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

Re: Wide screen cinema vs fulldome

Expand Messages
  • Paul
    ... Quite right, typo on my part. Of course I suggest that very little content that will ever be shown in stereo will be 8K, in the near future. Indeed the
    Message 1 of 10 , May 18, 2013
    View Source
    • 0 Attachment
      > A couple of clarifications. Macau Planetarium has an 8K stereo system

      Quite right, typo on my part.

      Of course I suggest that very little content that will ever be shown in stereo will be 8K, in the near future. Indeed the video sequences in Space Junk probably only exist as 2K wide resolution (?). I suspect they were filmed before 4K cameras were available. Indeed nothing in the whole production looked anywhere near 8K, more like 2K .... compared to something like DSA which I suspect was 4K but at least looked it.

      > As for Space Junk, Sky-Skan has not had a hand in supplying it, nor transforming it from flat screen to the dome and therefore, the transform is not to Sky-Skan's specifications.

      Accepted. But are you suggesting SkySkan does have a "specification" for transforming wide angle footage to fulldome?

      > I have seen this show on a flat screen and can say that 3D was clear and comfortable. The issues you are expressing, may be purely an artifact of the transform.

      But "purely an artefact" is pretty important. If standard content is stretched, as this is, to fill fulldome then bad things have to happen, both to the image and the stereo effect. The planets are a weird oblong shape, there was no detail towards the edges (by that I mean about 1/3 of dome radius) and the stereo failed totally ... by failed I mean was objectionable to watch.

      And then there are the usual problems of stereo content created for a smaller screen being scaled up to larger screens meaning that distant objects require the human eye to diverge, not something they are supposed to do.

      The thing I don't get is how/why the standards of the people producing this have been compromised. Did they ever see it in a stereo3D dome? If they did, had they seen any stereo3D fulldome that was done correctly for comparison? Are they listening in order to give some insight.
    • Ed Lantz
      Hi All, I d like to chime in on the issue of mapping Giant Screen films to domes. A small group of us have been working diligently to develop version 1.1 of
      Message 2 of 10 , May 18, 2013
      View Source
      • 0 Attachment
        Hi All,

        I'd like to chime in on the issue of mapping Giant Screen films to domes. A
        small group of us have been working diligently to develop version 1.1 of
        "Digital Immersive Giant Screen Specifications" for distribution of GS films
        to digital domes and flat screens (a special Task Force within the Giant
        Screen Cinema Association's Technical Committee). This will open the door
        for more and more GS films to enter the digital planetarium marketplace.

        DIGSS 1.1 Specs are open for comment here:
        http://bit.ly/18ULH45

        Note that three of us on the GSCA's Task Force are also heavily invested in
        the fulldome market (Martin Howe, Michael Daut and myself). We are doing
        our best to assure a clear distribution pathway for GS films into the
        fulldome market. Of course, distributors are not waiting for the DIGSS
        specification - a number of GS films have already been reformatted for
        fulldome distribution. One GS distributor expressed to me that the fulldome
        market has proven to be vital to their business model.

        As currently envisioned, DIGSS will clarify the process of packaging and
        distributing digital masters, and will set both minimal and aspirational
        goals for technical quality of digital pixel (and audio) delivery in the
        dome. Methods for reformatting from flat-screen to dome are currently seen
        as aesthetic processes, possibly varying on a scene-to-scene basis. GSCA
        has taken no official position on conversion methods between flat screen and
        dome as far as I know, leaving this issue to debate and personal opinions.
        Here is my opinion:

        Clearly, the most optimal way to shoot for fulldome in 2D is by collecting a
        field-of-view that matches the hemispheric field of view inherent in the
        dome master format. Without accurate geometric mapping, much of the value
        of fulldome (such as monoscopic depth cues including parallax) can be lost,
        resulting in sub-optimal use of the fulldome medium. Proper parallax in a
        fulldome production creates a near-autostereoscopic effect, leaving
        audiences asking how we created "3D without the glasses." IMAX and other GS
        films have sub-optimally exhibited flat-screen films in IMAX Dome format for
        years. In practice, similar liberties are routinely taken with fulldome CG
        rendered films on a scene-by-scene basis. I do believe that the freedom of
        "artistic license" should be left to producer's in such matters, and even
        blindly converted films can have those "lucky moments" where the scene
        appears to have been shot with a hemispheric lens.

        At the same time, we really need to encourage filmmakers distributing to
        domes to do their best, within reason, to exploit the unique advantages of
        the dome. Just as with several poorly converted 3D films (such as Prime
        Focus' rush job on Clash of the Titans which caused an uproar in the 3D film
        industry), poorly repurposed GS films could taint the fulldome experience
        for visitors and have a negative impact on the industry. Perhaps we should
        create guidelines (instead of "standards" or "specifications") for
        converting flatscreen films to fulldome. Ideally, GS filmmakers would plan
        for fulldome release during the film budgeting process (many of them already
        do!) - however they need a model or template for how to do this, and it
        would be very beneficial for us to tell them how to do it right. The
        aesthetics of flatscreen-to-dome conversion is a worthy topic for discussion
        in our community, and I would love to see studies and tests that compare
        various conversion techniques.

        On the other hand, I believe that fulldome 3D is a special case that does
        lend itself to standards/specifications. The eye can only tolerate so much
        horizontal (and vertical) disparity before it is difficult (or impossible)
        for the brain to converge a stereoscopic image. Presenting a stereo image
        to the brain that it is capable of fusing into a 3D scene is not only an
        issue of aesthetics or scene composition, it is a technical challenge that
        can be approached with the rigor of engineering and image science.

        Here are some ideas for further research:

        1) Dome versus Flatscreen Viewing. Let us assume that we map a 3D film
        to a dome without expanding its field-of-view. In other words, we project
        it as a flatscreen image covering only the front portion of the dome. How
        does this differ from flatscreen viewing? How does the stereo effect
        degrade as we move off axis into aisle or front row seats (versus
        degradation in a flatscreen theater)?

        2) Expanded FOV. As we expand the film's field-of-view on the dome -
        stretching it to fill more of the dome - how does the stereo effect degrade?

        3) Optimization. For geometrically expanded flat screen films, is there
        an algorithm that results in less distortion and more comfortable viewing
        over a wider range of seating?

        4) Guidelines. What are comfortable guidelines for 3D flat-to-fulldome
        conversion that producers and distributors can "blindly" follow and be
        assured of good results for most any 3D film? Or is a 3D flatscreen film
        best left for the flat screen?

        Inquiring minds want to know... Hint: this issue is worthy of research $
        investment from public sources with a vested interest in fulldomes as
        informal science education venues.

        Best,

        Ed Lantz
        ed@...
        www.vorteximmersion.com

        ._,___
      • petepopp
        Hi, There are omnidirectional 3D planetarium programs designed and produced originally for the dome. - Life of Trees fulldome 3D - Kaluoka hina the Enchanted
        Message 3 of 10 , May 19, 2013
        View Source
        • 0 Attachment
          Hi,

          There are omnidirectional 3D planetarium programs designed and produced originally for the dome.

          - Life of Trees fulldome 3D
          - Kaluoka' hina the Enchanted Reef fulldome 3D
          www.reef-distribution.com
          www.lifeoftrees.com

          Peter Popp
          Producer/director
          Softmachine
          www.softmachine.de
        • Roberto Ziche
          Another one is Astronaut 3D (http://astronaut-3d.com/), using an open source shader to render steroscopic dome images. More info here: http://bit.ly/15ZgRbV (
          Message 4 of 10 , May 19, 2013
          View Source
          • 0 Attachment
            Another one is Astronaut 3D (http://astronaut-3d.com/), using an open source shader to render steroscopic dome images.

            More info here: http://bit.ly/15ZgRbV
            ( http://fulldome.ning.com/forum/topics/stereoscopic-domemaster-images )


            Rob
          • Philip Groce
            Ed: A well reasoned response. Those GS films that have been repurposed for dome projection are for the most part a very poor second to showing these same
            Message 5 of 10 , May 19, 2013
            View Source
            • 0 Attachment
              Ed:

              A well reasoned response. Those GS films that have been repurposed
              for dome projection are for the most part a very poor second to
              showing these same films on a flat screen or in a truncated dome
              format. Nothing angered me more then when I saw "Sea Monsters" on a
              dome, so severely cropped that it would have disappointed even the
              most ardent fan of repurposing GS films. When the center of the image
              becomes the center of the dome, I want to just leave the theater.
              Now I realize that my reaction may not be that of a standard
              planetarium audience. Many audiences may be happy with this severely
              compromised image because Its big, and its thrilling, and the audience
              has no other way of seeing this film. Fulldome theaters are so
              desperate for content that they will show anything as long as it
              brings in audiences.

              However, as you rightly pointed out, we can do better than that.

              To me the simplest and most cost-effective answer for repurposing GS
              films is to show it as a truncated or partial dome image. Here the
              geometry of the dome does not severely compromise the image content
              with only the corners cropped. The truncated image doesn't have the
              severe cross-bounce issue of fulldome, so immediately, the image has
              greater contrast, greater color saturation and appears sharper. Using
              a fulldome system to show a partial dome image is a waste. For a
              fraction of the cost of a large dome fulldome system, one can install
              a dedicated single-projector truncated system. The resulting image
              will be brighter, sharper, and have better contrast that the
              equivalent resolution fulldome system.

              So let's "render on to Caesar that which is Caesar's." Let's let
              fulldome, be fulldome, giant screen be giant screen and planetariums
              be planetariums. Trying to do it all with a single projection system
              sounds on the surface a perfect design goal, but in reality, it is a
              fool's errand.

              Phil


              Philip Groce
              Helping Planetariums Succeed, LLC
              619 Orange Street
              Macon, GA 31201
              Voice: 478-750-7870
              Fax: 478-750-7826
              cell: 478-719-0661
              email work: hps4075@...
              email home: groce@...
            • dd4skyart
               A film shot in stereo should be shown so as to reproduce for the audience as well as possible the camera field of view. Trying to force a conventionally
              Message 6 of 10 , May 19, 2013
              View Source
              • 0 Attachment
                 A film shot in stereo should be shown so as to reproduce for the audience as well as possible the camera field of view. Trying to 'force' a conventionally filmed scene to fill too much of the dome is a mistake. Unless it such a scene is projected to imitate a conventional theater it will be terribly distorted, and how well a limited FOV projection would compare with a conventional projection setup is open to question.

                In general I don't like what fulldome 3D imposes on the choices of scenes we might want to use in shows. There seems to be an 'obligation' to have a foreground object or some other thing that shows the depth, partly to justify the use of the format.

                 In my experience this generally meant having a spacecraft always in the foreground, with the target planet, background stars, etc. all being at 'infinity', defined by the dome radius and viewer position. 
                 Rendering the foreground twice was done with twin cameras perpendicularly aligned to the subject of interest. The stereo breaks down at and beyond 45 degrees from your camera pointing direction, but is maintained along the forward vertical axis to and beyond the zenith as far as one can tilt ones view upwards. This means limiting any foreground spacecraft, including their long antennas, booms etc to that 'zone'. Camera moves were designed accordingly.

                 Depth perception in digitally modeled planetary landscapes that I moved the camera across looked better all around than I expected based on the above considerations. But there is no stereo information near each cameras view of the other camera, 90 degrees left and right.
                The content was produced to align its horizon with the true horizon when projected in a 22.5 degree tilted dome. I think tilted domes are where stereo should be considered, where the audience is looking more or less toward the same center of attention.

                The initial 3D movie wave of the 50s fumbled with cheap 'look, I'm in 3D' gimmickry with few notable works left behind. Now we are in a robust revival of cinematic 3D, with varying results. Clever use of depth perception can do amazing things and it will be intriguing to see what emerges as more try their hands at it on the dome.
                I can see especially great potential in 3D  for teaching concepts using diagrammatic stereo computer simulations, even of a vector graphic nature, of dynamic processes like convection, air movements, galaxy evolution and such, shown as bright centrally placed content in otherwise black backgrounds.


                  Don Davis
              • aaronbradbury
                ... A film shot in stereo should be shown so as to reproduce for the audience as well as possible the camera field of view. Trying to force a conventionally
                Message 7 of 10 , May 20, 2013
                View Source
                • 0 Attachment
                  --- In fulldome@yahoogroups.com, "dd4skyart" wrote:

                  A film shot in stereo should be shown so as to reproduce for the
                  audience as well as possible the camera field of view. Trying to 'force'
                  a conventionally filmed scene to fill too much of the dome is a mistake.
                  Unless it such a scene is projected to imitate a conventional theater it
                  will be terribly distorted, and how well a limited FOV projection would
                  compare with a conventional projection setup is open to question.

                  In general I don't like what fulldome 3D imposes on the choices of
                  scenes we might want to use in shows. There seems to be an 'obligation'
                  to have a foreground object or some other thing that shows the depth,
                  partly to justify the use of the format.

                  In my experience this generally meant having a spacecraft always in
                  the foreground, with the target planet, background stars, etc. all being
                  at 'infinity', defined by the dome radius and viewer position.

                  Rendering the foreground twice was done with twin cameras
                  perpendicularly aligned to the subject of interest. The stereo breaks
                  down at and beyond 45 degrees from your camera pointing direction, but
                  is maintained along the forward vertical axis to and beyond the zenith
                  as far as one can tilt ones view upwards. This means limiting any
                  foreground spacecraft, including their long antennas, booms etc to that
                  'zone'. Camera moves were designed accordingly.

                  Depth perception in digitally modeled planetary landscapes that I
                  moved the camera across looked better all around than I expected based
                  on the above considerations. But there is no stereo information near
                  each cameras view of the other camera, 90 degrees left and right.
                  > The content was produced to align its horizon with the true horizon
                  when projected in a 22.5 degree tilted dome. I think tilted domes are
                  where stereo should be considered, where the audience is looking more or
                  less toward the same center of attention.

                  The initial 3D movie wave of the 50s fumbled with cheap 'look, I'm in
                  3D' gimmickry with few notable works left behind. Now we are in a robust
                  revival of cinematic 3D, with varying results. Clever use of depth
                  perception can do amazing things and it will be intriguing to see what
                  emerges as more try their hands at it on the dome.

                  I can see especially great potential in 3D for teaching concepts
                  using diagrammatic stereo computer simulations, even of a vector graphic
                  nature, of dynamic processes like convection, air movements, galaxy
                  evolution and such, shown as bright centrally placed content in
                  otherwise black backgrounds.

                  Don Davis

                  ===

                  Regarding wide-screen to fulldome conversion:
                  I agree that conversions of giant screen 3D footage to fulldome 3D is
                  problematic.
                  4) Guidelines. What are comfortable guidelines for 3D
                  flat-to-fulldomeconversion that producers and distributors can "blindly"
                  follow and beassured of good results for most any 3D film? Or is a 3D
                  flatscreen filmbest left for the flat screen? (Ed Lantz)
                  Fulldome 3D is a big leap from flat-screen 3D, and having worked in both
                  formats I would probably not enjoy the task of trying to make flatscreen
                  work in fulldome. I would also say that it's not out of the question,
                  maybe the tools to make this possible just haven't been developed yet. I
                  haven't seen Space Junk, so I can't comment from personal experience,
                  but it sounds like Paul had a pretty bad experience, and where diverging
                  eyes are concerned, this doesn't have to be to extreme to give people
                  headaches. I guess we're stuck in that awkward place at the moment where
                  people are trying to push the boundaries and figure out the ways to make
                  it work and the results are having a negative impact, thus hindering
                  progression. It's a bit of a catch-22, but it may only takes a few
                  really good moments and clever ideas to really get the ball rolling.
                  Regarding general fulldome 3D:
                  >In general I don't like what fulldome 3D imposes on the choices of
                  scenes wemight want to use in shows. There seems to be an 'obligation'
                  to have aforeground object or some other thing that shows the depth,
                  partly to justifythe use of the format. (Don Davis)
                  Fulldome 3D is definitely a different process to fulldome 2D. There are
                  many other forces at work and many other options available. I prefer to
                  think of these as exciting challenges rather than imposed obstructions.
                  At NSC Creative, we pick our moments for 3D, so I wouldn't say that we
                  are pushing the stereoscopic effects in every shot. I think it helps to
                  have a mixture of wide open shots and some of the more intimate 3D
                  moments. There is even a time and a place for the poking-the-audience
                  effect if the motivation is right.
                  Interestingly, we find that a lot of our ideas for 3D compositions are
                  actually similar to that of those which aren't in 3D. When producing We
                  Are Aliens in 3D, we thought about it stereoscopically right from the
                  script through to the lighting and post effects. Considering this, we
                  still needed to be sure we were making the best possible show for a 2D
                  audience (the majority), but we made very few compromises between the
                  two.
                  >And still, there no real library 3-D fulldome content to show.
                  Untilthere are more 3-D (stereo) fulldome theaters, the content is
                  simplytoo expensive to produce. Supporters of 3-D Fulldome point out
                  thatthe same was true for the first few fulldome theaters. However, the
                  cost/benefit ratio is a much steeper climb for 3-D fulldome. 3-D
                  sounds great as a PR device, but, in reality, I fear it is an
                  expensive dead-end.(Philip Groce)
                  I describe my first experience of fulldome 3D at IPS in Chicago as a
                  spiritual one. It has been a driving force behind my own personal work
                  and the 3D work we do at NSC Creative. We've spent a huge amount of time
                  and money on figuring it all out, working with people from all over the
                  world. It's is a big mountain to climb in terms of challenges, but the
                  pay-off for me is worth the sweat.
                  Clever use of depth perception can doamazing things and it will be
                  intriguing to see what emerges as more try theirhands at it on the dome.
                  (Don Davis)
                  I am very excited about this. With more 3D domes popping up and more
                  shows being created in 3D, fulldome 3D will hopefully continue to
                  develop beyond it's infancy into something very special.
                  Regards,
                  Aaron Bradbury

                  CG Supervisor - NSC creative
                  National Space Centre, Exploration Drive, Leicester, LE4 5NS, UK
                  Tel: +44 (0) 116 2582151
                  http://www.NSCcreative.com/ <http://www.nsccreative.com/>
                Your message has been successfully submitted and would be delivered to recipients shortly.