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

Panos in Very Dimly Lit Small Rooms?

Expand Messages
  • Wheaton, Simon
    I want to shoot some panos of the interiors of Australian mountain huts, such as this one...
    Message 1 of 28 , Apr 19, 2007
      I want to shoot some panos of the interiors of Australian mountain huts,
      such as this one...

      http://geoimages.berkeley.edu/worldwidepanorama/wwp307/html/SimonWheaton
      .html

      (This pano has some issues, I was rushed getting it in by the deadline
      and was unable to fix it up to my satisfaction).



      These huts have no artificial lighting at all, some have no windows,
      some might have a single window per room.



      Any suggestions on how to deal with this situation?



      Should I try to use an HDR method (manual or automated), with no
      artificial lighting at all, to deal with the difference between the
      brightness of light coming through the window and the darkness of the
      room interior?



      Or, should I try to use some sort of artificial light, flash or
      otherwise, to bring the interior light level closer to the light from
      the windows?



      On the huts with no windows, I'm not sure if I'll even have enough light
      to shoot without flash, though they are not well sealed and have some
      light leaking in through cracks in the walls and roof, and I could shoot
      with the door open to provide some light, as in the situation where
      there is a window.



      If flash is the answer, any special requirements on how to set it up? I
      guess that the built-in camera flash wouldn't be any good, as the
      shadows would move as the camera is rotated, causing problems in
      stitching.



      I have to ride into these huts on my mountain bike along fire/management
      trails, so I can't carry much in the way of extra lighting gear.



      Also, I normally shoot outdoor panos, with a Canon 10-22mm lens on a
      350D. When shooting outdoor panos I use an aperture of f/16 to give
      maximum depth of field while still retaining some sharpness. When
      shooting in a small room like this, should I still use f/16, or can I
      open up the aperture to get more light onto the sensor?



      Any advice most appreciated.



      Thanks,

      Simon

      Canberra
      AUSTRALIA

      -----------------------------------------------------------------------
      This email, and any attachments, may be confidential and also privileged. If you are not the intended recipient, please notify the sender and delete all copies of this transmission along with any attachments immediately. You should not copy or use it for any purpose, nor disclose its contents to any other person.
      -----------------------------------------------------------------------


      [Non-text portions of this message have been removed]
    • erik leeman
      Why not use natural artificial light? What would the inhabitants of those huts use (or have used)? A candle or an old lamp could add something meaningfull to
      Message 2 of 28 , Apr 20, 2007
        Why not use 'natural' artificial light? What would the inhabitants of
        those huts use (or have used)? A candle or an old lamp could add
        something meaningfull to the atmosphere of the setting. Your biggest
        problem will be your own shadow, but if you position yourself between
        the window (or door opening) and that lamp, you might be able
        to 'obscure' your shadow(s) with light.
      • Kathy Wheeler
        ... Fascinating challenge. Definitely use a tripod, good pano head with reliable stops on it, and a variety of lighting options. Natural light exposing for
        Message 3 of 28 , Apr 20, 2007
          On 20/04/2007, at 4:54 PM, Wheaton, Simon wrote:
          > These huts have no artificial lighting at all, some have no windows,
          > some might have a single window per room.

          Fascinating challenge.

          Definitely use a tripod, good pano head with reliable stops on it,
          and a variety of lighting options. Natural light exposing for inside
          and outside, soft flash (got an umbrella flash you could take?), and
          perhaps a "period" light - hurricane lamp or similar, would set the
          mood beautifully and be a great prop, especially if shot at night
          (might not be much fun getting back unless you're camping out there).

          There have been some great posts on combining exposures lately (make
          sure your pano-head doesn't creep or tripod move ... ). It might mean
          a fair bit of photoshop work but I think it would be worth it.

          > When
          > shooting in a small room like this, should I still use f/16, or can I
          > open up the aperture to get more light onto the sensor?

          If the rooms are that small f/16 depth of field may be wasted anyway.
          Pace (measure) the room and do the math - see what you can get away
          with.

          Good luck,
          KathyW.
        • Wheaton, Simon
          Thanks for the reply Kathy, and everyone else. I do not own any flash units at the moment, besides the built-in flash on my 350D. Any recommendations for the
          Message 4 of 28 , Apr 20, 2007
            Thanks for the reply Kathy, and everyone else.

            I do not own any flash units at the moment, besides the built-in flash on my 350D. Any recommendations for the most suitable type/brand/model of flash to buy for this sort of situtation?

            Yes, shooting at night might make it easier, avoiding the need for an HDR process to cope with the dark interior and bright light through the window.

            Seeing as winter is fast approaching us down here, I could also light the open fireplace for more light (and atmosphere). And the 'period' light or candle could be positioned to lessen the shadow from myself and the tripod that would be created by the fire.

            Getting back out in the dark wouldn't be a problem, I have a great bike light. To take that WWP Cascade Hut pano I had to ride in from the trailhead carpark in the dark before sunrise, 10kilometres/1hr & 15 minutes to get in there. And as you say, I could just camp the night.

            I'll have to search the net for info regarding setting the aperture dependant on the room size, any idea where to start? I guess that I need to determine the maximum distance and minimum distances of objects in the room (and the room itself) from the camera, what do I do then?

            I have offered to shoot panos for free of the huts for the Kosciuszko Huts Association, the volunteer group that maintains these huts, to use on their website...
            http://www.kosciuskohuts.org.au/ <http://www.kosciuskohuts.org.au/>
            ...they have a page for each hut with info and a photo, I think it would improve their site to have VR panos of the huts as well, and the inside of the hut would really add to it.

            Many thanks,
            Simon
            Canberra
            AUSTRALIA

            ________________________________

            From: Kathy Wheeler
            Sent: Fri 20/04/2007 5:27 PM


            On 20/04/2007, at 4:54 PM, Wheaton, Simon wrote:
            > These huts have no artificial lighting at all, some have no windows,
            > some might have a single window per room.

            Fascinating challenge.

            Definitely use a tripod, good pano head with reliable stops on it,
            and a variety of lighting options. Natural light exposing for inside
            and outside, soft flash (got an umbrella flash you could take?), and
            perhaps a "period" light - hurricane lamp or similar, would set the
            mood beautifully and be a great prop, especially if shot at night
            (might not be much fun getting back unless you're camping out there).

            There have been some great posts on combining exposures lately (make
            sure your pano-head doesn't creep or tripod move ... ). It might mean
            a fair bit of photoshop work but I think it would be worth it.

            > When
            > shooting in a small room like this, should I still use f/16, or can I
            > open up the aperture to get more light onto the sensor?

            If the rooms are that small f/16 depth of field may be wasted anyway.
            Pace (measure) the room and do the math - see what you can get away
            with.

            Good luck,
            KathyW.

            -----------------------------------------------------------------------
            This email, and any attachments, may be confidential and also privileged. If you are not the intended recipient, please notify the sender and delete all copies of this transmission along with any attachments immediately. You should not copy or use it for any purpose, nor disclose its contents to any other person.
            -----------------------------------------------------------------------


            [Non-text portions of this message have been removed]
          • Wheaton, Simon
            Thanks for the reply Erik. Working with the shadows will be a problem, thanks for the idea regarding positioning in relation to the candle/lamp and door/window
            Message 5 of 28 , Apr 20, 2007
              Thanks for the reply Erik.

              Working with the shadows will be a problem, thanks for the idea regarding positioning in relation to the candle/lamp and door/window lighting.

              Thanks,
              Simon
              Canberra
              AUSTRALIA

              ________________________________

              From: erik leeman
              Sent: Fri 20/04/2007 5:13 PM

              Why not use 'natural' artificial light? What would the inhabitants of
              those huts use (or have used)? A candle or an old lamp could add
              something meaningfull to the atmosphere of the setting. Your biggest
              problem will be your own shadow, but if you position yourself between
              the window (or door opening) and that lamp, you might be able
              to 'obscure' your shadow(s) with light.

              -----------------------------------------------------------------------
              This email, and any attachments, may be confidential and also privileged. If you are not the intended recipient, please notify the sender and delete all copies of this transmission along with any attachments immediately. You should not copy or use it for any purpose, nor disclose its contents to any other person.
              -----------------------------------------------------------------------


              [Non-text portions of this message have been removed]
            • Will Brown
              ... [Non-text portions of this message have been removed]
              Message 6 of 28 , Apr 20, 2007
                > How about Large white board such as Fomecore to reflect the light that is
                > coming in from the window? The obvious problem is when shooting into the
                > window is to bracket the exposure and blend the images in photoshop. Even a
                > flashlight (or do you call them torches down under?) that has a
                > omni-directional lens might be of use. If you use a candle or a lantern, it
                > might add a nice accent, but you will still need some sort of fill light for a
                > natural look.
                >
                > Will
                >
                >
                >
                >
                > I want to shoot some panos of the interiors of Australian mountain huts,
                > such as this one...
                >
                > http://geoimages.berkeley.edu/worldwidepanorama/wwp307/html/SimonWheaton
                > .html
                >
                > (This pano has some issues, I was rushed getting it in by the deadline
                > and was unable to fix it up to my satisfaction).
                >
                > These huts have no artificial lighting at all, some have no windows,
                > some might have a single window per room.
                >
                > Any suggestions on how to deal with this situation?
                >
                > Should I try to use an HDR method (manual or automated), with no
                > artificial lighting at all, to deal with the difference between the
                > brightness of light coming through the window and the darkness of the
                > room interior?
                >
                > Or, should I try to use some sort of artificial light, flash or
                > otherwise, to bring the interior light level closer to the light from
                > the windows?
                >
                > On the huts with no windows, I'm not sure if I'll even have enough light
                > to shoot without flash, though they are not well sealed and have some
                > light leaking in through cracks in the walls and roof, and I could shoot
                > with the door open to provide some light, as in the situation where
                > there is a window.
                >
                > If flash is the answer, any special requirements on how to set it up? I
                > guess that the built-in camera flash wouldn't be any good, as the
                > shadows would move as the camera is rotated, causing problems in
                > stitching.
                >
                > I have to ride into these huts on my mountain bike along fire/management
                > trails, so I can't carry much in the way of extra lighting gear.
                >
                > Also, I normally shoot outdoor panos, with a Canon 10-22mm lens on a
                > 350D. When shooting outdoor panos I use an aperture of f/16 to give
                > maximum depth of field while still retaining some sharpness. When
                > shooting in a small room like this, should I still use f/16, or can I
                > open up the aperture to get more light onto the sensor?
                >
                > Any advice most appreciated.
                >
                > Thanks,
                >
                > Simon
                >
                > Canberra
                > AUSTRALIA
                >
                > ----------------------------------------------------------
                > This email, and any attachments, may be confidential and also privileged. If
                > you are not the intended recipient, please notify the sender and delete all
                > copies of this transmission along with any attachments immediately. You should
                > not copy or use it for any purpose, nor disclose its contents to any other
                > person.
                > ----------------------------------------------------------
                >
                > [Non-text portions of this message have been removed]
                >



                [Non-text portions of this message have been removed]
              • Carl von Einem
                Hi Simon, Posted by: Wheaton, Simon ... Halogen light might be easier to control and will also have the right color temperature to fill some interesting
                Message 7 of 28 , Apr 20, 2007
                  Hi Simon,

                  Posted by: "Wheaton, Simon"
                  >
                  > I do not own any flash units at the moment, besides the built-in flash
                  > on my 350D. Any recommendations for the most suitable type/brand/model
                  > of flash to buy for this sort of situtation?

                  Halogen light might be easier to control and will also have the right
                  color temperature to fill some interesting corners of the hut.
                  Experiment with "painting" the room with such a light using a long exposure.
                  You can try one layer of a paper handkerchief to soften the light. Put
                  that in front of the reflector of an electric torch light or a halogen
                  headlight that you can also use for biking (the most extreme product
                  available in Germany for biking is surely the "Big Bang":
                  <http://www.bumm.de/index-e.html?docu/197e.htm> )

                  > Yes, shooting at night might make it easier, avoiding the need for an
                  > HDR process to cope with the dark interior and bright light through the
                  > window.

                  Late afternoon or evening light coming in from the door can also be a
                  nice contrast to the light from a little fire in the fireplace and some
                  candles or a lantern will also add to the atmosphere, as noted before.

                  > Seeing as winter is fast approaching us down here, I could also light
                  > the open fireplace for more light (and atmosphere). And the 'period'
                  > light or candle could be positioned to lessen the shadow from myself and
                  > the tripod that would be created by the fire.

                  Your own shadow will not show up if you use a self timer or a
                  remote-control release from the outside.

                  > I'll have to search the net for info regarding setting the aperture
                  > dependant on the room size, any idea where to start? I guess that I need
                  > to determine the maximum distance and minimum distances of objects in
                  > the room (and the room itself) from the camera, what do I do then?

                  Just try that in your bathroom.

                  > I have offered to shoot panos for free of the huts for the Kosciuszko
                  > Huts Association, the volunteer group that maintains these huts, to use
                  > on their website... <http://www.kosciuskohuts.org.au/>
                  > ...they have a page for each hut with info and a photo, I think it would
                  > improve their site to have VR panos of the huts as well, and the inside
                  > of the hut would really add to it.

                  A series of panoramas from those huts will be really interesting!

                  I just see a photo on <http://kosciuskohuts.org.au> showing a hut
                  wrapped in foil. Looks like this is meant as a protection against
                  bushfires... keeping something cool can definitely look cool :-)
                  Imagine a pano of a sunset with such a hut reflecting the sunset in the
                  foil or a VR object of that hut!

                  Carl
                • Wheaton, Simon
                  That Big Bang bike light looks good, doesn t seem to mention a price on the website though. Yes, that hut wrapped in foil on the Kosciuszko Huts Association
                  Message 8 of 28 , Apr 20, 2007
                    That Big Bang bike light looks good, doesn't seem to mention a price on the website though.

                    Yes, that hut wrapped in foil on the Kosciuszko Huts Association home page is actually Cascade Hut, the same hut as in my WWP submission.

                    They wrapped Cascade Hut and Tin Mines Hut (further down the same fire trail) in fire resistant foil, to prevent them possibly burning down from a bushfire that was threatening the area this past summer. Thankfully the fires didn't reach them. These efforts to protect the huts are in response to the loss of many mountain huts to fire in some very large bushfires back in 2003. It is great that they went to the trouble of doing that to try and protect them from fire, it would be a terrible loss to lose more of these huts.

                    Thanks for your advice,
                    Simon
                    Canberra
                    AUSTRALIA

                    ________________________________

                    From: Carl von Einem
                    Sent: Fri 20/04/2007 11:22 PM

                    Posted by: "Wheaton, Simon"
                    > I have offered to shoot panos for free of the huts for the Kosciuszko
                    > Huts Association, the volunteer group that maintains these huts, to use
                    > on their website... <http://www.kosciuskohuts.org.au/>
                    > ...they have a page for each hut with info and a photo, I think it would
                    > improve their site to have VR panos of the huts as well, and the inside
                    > of the hut would really add to it.

                    A series of panoramas from those huts will be really interesting!

                    I just see a photo on <http://kosciuskohuts.org.au <http://kosciuskohuts.org.au/> > showing a hut
                    wrapped in foil. Looks like this is meant as a protection against
                    bushfires... keeping something cool can definitely look cool :-)
                    Imagine a pano of a sunset with such a hut reflecting the sunset in the
                    foil or a VR object of that hut!

                    Carl

                    -----------------------------------------------------------------------
                    This email, and any attachments, may be confidential and also privileged. If you are not the intended recipient, please notify the sender and delete all copies of this transmission along with any attachments immediately. You should not copy or use it for any purpose, nor disclose its contents to any other person.
                    -----------------------------------------------------------------------


                    [Non-text portions of this message have been removed]
                  • Carl von Einem
                    I think the price is why they called the light Big Bang . That will hit you the very moment when you ask for it in the next store: it s more than 600 Euros!
                    Message 9 of 28 , Apr 21, 2007
                      I think the price is why they called the light "Big Bang". That will hit
                      you the very moment when you ask for it in the next store: it's more
                      than 600 Euros! BANG :-)

                      Pretty much money indeed for "just a lamp", even if it will serve your
                      bike with the same light technology that some of the modern cars use.

                      Anyhow, one of those standard head lights will suffice for the "dual
                      use" biking and hut photography. I also like the idea with the mobile
                      video lamp that was mentioned.

                      Carl

                      Posted by: "Wheaton, Simon"
                      >
                      > That Big Bang bike light looks good, doesn't seem to
                      > mention a price on the website though.
                    • Milko Amorth
                      Hi Carl, ... Makes you go mmmhhh? Bundle a few cheap LEDs wrap them with aluminum foil and put a battery power source to it and charge big bang bucks. Cheers,
                      Message 10 of 28 , Apr 21, 2007
                        Hi Carl,

                        > I think the price is why they called the light "Big Bang". That will hit
                        > you the very moment when you ask for it in the next store: it's more
                        > than 600 Euros! BANG

                        Makes you go mmmhhh? Bundle a few cheap LEDs wrap them with aluminum foil
                        and put a battery power source to it and charge big bang bucks.

                        Cheers, Milko

                        --
                        Milko Amorth
                        360° Immersive Imaging
                        Photographic Virtual Reality
                        VRCanada.ca
                        604.561.5101
                        Skype me @ vrdundee
                        Member of IVRPA.org
                        Contributor to the World Wide Panorama Project
                      • Kathy Wheeler
                        ... Ouch!!! Simon, do a Google search for 10 million candle power . They are great torches, effectively a car headlight!! With some sort of easy to carry
                        Message 11 of 28 , Apr 21, 2007
                          On 21/04/2007, at 5:49 PM, Carl von Einem wrote:
                          > "Big Bang": it's more
                          > than 600 Euros! BANG :-)

                          Ouch!!!

                          Simon, do a Google search for "10 million candle power". They are
                          great torches, effectively a car headlight!! With some sort of easy
                          to carry reflector (white sheet??) it might do the job and be a
                          useful torch at the same time! We have one of these:
                          http://cgi.ebay.com.au/ws/eBayISAPI.dll?ViewItem&item=160107989086
                          and it's brilliant, also great for catching eye-shine off nocturnal
                          wildlife esp kangaroos and wallabys ;-)

                          Cheers,
                          KathyW.
                        • d9616
                          A really useful program for determining the best aperture for your lens/ camera combination is Barnack - found here at http://www.stegmann.dk/mikkel/barnack/
                          Message 12 of 28 , Apr 21, 2007
                            A really useful program for determining the best aperture for your
                            lens/ camera combination is Barnack - found here at

                            http://www.stegmann.dk/mikkel/barnack/

                            simply select your camera/lens combination and adjust the sliders
                            until you find the optimum aperture - in your case F5.6

                            A quick check shows that with the 350D and the zoom set to 10mm
                            (manually set your focus to 1m and keep it there - 1m is the
                            hyperfocal distance) then everything from 0.51m to infinity will be
                            in focus.

                            So set your camera to fully manual - select F5.6 and select 1m on the
                            focus ring - this should be your set up every time you take a pano -
                            I then use a handheld meter to determine my shutter speed at F5.6 -
                            this is the only variable.

                            Just be careful as you get used to using this combo that you dont
                            accidentally alter the settings

                            Hope this is of some help

                            Dave H





                            --- In PanoToolsNG@yahoogroups.com, Kathy Wheeler <kathyw@...> wrote:
                            >
                            >
                            > On 20/04/2007, at 4:54 PM, Wheaton, Simon wrote:
                            > > These huts have no artificial lighting at all, some have no
                            windows,
                            > > some might have a single window per room.
                            >
                            > Fascinating challenge.
                            >
                            > Definitely use a tripod, good pano head with reliable stops on it,
                            > and a variety of lighting options. Natural light exposing for
                            inside
                            > and outside, soft flash (got an umbrella flash you could take?),
                            and
                            > perhaps a "period" light - hurricane lamp or similar, would set
                            the
                            > mood beautifully and be a great prop, especially if shot at night
                            > (might not be much fun getting back unless you're camping out
                            there).
                            >
                            > There have been some great posts on combining exposures lately
                            (make
                            > sure your pano-head doesn't creep or tripod move ... ). It might
                            mean
                            > a fair bit of photoshop work but I think it would be worth it.
                            >
                            > > When
                            > > shooting in a small room like this, should I still use f/16, or
                            can I
                            > > open up the aperture to get more light onto the sensor?
                            >
                            > If the rooms are that small f/16 depth of field may be wasted
                            anyway.
                            > Pace (measure) the room and do the math - see what you can get
                            away
                            > with.
                            >
                            > Good luck,
                            > KathyW.
                            >
                          • Georgia Real Tours
                            ... Hi Simon, I don t have much to offer technically, but that s not why I m replying. Rather, I d like to say your project seems to be desperately needed, as
                            Message 13 of 28 , Apr 23, 2007
                              On 4/20/07, Wheaton, Simon <simon.wheaton@...> wrote:
                              >
                              > I want to shoot some panos of the interiors of Australian mountain huts,
                              > such as this one...
                              >
                              > http://geoimages.berkeley.edu/worldwidepanorama/wwp307/html/SimonWheaton
                              > .html

                              Hi Simon,

                              I don't have much to offer technically, but that's not why I'm
                              replying. Rather, I'd like to say your project seems to be
                              desperately needed, as these structures are obviously so susceptible
                              to fire. It's one thing to be able to see a photo of the outside of
                              the hut yet an entirely different situation to be able to archive the
                              inside of these using the pano technology.

                              Once these things are gone, they're gone. I see some of them can be
                              rebuilt, probably many times over the course of time, but once they
                              are completely consumed in those frequent fires there literally is
                              nothing left but ashes in many cases.

                              Your work would obviously be cherished by KHA, as well as many people
                              and groups down there.

                              I do have one recommendation, and that is to see if you can get
                              permission to drive up there to conduct your project(s). This means
                              pretty much any equipment you need (and camping stuff too) as well an
                              optional generator if needed. It may be against the rules, but if
                              your project is in an 'official' capacity some rules are usually not
                              applicable. Just ask; you'll probably be offered a ride if you aren't
                              allowed by yourself. ;c)

                              Something to consider.

                              Best wishes, and looking forward to your panos!

                              Robert Reese~

                              --
                              Mid GA: 478-599-1300
                              ATL: 678-438-6955
                              garealtours.com
                            • Wheaton, Simon
                              Thanks for this info, and the link to the DOF calculator. The default settings in the calculator are for a film circle of confusion (COC) at .030mm. There are
                              Message 14 of 28 , Apr 25, 2007
                                Thanks for this info, and the link to the DOF calculator.

                                The default settings in the calculator are for a film circle of
                                confusion (COC) at .030mm.

                                There are some links to pages in the calculator help file that mention
                                the COC for a DSLR should be 5 times smaller than film, they suggest a
                                0.0065mm COC...
                                http://www.wrotniak.net/photo/tech/dof.html

                                Any idea which size COC to choose when calculating hyperfocal distances
                                on a DSLR such as a Canon 350D?

                                I would normally set shutter speed according to some test shots and
                                checking the camera histogram, to check for blown highlights in the
                                brightest frame of the pano sequence. Is it better to use a handheld
                                meter for this?

                                Thanks,
                                Simon
                                Canberra
                                AUSTRALIA

                                -----Original Message-----
                                From: d9616
                                Sent: Saturday, 21 April 2007 10:48 PM

                                A really useful program for determining the best aperture for your
                                lens/ camera combination is Barnack - found here at

                                http://www.stegmann.dk/mikkel/barnack/

                                simply select your camera/lens combination and adjust the sliders
                                until you find the optimum aperture - in your case F5.6

                                A quick check shows that with the 350D and the zoom set to 10mm
                                (manually set your focus to 1m and keep it there - 1m is the
                                hyperfocal distance) then everything from 0.51m to infinity will be
                                in focus.

                                So set your camera to fully manual - select F5.6 and select 1m on the
                                focus ring - this should be your set up every time you take a pano -
                                I then use a handheld meter to determine my shutter speed at F5.6 -
                                this is the only variable.

                                Dave H

                                -----------------------------------------------------------------------
                                This email, and any attachments, may be confidential and also privileged. If you are not the intended recipient, please notify the sender and delete all copies of this transmission along with any attachments immediately. You should not copy or use it for any purpose, nor disclose its contents to any other person.
                                -----------------------------------------------------------------------
                              • d9616
                                Hi Simon, I see your point about the COC reference point! I didn t have that problem as I have the full frame equivalent sensor. I did a little browsing for
                                Message 15 of 28 , Apr 26, 2007
                                  Hi Simon,

                                  I see your point about the COC reference point!
                                  I didn't have that problem as I have the full frame equivalent sensor.
                                  I did a little browsing for you and found this on wikipedia

                                  http://en.wikipedia.org/wiki/Circle_of_confusion

                                  For an APS sized sensor of 22.5 mm x 15.0 mm the COC is calculated
                                  at 0.018 mm , I see the sensor size on your 350 is 22.2 x 14.8 mm
                                  which would make the value for yours slightly smaller say - 0.016

                                  (This is my best guess-timate - i got about a third of the way down
                                  the wiki page before my brain fell out lol)

                                  Using the 0.016 in the Barnack calculator, the focal length at
                                  10mm, an aperture of F11 and the focus distance at 1m - you should
                                  have everything in focus from 0.5m to infinity.

                                  Maybe a few test shots at these settings is worth a try? failing
                                  that maybe a lurking physics professor on the list might care to lend
                                  a hand with a more precise answer based on the wiki formulas?

                                  I prefer to use a handheld meter to take an average reading for
                                  speed really. If i have the time then I might take spot readings from
                                  highlight and shadow areas and work out an average - I will usually
                                  bias the exposure towards the highlights as once detail has gone in
                                  those - it has gone for good. This is just how I prefer to work -
                                  Your method of checking for blown highlights on the histogram seems
                                  fine to me.
                                  I shoot RAW, which gives me a little lattitude during RAW
                                  conversion - I use Capture one which allows the user to perform
                                  uniform adjustments to the full set of pre stitched images at the
                                  same time. In the past i have tried altering just single frames
                                  seperately - but this usually shows up as stitching lines in the
                                  final image.
                                  Once stitched i can then bring out some more details in the final
                                  equirectangular using the shadow/highlight tool or a simple curves
                                  adjustment.

                                  Kindest Regards

                                  Dave H


                                  >
                                  > Thanks for this info, and the link to the DOF calculator.
                                  >
                                  > The default settings in the calculator are for a film circle of
                                  > confusion (COC) at .030mm.
                                  >
                                  > There are some links to pages in the calculator help file that
                                  mention
                                  > the COC for a DSLR should be 5 times smaller than film, they
                                  suggest a
                                  > 0.0065mm COC...
                                  > http://www.wrotniak.net/photo/tech/dof.html
                                  >
                                  > Any idea which size COC to choose when calculating hyperfocal
                                  distances
                                  > on a DSLR such as a Canon 350D?
                                  >
                                  > I would normally set shutter speed according to some test shots and
                                  > checking the camera histogram, to check for blown highlights in the
                                  > brightest frame of the pano sequence. Is it better to use a handheld
                                  > meter for this?
                                  >
                                  > Thanks,
                                  > Simon
                                  > Canberra
                                  > AUSTRALIA
                                  >
                                  > -----Original Message-----
                                  > From: d9616
                                  > Sent: Saturday, 21 April 2007 10:48 PM
                                  >
                                  > A really useful program for determining the best aperture for your
                                  > lens/ camera combination is Barnack - found here at
                                  >
                                  > http://www.stegmann.dk/mikkel/barnack/
                                  >
                                  > simply select your camera/lens combination and adjust the sliders
                                  > until you find the optimum aperture - in your case F5.6
                                  >
                                  > A quick check shows that with the 350D and the zoom set to 10mm
                                  > (manually set your focus to 1m and keep it there - 1m is the
                                  > hyperfocal distance) then everything from 0.51m to infinity will be
                                  > in focus.
                                  >
                                  > So set your camera to fully manual - select F5.6 and select 1m on
                                  the
                                  > focus ring - this should be your set up every time you take a pano -

                                  > I then use a handheld meter to determine my shutter speed at F5.6 -
                                  > this is the only variable.
                                  >
                                  > Dave H
                                  >
                                  > --------------------------------------------------------------------
                                  ---
                                  > This email, and any attachments, may be confidential and also
                                  privileged. If you are not the intended recipient, please notify the
                                  sender and delete all copies of this transmission along with any
                                  attachments immediately. You should not copy or use it for any
                                  purpose, nor disclose its contents to any other person.
                                  > --------------------------------------------------------------------
                                  ---
                                  >
                                • Mark D. Fink
                                  I ve been tinkering with this, (Barnack software), and trying my best to get my head wrapped around the Wiki article. One thing that occurred to me is that the
                                  Message 16 of 28 , Apr 26, 2007
                                    I've been tinkering with this, (Barnack software), and trying my best to
                                    get my head wrapped around the Wiki article. One thing that occurred to
                                    me is that the COC seems to be dependent on the magnification factor of
                                    the final print. OK, this I can follow. A smaller sensor will need to be
                                    enlarged more than a larger sensor that has the same size photosites.

                                    However, we panoheads don't follow these rules, do we. No! We drive our
                                    spouses or significant others to tears by taking dozens of images and
                                    stitching them together into a mosaic. Because of that, what COC do we
                                    use? In effect, we have created an image sensor with a much larger
                                    surface area, and therefore, the final image doesn't need to be enlarged
                                    nearly as much. Actually, there are many times when I am reducing the
                                    size of the image for printing, simply because I got carried away while
                                    shooting. :o)

                                    So, how does this affect COC?

                                    Mark
                                    www.pinnacle-vr.com
                                    www.northernlight.net

                                    -----Original Message-----
                                    From: PanoToolsNG@yahoogroups.com [mailto:PanoToolsNG@yahoogroups.com]
                                    On Behalf Of d9616
                                    Sent: Thursday, April 26, 2007 8:03 AM
                                    To: PanoToolsNG@yahoogroups.com
                                    Subject: [PanoToolsNG] Re: Panos in Very Dimly Lit Small Rooms?



                                    Hi Simon,

                                    I see your point about the COC reference point!
                                    I didn't have that problem as I have the full frame equivalent sensor.
                                    I did a little browsing for you and found this on wikipedia

                                    http://en.wikipedia <http://en.wikipedia.org/wiki/Circle_of_confusion>
                                    .org/wiki/Circle_of_confusion

                                    For an APS sized sensor of 22.5 mm x 15.0 mm the COC is calculated
                                    at 0.018 mm , I see the sensor size on your 350 is 22.2 x 14.8 mm
                                    which would make the value for yours slightly smaller say - 0.016

                                    (This is my best guess-timate - i got about a third of the way down
                                    the wiki page before my brain fell out lol)

                                    Using the 0.016 in the Barnack calculator, the focal length at
                                    10mm, an aperture of F11 and the focus distance at 1m - you should
                                    have everything in focus from 0.5m to infinity.

                                    Maybe a few test shots at these settings is worth a try? failing
                                    that maybe a lurking physics professor on the list might care to lend
                                    a hand with a more precise answer based on the wiki formulas?

                                    I prefer to use a handheld meter to take an average reading for
                                    speed really. If i have the time then I might take spot readings from
                                    highlight and shadow areas and work out an average - I will usually
                                    bias the exposure towards the highlights as once detail has gone in
                                    those - it has gone for good. This is just how I prefer to work -
                                    Your method of checking for blown highlights on the histogram seems
                                    fine to me.
                                    I shoot RAW, which gives me a little lattitude during RAW
                                    conversion - I use Capture one which allows the user to perform
                                    uniform adjustments to the full set of pre stitched images at the
                                    same time. In the past i have tried altering just single frames
                                    seperately - but this usually shows up as stitching lines in the
                                    final image.
                                    Once stitched i can then bring out some more details in the final
                                    equirectangular using the shadow/highlight tool or a simple curves
                                    adjustment.

                                    Kindest Regards

                                    Dave H

                                    >
                                    > Thanks for this info, and the link to the DOF calculator.
                                    >
                                    > The default settings in the calculator are for a film circle of
                                    > confusion (COC) at .030mm.
                                    >
                                    > There are some links to pages in the calculator help file that
                                    mention
                                    > the COC for a DSLR should be 5 times smaller than film, they
                                    suggest a
                                    > 0.0065mm COC...
                                    > http://www.wrotniak <http://www.wrotniak.net/photo/tech/dof.html>
                                    .net/photo/tech/dof.html
                                    >
                                    > Any idea which size COC to choose when calculating hyperfocal
                                    distances
                                    > on a DSLR such as a Canon 350D?
                                    >
                                    > I would normally set shutter speed according to some test shots and
                                    > checking the camera histogram, to check for blown highlights in the
                                    > brightest frame of the pano sequence. Is it better to use a handheld
                                    > meter for this?
                                    >
                                    > Thanks,
                                    > Simon
                                    > Canberra
                                    > AUSTRALIA
                                    >
                                    > -----Original Message-----
                                    > From: d9616
                                    > Sent: Saturday, 21 April 2007 10:48 PM
                                    >
                                    > A really useful program for determining the best aperture for your
                                    > lens/ camera combination is Barnack - found here at
                                    >
                                    > http://www.stegmann <http://www.stegmann.dk/mikkel/barnack/>
                                    .dk/mikkel/barnack/
                                    >
                                    > simply select your camera/lens combination and adjust the sliders
                                    > until you find the optimum aperture - in your case F5.6
                                    >
                                    > A quick check shows that with the 350D and the zoom set to 10mm
                                    > (manually set your focus to 1m and keep it there - 1m is the
                                    > hyperfocal distance) then everything from 0.51m to infinity will be
                                    > in focus.
                                    >
                                    > So set your camera to fully manual - select F5.6 and select 1m on
                                    the
                                    > focus ring - this should be your set up every time you take a pano -

                                    > I then use a handheld meter to determine my shutter speed at F5.6 -
                                    > this is the only variable.
                                    >
                                    > Dave H
                                    >
                                    > ----------------------------------------------------------
                                    ---
                                    > This email, and any attachments, may be confidential and also
                                    privileged. If you are not the intended recipient, please notify the
                                    sender and delete all copies of this transmission along with any
                                    attachments immediately. You should not copy or use it for any
                                    purpose, nor disclose its contents to any other person.
                                    > ----------------------------------------------------------
                                    ---
                                    >







                                    [Non-text portions of this message have been removed]
                                  • Victor
                                    Mark, ... Perhaps I m just oversimplifying, but it seems to me that the fact we re stitching a few, or dozens, or sometimes many more photos together doesn t
                                    Message 17 of 28 , Apr 26, 2007
                                      Mark,

                                      > In effect, we have created an image sensor with a much larger
                                      > surface area, and therefore, the final image doesn't need to
                                      > be enlarged nearly as much.

                                      Perhaps I'm just oversimplifying, but it seems to me that the fact
                                      we're stitching a few, or dozens, or sometimes many more photos
                                      together doesn't matter a whit. If an object is not in focus on any
                                      ONE photo, combining many photos won't matter ... unless you're
                                      combining photos taken at different focus positions and merging for
                                      enhanced depth of field, much like merging different exposures to get
                                      enhanced dynamic range.

                                      The fact that you don't need to enlarge as much helps, to be sure. As
                                      I see it however, at the end of the day, if it ain't in focus, it
                                      simply ain't in focus.

                                      Depth of field is really just a measure of how far OUT of focus you're
                                      willing to tolerate. There's really only one plane of focus.
                                      Everything in front of or behind that plane it is out of focus,
                                      period. COC is a way of getting the math to work out to your level
                                      of tolerance.

                                      To sum up, use the COC for the sensor/film size you are using.


                                      Victor
                                    • Mark D. Fink
                                      Hi Victor, ... Yes, that s exactly the way I m looking at it - combining multiple photos and merging for enhanced depth of field. ... you re ... Agreed, but
                                      Message 18 of 28 , Apr 26, 2007
                                        Hi Victor,

                                        > Mark,

                                        > > In effect, we have created an image sensor with a much larger
                                        > > surface area, and therefore, the final image doesn't need to
                                        > > be enlarged nearly as much.

                                        > Perhaps I'm just oversimplifying, but it seems to me that the fact
                                        > we're stitching a few, or dozens, or sometimes many more photos
                                        > together doesn't matter a whit. If an object is not in focus on any
                                        > ONE photo, combining many photos won't matter ... unless you're
                                        > combining photos taken at different focus positions and merging for
                                        > enhanced depth of field, much like merging different exposures to get
                                        > enhanced dynamic range.

                                        Yes, that's exactly the way I'm looking at it - combining multiple
                                        photos and merging for enhanced depth of field.

                                        > The fact that you don't need to enlarge as much helps, to be sure. As
                                        > I see it however, at the end of the day, if it ain't in focus, it
                                        > simply ain't in focus.

                                        > Depth of field is really just a measure of how far OUT of focus
                                        you're
                                        > willing to tolerate. There's really only one plane of focus.
                                        > Everything in front of or behind that plane it is out of focus,
                                        > period. COC is a way of getting the math to work out to your level
                                        > of tolerance.

                                        Agreed, but using the Barnack calculator, changing the COC (i.e. for a
                                        Canon 1Ds Mark II vs. a Canon 10D) dramatically affects the "acceptable"
                                        depth of field. As an extreme example, a 50mm lens at f8 focused at 24
                                        feet will give you a depth of field of 9.21 ft. using a COC of .01mm.
                                        Change the COC to .04mm, and the depth of field (according to the
                                        calculator) goes up to 78.87 ft. What do I choose? I have no idea, but
                                        it seems to me that the .016 suggested for the 10D is too conservative
                                        when shooting a mosaic.

                                        > To sum up, use the COC for the sensor/film size you are using.

                                        > Victor

                                        Mark
                                        www.pinnacle-vr.com
                                        www.northernlight.net



                                        [Non-text portions of this message have been removed]
                                      • Wheaton, Simon
                                        If creating online panos, then you might need to enlarge much more than print, depending on how far you allow the viewer to zoom in. Or, with printed panos,
                                        Message 19 of 28 , Apr 26, 2007
                                          If creating online panos, then you might need to enlarge much more than print, depending on how far you allow the viewer to zoom in.

                                          Or, with printed panos, you are probably going to print a very wide image, possibly the vertical height might be the same, therefore possibly the same amount of enlargement.

                                          What is the COC for the sensor/film size, is it the same for digital and film, just depending on the size?

                                          Simon
                                          Canberra
                                          AUSTRALIA

                                          ________________________________

                                          From: Victor
                                          Sent: Fri 27/04/2007 1:08 AM



                                          The fact that you don't need to enlarge as much helps, to be sure.

                                          ...

                                          To sum up, use the COC for the sensor/film size you are using.

                                          Victor

                                          -----------------------------------------------------------------------
                                          This email, and any attachments, may be confidential and also privileged. If you are not the intended recipient, please notify the sender and delete all copies of this transmission along with any attachments immediately. You should not copy or use it for any purpose, nor disclose its contents to any other person.
                                          -----------------------------------------------------------------------


                                          [Non-text portions of this message have been removed]
                                        • d9616
                                          OK Ok I think I m getting somewhere, Using the mentioned Zeiss formula as quoted on the wiki page: d is the diagonal measurement for the sensor sensor size
                                          Message 20 of 28 , Apr 26, 2007
                                            OK Ok I think I'm getting somewhere,

                                            Using the mentioned "Zeiss formula" as quoted on the wiki page:

                                            d is the diagonal measurement for the sensor

                                            sensor size on 350d is 22.2 x 14.8mm

                                            using this useful calculator i found 'd' to be 26.68mm

                                            http://www.csgnetwork.com/righttricalc.html

                                            so using zeiss = d/1500 = 0.017786666

                                            i guess you could round that up to a 0.18 COC Reference

                                            If the zeiss formula is taken as d/1730

                                            then 26.68/1730 = 0.0154219 or 0.15 COC reference

                                            However as to which to use version of the formaula to use is still a
                                            bit of a mystery, wikipedia only offers:
                                            "The Zeiss formula is apocryphal, in the sense that it has grown to
                                            be a well-known named concept by propagation through the internet,
                                            even though it has no official origin, little connection to Carl
                                            Zeiss Company, and no recognition or usage in the photographic
                                            industry outside the web community."

                                            but seems to support more the d/1730 version rather than the d/1500

                                            Hope this helps a little

                                            Dave H







                                            >
                                            > > > In effect, we have created an image sensor with a much larger
                                            > > > surface area, and therefore, the final image doesn't need to
                                            > > > be enlarged nearly as much.
                                            >
                                            > > Perhaps I'm just oversimplifying, but it seems to me that the
                                            fact
                                            > > we're stitching a few, or dozens, or sometimes many more photos
                                            > > together doesn't matter a whit. If an object is not in focus on
                                            any
                                            > > ONE photo, combining many photos won't matter ... unless you're
                                            > > combining photos taken at different focus positions and merging
                                            for
                                            > > enhanced depth of field, much like merging different exposures
                                            to get
                                            > > enhanced dynamic range.
                                            >
                                            > Yes, that's exactly the way I'm looking at it - combining multiple
                                            > photos and merging for enhanced depth of field.
                                            >
                                            > > The fact that you don't need to enlarge as much helps, to be
                                            sure. As
                                            > > I see it however, at the end of the day, if it ain't in focus, it
                                            > > simply ain't in focus.
                                            >
                                            > > Depth of field is really just a measure of how far OUT of focus
                                            > you're
                                            > > willing to tolerate. There's really only one plane of focus.
                                            > > Everything in front of or behind that plane it is out of focus,
                                            > > period. COC is a way of getting the math to work out to your
                                            level
                                            > > of tolerance.
                                            >
                                            > Agreed, but using the Barnack calculator, changing the COC (i.e.
                                            for a
                                            > Canon 1Ds Mark II vs. a Canon 10D) dramatically affects
                                            the "acceptable"
                                            > depth of field. As an extreme example, a 50mm lens at f8 focused at
                                            24
                                            > feet will give you a depth of field of 9.21 ft. using a COC
                                            of .01mm.
                                            > Change the COC to .04mm, and the depth of field (according to the
                                            > calculator) goes up to 78.87 ft. What do I choose? I have no idea,
                                            but
                                            > it seems to me that the .016 suggested for the 10D is too
                                            conservative
                                            > when shooting a mosaic.
                                            >
                                            > > To sum up, use the COC for the sensor/film size you are using.
                                            >
                                            > > Victor
                                            >
                                            > Mark
                                            > www.pinnacle-vr.com
                                            > www.northernlight.net
                                            >
                                            >
                                            >
                                            > [Non-text portions of this message have been removed]
                                            >
                                          • d9616
                                            CORRECTION that should read when d/1500 then the COC is 0.018 when d/1730 then the COC is 0.015 wong decimal place there! ( I don t think I was *that* far out
                                            Message 21 of 28 , Apr 26, 2007
                                              CORRECTION that should read

                                              when d/1500 then the COC is 0.018

                                              when d/1730 then the COC is 0.015

                                              wong decimal place there!

                                              ( I don't think I was *that* far out with that guesstimate of 0.016)

                                              Dave H


                                              >
                                              > OK Ok I think I'm getting somewhere,
                                              >
                                              > Using the mentioned "Zeiss formula" as quoted on the wiki page:
                                              >
                                              > d is the diagonal measurement for the sensor
                                              >
                                              > sensor size on 350d is 22.2 x 14.8mm
                                              >
                                              > using this useful calculator i found 'd' to be 26.68mm
                                              >
                                              > http://www.csgnetwork.com/righttricalc.html
                                              >
                                              > so using zeiss = d/1500 = 0.017786666
                                              >
                                              > i guess you could round that up to a 0.18 COC Reference
                                              >
                                              > If the zeiss formula is taken as d/1730
                                              >
                                              > then 26.68/1730 = 0.0154219 or 0.15 COC reference
                                              >
                                              > However as to which to use version of the formaula to use is still
                                              a
                                              > bit of a mystery, wikipedia only offers:
                                              > "The Zeiss formula is apocryphal, in the sense that it has grown to
                                              > be a well-known named concept by propagation through the internet,
                                              > even though it has no official origin, little connection to Carl
                                              > Zeiss Company, and no recognition or usage in the photographic
                                              > industry outside the web community."
                                              >
                                              > but seems to support more the d/1730 version rather than the d/1500
                                              >
                                              > Hope this helps a little
                                              >
                                              > Dave H
                                              >
                                              >
                                              >
                                              >
                                              >
                                              >
                                              >
                                              > >
                                              > > > > In effect, we have created an image sensor with a much larger
                                              > > > > surface area, and therefore, the final image doesn't need to
                                              > > > > be enlarged nearly as much.
                                              > >
                                              > > > Perhaps I'm just oversimplifying, but it seems to me that the
                                              > fact
                                              > > > we're stitching a few, or dozens, or sometimes many more photos
                                              > > > together doesn't matter a whit. If an object is not in focus
                                              on
                                              > any
                                              > > > ONE photo, combining many photos won't matter ... unless you're
                                              > > > combining photos taken at different focus positions and
                                              merging
                                              > for
                                              > > > enhanced depth of field, much like merging different exposures
                                              > to get
                                              > > > enhanced dynamic range.
                                              > >
                                              > > Yes, that's exactly the way I'm looking at it - combining
                                              multiple
                                              > > photos and merging for enhanced depth of field.
                                              > >
                                              > > > The fact that you don't need to enlarge as much helps, to be
                                              > sure. As
                                              > > > I see it however, at the end of the day, if it ain't in focus,
                                              it
                                              > > > simply ain't in focus.
                                              > >
                                              > > > Depth of field is really just a measure of how far OUT of focus
                                              > > you're
                                              > > > willing to tolerate. There's really only one plane of focus.
                                              > > > Everything in front of or behind that plane it is out of focus,
                                              > > > period. COC is a way of getting the math to work out to your
                                              > level
                                              > > > of tolerance.
                                              > >
                                              > > Agreed, but using the Barnack calculator, changing the COC (i.e.
                                              > for a
                                              > > Canon 1Ds Mark II vs. a Canon 10D) dramatically affects
                                              > the "acceptable"
                                              > > depth of field. As an extreme example, a 50mm lens at f8 focused
                                              at
                                              > 24
                                              > > feet will give you a depth of field of 9.21 ft. using a COC
                                              > of .01mm.
                                              > > Change the COC to .04mm, and the depth of field (according to the
                                              > > calculator) goes up to 78.87 ft. What do I choose? I have no
                                              idea,
                                              > but
                                              > > it seems to me that the .016 suggested for the 10D is too
                                              > conservative
                                              > > when shooting a mosaic.
                                              > >
                                              > > > To sum up, use the COC for the sensor/film size you are using.
                                              > >
                                              > > > Victor
                                              > >
                                              > > Mark
                                              > > www.pinnacle-vr.com
                                              > > www.northernlight.net
                                              > >
                                              > >
                                              > >
                                              > > [Non-text portions of this message have been removed]
                                              > >
                                              >
                                            • Victor
                                              Simon, ... True, but I think COC would be swamped by compression artifacts if zoomed to a high degree for most online panos. ... Yes. To paraphrase from a
                                              Message 22 of 28 , Apr 26, 2007
                                                Simon,

                                                > If creating online panos, then you might need to enlarge much
                                                > more than print, depending on how far you allow the viewer to
                                                > zoom in.

                                                True, but I think COC would be swamped by compression artifacts if
                                                zoomed to a high degree for most online panos.

                                                > What is the COC for the sensor/film size, is it the same
                                                > for digital and film, just depending on the size?

                                                Yes. To paraphrase from a variety of sources:

                                                The math behind determining the COC and, by extension, DOF is based on
                                                visual acuity. It is generlly accepted that the smallest detail the
                                                human eye can resolve is approximately 1' (one minute) of arc, this
                                                works out to something like 1/16th of a millimeter separation at
                                                "normal" reading distances. Put two lines closer together than this
                                                and they appear to be one line.

                                                Lens makers assume that the print you make will require a 5X
                                                enlargement from the original size "in the camera" (be that sensor,
                                                film, whatever). Taking 1/16th divided by 5 comes out to 0.033 which,
                                                by the way, is very close to what Canon uses for their lenses.

                                                A 5X enlargement of a 35mm film image is approximately a 5x7 print,
                                                doesn't sound very "enlarged". However, with larger prints, it is
                                                assumed that the normal viewing distance will also be longer (farther)
                                                so the relationship, or separation, between the smallest details
                                                should remain approximately the same.

                                                BTW, I remember reading that Leica uses 0.025 for the COC calculations
                                                ... smaller DOF ... tighter definition of what's "in focus". :-)

                                                The math is the same regardless of digital or film. It's all a
                                                question of the amount of enlargement. That's why COC numbers for
                                                large format film lenses can be much larger ... less enlargement is
                                                required to get to a given print size.

                                                As I stated previously, however, DOF is an illusion. Only one plane
                                                of focus is "real". The math used to define the distances where the
                                                eye can't distinguish that things in front of or behind that plane are
                                                out of focus is constrained by the COC used.

                                                Of course, there's a disjunction between acuity and focus. All of the
                                                above talks about the ability to distinguish fine detail, whereas DOF
                                                is about whether things are fuzzy or not. However, you've got to base
                                                the math on something, right? So acuity is as good as anything else.
                                                Ultimately though, it boils down to how "fuzzy" is "too" fuzzy and
                                                that, unfortunately, is a subjective answer that no mathematics can
                                                provide an answer for.

                                                Victor
                                              • Serge Maandag
                                                ... I think it is quite simple, if it s explained in laymens terms: A sharp point that is in focus will be a sharp point on the sensor. A sharp point that is
                                                Message 23 of 28 , Apr 26, 2007
                                                  Mark D. Fink wrote:
                                                  > I've been tinkering with this, (Barnack software), and trying my best to
                                                  > get my head wrapped around the Wiki article. One thing that occurred to
                                                  > me is that the COC seems to be dependent on the magnification factor of
                                                  > the final print. OK, this I can follow. A smaller sensor will need to be
                                                  > enlarged more than a larger sensor that has the same size photosites
                                                  I think it is quite simple, if it's explained in laymens terms:

                                                  A sharp point that is in focus will be a sharp point on the sensor.
                                                  A sharp point that is not in focus will be a small disc shaped spot on
                                                  the sensor. That disc will project the original point as a blur over
                                                  multiple pixels.
                                                  The disc is called the Circle of Confusion (CoC).

                                                  So how do you determine the maximum size for that CoC? It's determined
                                                  by the amount of pixels you can discern with your eye.

                                                  If you print the image, that is decided by:
                                                  - The size of the print
                                                  - The distance from which you view the print
                                                  - The quality of your eyes.

                                                  A bit of math and an estimate for the average eye quality of human
                                                  beings will get you the maximum allowed CoC.

                                                  With panoramas you can skip the math in most cases. Usually you allow
                                                  your viewers to zoom in on the panorama until they can discern every
                                                  pixel. So the simple rule is that the CoC is not allowed to be big
                                                  enough to influence other pixels. Now most of the sensors do not have
                                                  individual pixels. Instead they have a so called bayer array where the
                                                  receptors for red, green and blue are adjacent and there are more green
                                                  receptors. So usually it is said that the CoC can be a bit bigger. In
                                                  the Barnack program there's a checkbox called "CoC from pixel pitch",
                                                  it seems to set the CoC to a width of 2 pixels.

                                                  And of course the CoC is determined not only by the sensor size, but
                                                  also by the sensor resolution.

                                                  For my Nikon D70s I usually use a CoC of 0.02mm, although according to
                                                  the Barnack calculator, I should be using 0.0158mm.

                                                  Serge.

                                                  p.s. The tighter you set the CoC, the harder it will be to get
                                                  hyperfocus. You can of course open the aperture further, but beware of
                                                  diffraction. A smaller sensor demands a smaller CoC, but it also suffers
                                                  from diffraction quicker. So it's a bit of a catch 20-20 problem.
                                                • Erik Krause
                                                  ... Did you read http://wiki.panotools.org/Depth_of_Field ? It should all be answered there, especially in the paragraphs Considerations for VR panoramas and
                                                  Message 24 of 28 , Apr 26, 2007
                                                    On Thursday, April 26, 2007 at 10:51, Mark D. Fink wrote:

                                                    > Because of that, what COC do we
                                                    > use? In effect, we have created an image sensor with a much larger
                                                    > surface area, and therefore, the final image doesn't need to be enlarged
                                                    > nearly as much. Actually, there are many times when I am reducing the
                                                    > size of the image for printing, simply because I got carried away while
                                                    > shooting. :o)

                                                    Did you read http://wiki.panotools.org/Depth_of_Field ? It should all
                                                    be answered there, especially in the paragraphs "Considerations for
                                                    VR panoramas" and "Hyperfocal distance". If it is unclear, please let
                                                    me know...

                                                    best regards
                                                    --
                                                    Copyright (c) 2007 Erik Krause
                                                    Verbatim copying and distribution strictly forbidden
                                                    except those allowed in wiki.panotools.org/User_Guidelines
                                                  • Erik Krause
                                                    ... True, but you have to look at the resulting pixels in the pano, not the sensor pixels - especially if you downsize the pano. The formula is given in the
                                                    Message 25 of 28 , Apr 26, 2007
                                                      On Thursday, April 26, 2007 at 21:26, Serge Maandag wrote:

                                                      > With panoramas you can skip the math in most cases. Usually you allow
                                                      > your viewers to zoom in on the panorama until they can discern every
                                                      > pixel. So the simple rule is that the CoC is not allowed to be big
                                                      > enough to influence other pixels.

                                                      True, but you have to look at the resulting pixels in the pano, not
                                                      the sensor pixels - especially if you downsize the pano. The formula
                                                      is given in the wiki.

                                                      > Now most of the sensors do not have individual pixels. Instead they
                                                      > have a so called bayer array where the receptors for red, green and
                                                      > blue are adjacent and there are more green receptors. So usually it is
                                                      > said that the CoC can be a bit bigger. In the Barnack program there's
                                                      > a checkbox called "CoC from pixel pitch", it seems to set the CoC to
                                                      > a width of 2 pixels.

                                                      The reason for taking 2 pixels is not the bayer pattern, it is the
                                                      simple fact that one pixel can not show blur - you need at least two.

                                                      There is a common misunderstanding regarding the bayer pattern. You
                                                      can often read that the resulting resolution is lower because of the
                                                      pattern. This is only partly true: The color resolution is lower
                                                      indeed, but the luminance resolution is not. This is because all
                                                      pixels contribute to the luminance information but only the colored
                                                      ones to the respective color.

                                                      In most images most of the information is in the luminance channel,
                                                      hence the bayer pattern has not such a bad influence. You can see
                                                      that if you change an image into Lab color space and look separately
                                                      on the L channel (which contains the luminance info) and the a and b
                                                      channels (which contain color info).

                                                      best regards
                                                      --
                                                      Copyright (c) 2007 Erik Krause
                                                      Verbatim copying and distribution strictly forbidden
                                                      except those allowed in wiki.panotools.org/User_Guidelines
                                                    • Pat Swovelin
                                                      ... Opening the aperture will worsen the DoF (i.e., worsen the hyperfocal distance), you really mean close the aperture to increase the DoF (i.e., giving you
                                                      Message 26 of 28 , Apr 26, 2007
                                                        Serge Maandag wrote:
                                                        > p.s. The tighter you set the CoC, the harder it will be to get
                                                        > hyperfocus. You can of course open the aperture further,

                                                        Opening the aperture will worsen the DoF (i.e., worsen the hyperfocal
                                                        distance), you really mean close the aperture to increase the DoF
                                                        (i.e., giving you a better hyperfocal distance), right?

                                                        > but beware of
                                                        > diffraction. A smaller sensor demands a smaller CoC, but it also suffers
                                                        > from diffraction quicker. So it's a bit of a catch 20-20 problem.
                                                        That's Catch 22...





                                                        Pat Swovelin
                                                        Cool Guy @ Large
                                                      • Sacha Griffin
                                                        I think its catch 20-20 when its really obvious and you can see it. ;) Worsen is a bad/negative word. Shorten would be more appropriate. I encourage everyone
                                                        Message 27 of 28 , Apr 26, 2007
                                                          I think its catch 20-20 when its really obvious and you can see it. ;)



                                                          Worsen is a bad/negative word. Shorten would be more appropriate.



                                                          I encourage everyone not to rely on calculators/measurements, but actually
                                                          do tests and verify.



                                                          Its easy to get caught in some measurement on a website, but many variables
                                                          can affect what optimal settings you need.

                                                          Adapters/manufacturing slop etc.



                                                          Sacha Griffin
                                                          Southern Digital Solutions LLC
                                                          www.southern-digital.com
                                                          www.seeit360.net
                                                          www.ezphotosafe.com
                                                          404-551-4275
                                                          404-731-7798

                                                          _____

                                                          From: Pat Swovelin [mailto:Panoramas@...]
                                                          Sent: Thursday, April 26, 2007 6:32 PM
                                                          To: PanoToolsNG@yahoogroups.com
                                                          Subject: Re: Depth of field calculations (was RE: [PanoToolsNG] Re: Panos in
                                                          Very Dimly Lit Small Rooms?)



                                                          Serge Maandag wrote:
                                                          > p.s. The tighter you set the CoC, the harder it will be to get
                                                          > hyperfocus. You can of course open the aperture further,

                                                          Opening the aperture will worsen the DoF (i.e., worsen the hyperfocal
                                                          distance), you really mean close the aperture to increase the DoF
                                                          (i.e., giving you a better hyperfocal distance), right?

                                                          > but beware of
                                                          > diffraction. A smaller sensor demands a smaller CoC, but it also suffers
                                                          > from diffraction quicker. So it's a bit of a catch 20-20 problem.
                                                          That's Catch 22...

                                                          Pat Swovelin
                                                          Cool Guy @ Large





                                                          [Non-text portions of this message have been removed]
                                                        • PanoToolsNG.10.m8@spamgourmet.com
                                                          Hmm... It feels like you are trying to combine two metaphors.. Common usage of 20-20 vision is, perfect vision , which implies, obvious . (this actually is
                                                          Message 28 of 28 , Apr 26, 2007
                                                            Hmm...
                                                            It feels like you are trying to combine two metaphors..

                                                            Common usage of "20-20 vision" is, "perfect vision", which implies,
                                                            "obvious".
                                                            (this actually is a measurement scale for eyesight)

                                                            Common usage of "catch 22" is, "no good choice", or
                                                            "it's bad if you do and it's bad if you don't"

                                                            Cheers,
                                                            Darren.


                                                            )-----Original Message-----
                                                            )From: Sacha Griffin
                                                            )
                                                            )I think its catch 20-20 when its really obvious and you can see it. ;)
                                                            )
                                                            )>From: Pat Swovelin
                                                            )>
                                                            )>> but beware of
                                                            )>> diffraction. A smaller sensor demands a smaller CoC, but it also
                                                            )>> suffers from diffraction quicker. So it's a bit of a catch 20-20
                                                            problem.
                                                            )>That's Catch 22...
                                                            )

                                                            No virus found in this outgoing message.
                                                            Checked by AVG Free Edition.
                                                            Version: 7.5.463 / Virus Database: 269.6.1/776 - Release Date: 25/04/2007
                                                            12:19
                                                          Your message has been successfully submitted and would be delivered to recipients shortly.