > >
          > >
          > > Hi all,
          > >
          > > This is a recurrent subject of discussion here and here.
          > > I could not make my mind clear about the actual shape of the region
          > > in space -when shooting with the widest aperture stop set- where the
          > > object can be predicted to be viewed eventually sharp (i.e. in-focus)
          > > on the image.
          > > Almost all responses that I got from searching the Web is how huge is
          > > the DOF with fisheye lenses. This is obviously misleading, at least
          > > with some lenses and at close to moderate shooting range. So I have
          > > attempted to perform a study by myself and to report it:
          > > http://michel.thoby.free.fr/Fisheyes_Focus/Focus_mapping.html
          > > or http://tinyurl.com/nn3e7y
          > >
          > > Regards,
          > >
          > > Michel Thoby
          > >
          >
          >
          > [Non-text portions of this message have been removed]
          >
        • Erik Krause
          ... Many thanks for your big efforts. This was something which needed to be worked out for ages - especially in times of high resolution full frame sensors
          Message 4 of 7 , Sep 8, 2009
          • 0 Attachment
            michel thoby wrote:

            > So I have
            > attempted to perform a study by myself and to report it:
            > http://michel.thoby.free.fr/Fisheyes_Focus/Focus_mapping.html
            > or http://tinyurl.com/nn3e7y

            Many thanks for your big efforts. This was something which needed to be
            worked out for ages - especially in times of high resolution full frame
            sensors where DOF is an issue not only for fisheyes.

            You estimated the sharpness in the fisheye image directly. Since if a
            fisheye image is used in a spherical panorama some portions of the
            images are scaled heavily it would be interesting how sharpness
            estimates after remapping to spherical.

            And since spherical images are normally viewed in rectilinear portions
            it would be most interesting to estimate the sharpness this way.
            Probably DevalVR in it's stitchless mode could be used best to display
            fisheye images in their final rectilinear resolution...

            best regards
            --
            Erik Krause
            http://www.erik-krause.de
          • Bernhard Vogl
            Thanks Michel for publishing the results of your study. It is very interesting to see the actual shape of the focus area. For me it would be also very
            Message 5 of 7 , Sep 8, 2009
            • 0 Attachment
              Thanks Michel for publishing the results of your study.
              It is very interesting to see the actual shape of the focus area.
              For me it would be also very interesting to see how the in-focus area
              changes for a longer (16mm) fisheye on a FF camera. From my bitter
              experience, i can say, that even stopped-down the DOF is pretty narrow.

              Best regards
              Bernhard
              > This is a recurrent subject of discussion here and here.
              > I could not make my mind clear about the actual shape of the region
              > in space -when shooting with the widest aperture stop set- where the
              > object can be predicted to be viewed eventually sharp (i.e. in-focus)
              > on the image.
              > Almost all responses that I got from searching the Web is how huge is
              > the DOF with fisheye lenses. This is obviously misleading, at least
              > with some lenses and at close to moderate shooting range. So I have
              > attempted to perform a study by myself and to report it:
              > http://michel.thoby.free.fr/Fisheyes_Focus/Focus_mapping.html
              > or http://tinyurl.com/nn3e7y
              >
            • panoramicsdk
              ... It is easy to do a test for it. Here is one for Canon 15mm First one focused at 2m for DOF from 1m-infinity. Distance to wall 1m and to the floor 1.4m You
              Message 6 of 7 , Sep 8, 2009
              • 0 Attachment
                --- In PanoToolsNG@yahoogroups.com, Bernhard Vogl <bvogl@...> wrote:
                >
                > Thanks Michel for publishing the results of your study.
                > It is very interesting to see the actual shape of the focus area.
                > For me it would be also very interesting to see how the in-focus area
                > changes for a longer (16mm) fisheye on a FF camera. From my bitter
                > experience, i can say, that even stopped-down the DOF is pretty narrow.

                It is easy to do a test for it.
                Here is one for Canon 15mm
                First one focused at 2m for DOF from 1m-infinity.
                Distance to wall 1m and to the floor 1.4m You can see that causes unfocus at the floor.
                http://www.panoramas.dk/panorama/fisheye-focus/f4-2m.jpg

                Same focus but stopped down to f8. That gives DOF from 0.64 which is the flat focusdistance to the closest area on the floor.
                http://www.panoramas.dk/panorama/fisheye-focus/f8-2m.jpg

                The last one is focused at 0.6m and at f4 to show the actual sharpness at the edge with f4.
                http://www.panoramas.dk/panorama/fisheye-focus/f4-0.6m.jpg

                Problem with fullframe is not that large as it only affects the ground but if you use one like the Nikkor 10.5 or the Tokina shaved on a fullframe camera you get the big problem as it is not possible to do a sharp zenith without a zenith shot unless you stop down to f11.

                Hans

                >
                > Best regards
                > Bernhard
                > > This is a recurrent subject of discussion here and here.
                > > I could not make my mind clear about the actual shape of the region
                > > in space -when shooting with the widest aperture stop set- where the
                > > object can be predicted to be viewed eventually sharp (i.e. in-focus)
                > > on the image.
                > > Almost all responses that I got from searching the Web is how huge is
                > > the DOF with fisheye lenses. This is obviously misleading, at least
                > > with some lenses and at close to moderate shooting range. So I have
                > > attempted to perform a study by myself and to report it:
                > > http://michel.thoby.free.fr/Fisheyes_Focus/Focus_mapping.html
                > > or http://tinyurl.com/nn3e7y
                > >
                >
              • Bernhard Vogl
                Hmm - interesting to see for me. So, at f8 you get sharp images from 0.64 to infinity with the 15mm. This is not the case with my Zenitar 16mm on the D700 at
                Message 7 of 7 , Sep 8, 2009
                • 0 Attachment
                  Hmm - interesting to see for me. So, at f8 you get sharp images from
                  0.64 to infinity with the 15mm.
                  This is not the case with my Zenitar 16mm on the D700 at f8. You can see
                  it e.g. in this recent panorama even at lower resolution:
                  http://www.austria-360.at/ooe/linz/ars2009/20090904-10.html (focused on
                  the robot - look outside the window, left behind the robot).
                  After carefully re-examining the images, i think it is a slight
                  mis-alignment of my lens' focal plane settings (maybe the adapter).
                  Thanks for the enlightment!
                  Bernhard
                Your message has been successfully submitted and would be delivered to recipients shortly.