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

Re: [fulldome] Some initial Full-Dome video thoughts

Expand Messages
  • J. Scott Miller
    ... So, what you are saying is that rather than reproduce a star field that shows the subtle but vivid colors that stars show in the real sky, one should
    Message 1 of 7 , Oct 1, 2000
      Don Davis wrote:
      >
      > Full Dome video is the beginning of the golden age of domed theaters.
      > The ability to readily create an immersive enviornment will probably
      > result in the use of such domed theaters for many presentations unrelated
      > to astronomy so the design of new facilities should be free of the
      > constraints of earlier planetaria.
      > Specifically, since high def video can produce a reasonably good
      > starfield, the electro-mechanical star projectors are becoming outmoded,
      > unless having a starfield of the specific quality these machines can
      > deliver is a high priority. These machines are, in my view, a greater
      > libility than an asset because they deprive the audience of the best seats
      > in the house, those in the center of the dome where the viewing geometry is
      > optimised. Many of these machines also block much of the sky from view in
      > the seats closest to the ideal viewing geometry. No matter how fine the
      > stars projected with such machines appear, all you basically get for this
      > very expensive optical projector is one star field.

      So, what you are saying is that rather than reproduce a star field that shows
      the subtle but vivid colors that stars show in the real sky, one should simply
      settle for a mediocre star field, a star field where stars look like small
      snowballs instead of points of light.

      A better solution would be to keep the star ball but put it on an elevator.
      Remove it when it isn't necessary for programs in which astronomy is not the
      purpose, or even during programs of an astronomy nature in which the stars are
      acting like background anyway.

      > A tilted dome. perhaps 22.5 degrees, with the seats aligned toward the
      > front of the theater seems ideal, but I would remove the seats too far
      > removed from the center, with quality of presentation the goal and not
      > simply quantity of seats.

      Until quality is important to the bean counters and not quantity, I would say
      this is a pipe dream as well.

      > In old Planetarium designs the audiences were asked to accept the cove
      > line as a horizon, despite the resulting perspective of being in a shallow
      > pit. A tilted dome gives the viewer a chance to see an eye level horizon,
      > providing a more natural visual experience. The theme park rides I know of
      > using dome projection are generally highly tilted, so audiences look
      > forward as much as up.

      I been in both and from my own experience, each has advantages in its own use.
      But there are probably few audiences that want to feel like they are looking
      forward when they have come to see the stars. A slight tilt is fine (ours
      will be about 12 to 15 degrees) but I am glad we didn't opt for a dome tilt
      more than that.

      > Although useful as an aquisition medium, the days of film being used in
      > Planetarium projection are passing. I once had high hopes for Omnimax, but
      > the medium was too ungainly and expensive for general use. The better
      > efforts using fisheye film technology should be scanned and shown in
      > digital formats, they may well appear superior to the images projected on
      > film, especially if transferred at 30 FPS.

      Sorry, facts are facts. You don't get the resolution in video projection that
      you get in film. You can kid yourself into believing it, but when all is said
      and done, there is no comparison.

      > Now people can create images filling a dome using affordable home
      > computers and animation software. The days of machine shops under
      > Planetaria custom building projectors for each show are giving way to
      > artists and animators, on site and contracted, using software like Electric
      > Image, Lightwave, and the various Strata and 3D Studio incarnations to fill
      > the dome with digital enviornments.
      >
      > Don Davis

      And all have their place. But none are replacement.

      --
      J. Scott Miller, Program Coordinator Scott.Miller@...
      Gheens Science Hall and Rauch Planetarium
      http://www.louisville.edu/planetarium
      University of Louisville
    • Don Davis
      ... I believe the best all dome video can be good enough to take the place of many of the traditional planetarium projectors, although the electro-mechanical
      Message 2 of 7 , Oct 2, 2000
        J. Scott Miller wrote:

        >
        >So, what you are saying is that rather than reproduce a star field that shows
        >the subtle but vivid colors that stars show in the real sky, one should simply
        >settle for a mediocre star field, a star field where stars look like small
        >snowballs instead of points of light.

        I believe the best all dome video can be good enough to take the place of
        many of the traditional planetarium projectors, although the
        electro-mechanical star projectors with their decades of refinement
        obviously do one thing beautifully. High Def video is also capable of
        great subtlety as well as detail and colorful variety. I question depriving
        the audience of the chance of sitting at the center of the dome, where the
        view is least distorted, because of one specialized projector. Elevators
        also take up that ideal viewing position.

        >
        >But there are probably few audiences that want to feel like they are looking
        >forward when they have come to see the stars. A slight tilt is fine (ours
        >will be about 12 to 15 degrees) but I am glad we didn't opt for a dome tilt
        >more than that.

        Providing an eye level horizon line is important if a significant portion
        of the imagery attempts to reproduce an environment other than a starfield.
        At some point we are discussing preferences rather than quality or
        objective standards, and it's great we are living in a time where such
        choices are becoming increasingly available.


        >
        >Sorry, facts are facts. You don't get the resolution in video projection that
        >you get in film. You can kid yourself into believing it, but when all is said
        >and done, there is no comparison.

        Resolution is influenced by less obvious factors than the number of
        viewable line pairs on a piece of film. At the moment film is the
        unquestioned supreme acquisition medium for capturing real world moving
        scenery. I'm afraid that 70 mm film is necessary for this, which renders
        such footage all but impractical for people to obtain on demand. Once a
        sequence such as a time lapse is obtained it must be scanned (very
        expensive) and 'cleaned up' for high res video projection.

        I believe a filmed test subject scanned and shown on a well tuned multiple
        projection high def system would compare well with a 70mm film release
        print of the same subject projected with a fisheye lens. It would be
        interesting to see an actual comparison. Assuming the film camera has
        steady registration between frames, the video projection would be rock
        steady compared with whatever mechanical 'play' and focus variations a film
        projector might introduce. A 30 frame per second image has more 'temporal'
        resolution than film at 24 FPS, a fact that is especially noticeable when
        viewing dynamic subjects on a dome. In addition to visual performance
        issues the video image files would not gather scratches and other wear with
        each showing.

        Although importing wide angle real world moving images is still very
        difficult, computer generated images are where all dome video really takes
        off.

        Don Davis
      • Laurent Pellerin
        I have to agree with Scott. I spent some time two days ago at the FlorPlan Conference talking with the Spitz Representative about their system. Even he
        Message 3 of 7 , Oct 2, 2000
          I have to agree with Scott. I spent some time two days ago at the FlorPlan
          Conference talking with the Spitz Representative about their system. Even he
          agrees that video systems are a very long way from having the resolution
          necessary to replace the star projector. And even when they do, it will be
          even longer before the price will allow most domes to use them.

          The star field is the most important part of a planetarium. It is the only
          thing that makes a planetarium truly unique from all of the other
          entertainment and educational facilities out there. Therefore, a planetarium
          should get the best, most realistic star field that they can afford. Today,
          that still means star projectors, and will for years to come.

          But, the star field alone is not enough. Full dome video and all-skies
          certainly have their much needed place along side the star projector and slide
          projectors, etc., in domes that can afford them now, and in the future. But
          that doesn't mean that full dome video can replace the star projector any more
          than static all-skies could. They have the resolution, but not the movement,
          and video has the movement but not the resolution. So far, only the star
          projector has both.

          Laurent Pellerin
          Operations & Production
          Seminole Community College Planetarium
          Sanford, Florida
          pelleril@...



          ____________________________________________________________________
          Get your own FREE, personal Netscape WebMail account today at http://home.netscape.com/webmail
        • Ryan Wyatt
          In response to Don Davis s suggestion to remove star projectors from ... As a proud owner of a snowball projector (a.k.a. Digistar), I find the trade-off
          Message 4 of 7 , Oct 2, 2000
            In response to Don Davis's suggestion to remove star projectors from
            the optimal seating area in the planetarium, J. Scott Miller wrote:

            >So, what you are saying is that rather than reproduce a star field that shows
            >the subtle but vivid colors that stars show in the real sky, one should simply
            >settle for a mediocre star field, a star field where stars look like small
            >snowballs instead of points of light.

            As a proud owner of a snowball projector (a.k.a. Digistar), I find
            the trade-off perfectly acceptable: at least in terms of being able
            to put real-time 3D wireframe imagery on the dome. I can show a
            school group a model of the solar system (demonstrating clearly that
            the orbits lie in essentially the same plane), fly them around a 3D
            starfield, guide them through a cartoon of the Milky Way...

            What we show our visitors is *always* a caricature of the night sky.
            Even the much-vaunted Zeiss Mark 9.14159265 or whatever doesn't
            present a truly realistic image of the sky... The pinpoint stars
            look ab fab, but they stand out in too-high contrast against the
            background: it's not the real sky, and we shouldn't kid ourselves
            into thinking that way. (I'm not saying anyone on the list said
            that, BTW, but many of our colleagues seem to think that way.)

            That said, I haven't seen an acceptable starfield projected by
            full-dome video systems (I do have *some* standards, after all), with
            the possible exception of the Digital Galaxy at the Hayden.

            >A better solution would be to keep the star ball but put it on an elevator.
            >Remove it when it isn't necessary for programs in which astronomy is not the
            >purpose, or even during programs of an astronomy nature in which the stars are
            >acting like background anyway.

            That still doesn't address the fact that the star projector takes up
            the Best Seat in the House. When I visited the Hayden this past
            June, I had the chance, during a late-night session, to sit (well,
            actually, I laid flat on my back) in the center of the theater with
            the starball sunken and the SGI system rolling. Magnificent! That's
            where we need to be... Someday. Although I'm not sure when that day
            will come, it will come soon.

            As long as a system possesses real-time capability, it can perform
            the critical tasks to do a star show (my litmus test). But if you
            had to pre-animate star shows for a full-dome playback system, you'd
            never get anything else done.

            Denver will take the plunge with the first pitless planetarium. We
            can all see how well they sink or swim. :) I think the choice of
            laser video projection makes the most sense for reproducing a
            starfield, and their system will also work in real time. The best
            combination of choices for a planetarium application, IMO.

            Don wrote:

            >> In old Planetarium designs the audiences were asked to accept the cove
            >> line as a horizon, despite the resulting perspective of being in a shallow
            >> pit. A tilted dome gives the viewer a chance to see an eye level horizon,
            >> providing a more natural visual experience. The theme park rides I know of
            >> using dome projection are generally highly tilted, so audiences look
            >> forward as much as up.

            To which Scott (may I call you Scott?) replied:

            >I been in both and from my own experience, each has advantages in its own use.
            >But there are probably few audiences that want to feel like they are looking
            >forward when they have come to see the stars. A slight tilt is fine (ours
            >will be about 12 to 15 degrees) but I am glad we didn't opt for a dome tilt
            >more than that.

            I came on board at LodeStar after the theater design had been
            completed, and I initially dreaded working in the 25-degree tilt. I
            was particularly concerned about how the audience would adapt to the
            skewed horizon when doing a star show. My anecdotal perspective, six
            months down the line and 25,000 visitors later: dome tilt, no
            problem! I wouldn't do more than 30 degrees, but I'm quite happy
            with what I have to work with.

            And the full-dome video sequences definitely work better with the
            tilt. The only exceptions are the sequences that present a horizon
            parallel to the cove-line: then the audience feels like they're
            sunken into a bowl. But tilting the horizon in the animation (even
            less than the dome tilt) seems to remedy the problem.

            >> Although useful as an aquisition medium, the days of film being used in
            >> Planetarium projection are passing. I once had high hopes for Omnimax, but
            >> the medium was too ungainly and expensive for general use. The better
            >> efforts using fisheye film technology should be scanned and shown in
            >> digital formats, they may well appear superior to the images projected on
            >> film, especially if transferred at 30 FPS.

            I've had to deal with a 70-mm film projector in our SimEx theater,
            and let me tell you, I've never been a bigger fan of digital
            projection! Film is far, far, far too *analog* for me. Our
            projector requires more coddling and maintenance than any piece of
            machinery deserves to demand from a human. Plus, film has its own
            problems of uneven illumination and jitter.

            >Sorry, facts are facts. You don't get the resolution in video projection that
            >you get in film. You can kid yourself into believing it, but when all is said
            >and done, there is no comparison.

            And the fact is that full-dome video resolutions are already
            approaching or exceeding Imax(tm) resolutions on the dome. And the
            situation will only improve. Even if there's "no comparison" now
            (which I would dispute), there will be in the very near future.

            Can you say "Zulip"? Rhymes with tulip and julep... And although
            the demonstration we saw at IPS this year showed a technology in its
            infancy, it also showed tremendous promise. Mentally scaling up to
            the sizes and resolutions required for full-dome presentation, I
            convinced myself that (assuming they can correct the black-level
            issues and attain the resolutions they say they can) the technology
            should work beautifully for full-dome applications in a Denver-like
            time frame of two years.

            The greatest challenge with all these technologies right now is the
            price (n.b. that my favored technologies in this message have been
            SGIs and laser video projectors). But those prices will come down,
            in much the fashion Don referred to in his posting...

            >> Now people can create images filling a dome using affordable home
            >> computers and animation software. The days of machine shops under
            >> Planetaria custom building projectors for each show are giving way to
            >> artists and animators, on site and contracted, using software like Electric
            >> Image, Lightwave, and the various Strata and 3D Studio incarnations to fill
            >> the dome with digital enviornments.

            Amen!

            >And all have their place. But none are replacement.

            Not yet, but...


            Ryan.
          • J. Scott Miller
            ... I challenge such a comparison. Some of the best video star displays available today are the Digistars. I have sat in the domes that employ the latest and
            Message 5 of 7 , Oct 2, 2000
              Don Davis wrote:
              >
              > J. Scott Miller wrote:
              >
              > >
              > >So, what you are saying is that rather than reproduce a star field that
              > >shows the subtle but vivid colors that stars show in the real sky, one
              > >should simply settle for a mediocre star field, a star field where stars
              > >look like small snowballs instead of points of light.
              >
              > I believe the best all dome video can be good enough to take the place of
              > many of the traditional planetarium projectors, although the
              > electro-mechanical star projectors with their decades of refinement
              > obviously do one thing beautifully. High Def video is also capable of
              > great subtlety as well as detail and colorful variety. I question depriving
              > the audience of the chance of sitting at the center of the dome, where the
              > view is least distorted, because of one specialized projector. Elevators
              > also take up that ideal viewing position.

              I challenge such a comparison. Some of the best video star displays available
              today are the Digistars. I have sat in the domes that employ the latest and
              frankly remain unimpressed. One planetarium (the Astronaut Memorial
              Planetarium near Cape Canaveral) employed both a Digistar and a traditional
              Minolta projector in an attempt to have the best of both worlds. Not too many
              folks could afford to do that. If quality is the goal, as you imply, then
              high definition video will not be able to provide it.

              >
              > >
              > >But there are probably few audiences that want to feel like they are looking
              > >forward when they have come to see the stars. A slight tilt is fine (ours
              > >will be about 12 to 15 degrees) but I am glad we didn't opt for a dome tilt
              > >more than that.
              >
              > Providing an eye level horizon line is important if a significant portion
              > of the imagery attempts to reproduce an environment other than a starfield.
              > At some point we are discussing preferences rather than quality or
              > objective standards, and it's great we are living in a time where such
              > choices are becoming increasingly available.

              Well if quality and substandard quality are the choices, I think the
              discussion is already over. If the stars provide a back drop for an
              astronomy-related program, then there is some things that could be done better
              with interlaced video than say with all sky panorama systems. Again, the
              sacrifice is resolution. You still cannot beat the resolution of still
              photographic film with video. Even high def. video can't do it. If motion is
              necessary, then the resolution may not be as important (but, the Adler does
              something like this now and I was not impressed with their rendition of it in
              their tilted dome).

              >
              > >
              > >Sorry, facts are facts. You don't get the resolution in video projection
              > >that you get in film. You can kid yourself into believing it, but when all
              > >is said and done, there is no comparison.
              >
              > Resolution is influenced by less obvious factors than the number of
              > viewable line pairs on a piece of film. At the moment film is the
              > unquestioned supreme acquisition medium for capturing real world moving
              > scenery. I'm afraid that 70 mm film is necessary for this, which renders
              > such footage all but impractical for people to obtain on demand. Once a
              > sequence such as a time lapse is obtained it must be scanned (very
              > expensive) and 'cleaned up' for high res video projection.
              >
              > I believe a filmed test subject scanned and shown on a well tuned multiple
              > projection high def system would compare well with a 70mm film release
              > print of the same subject projected with a fisheye lens. It would be
              > interesting to see an actual comparison. Assuming the film camera has
              > steady registration between frames, the video projection would be rock
              > steady compared with whatever mechanical 'play' and focus variations a film
              > projector might introduce. A 30 frame per second image has more 'temporal'
              > resolution than film at 24 FPS, a fact that is especially noticeable when
              > viewing dynamic subjects on a dome. In addition to visual performance
              > issues the video image files would not gather scratches and other wear with
              > each showing.
              >
              > Although importing wide angle real world moving images is still very
              > difficult, computer generated images are where all dome video really takes
              > off.

              You are still hedging. I have seen both. Had to in order to be part of the
              team that put together our system. Today, the quality is not there for video
              versus film. I doubt it will ever be. But, if that is all that is offered,
              then the public will never know the difference, much to their loss.

              There is a place for video. It is not in the replacement of the star field
              produced by electromechanical means.

              --
              J. Scott Miller, Program Coordinator Scott.Miller@...
              Gheens Science Hall and Rauch Planetarium
              http://www.louisville.edu/planetarium
              University of Louisville
            • webb@adlernet.org
              ... Our system here at Adler projects real time imagery. Our resolution is limited (for now ) by what subject matter can be rendered in 1/30 second or less.
              Message 6 of 7 , Oct 2, 2000
                Scott.Miller@... writes:
                >If the stars provide a back drop for an astronomy-related program, then
                >there is some things that could be done better with interlaced video than
                >say with all sky panorama systems. Again, the sacrifice is resolution.
                >You still cannot beat the resolution of still photographic film with video.
                >Even high def. video can't do it. If motion is necessary, then the resolution
                >may not be as important (but, the Adler does something like this now and
                >I was not impressed with their rendition of it in their tilted dome).

                Our system here at Adler projects real time imagery. Our resolution is
                limited (for now ) by what subject matter can be rendered in 1/30 second
                or less. Don't confuse that with projector resolution or image quality on
                the screen.

                While I would agree that video on the dome is not up to film quality...
                yet... I would have to admit that it is a whole lot closer than I thought
                it would be. The Barco 1209 series projector is being discontinued at the
                end of this year. It will be replaced with a model called the BR909. The
                resolution of the BR909 is 3200x2560. If you put six of those projectors
                on the dome at full resolution you will have a full color star field that
                you can fly through. You and I could, probably, still tell the difference
                between that and a Zeiss. However, I'm willing to bet that your audience
                won't be commenting about the star field quality as they exit the theater.

                nightsight@... writes:
                >I spent some time two days ago at the FlorPlan Conference talking with the
                ***
                >Spitz Representative
                ***
                >about their system. Even he agrees that video systems are a very long way
                >from having the resolution necessary to replace the star projector. And
                >even when they do, it will be even longer before the price will allow most
                >domes to use them.

                No offense but I would guess that Spitz wants people to have full dome
                video AND a star ball in their theater. Ask a vendor that does not sell
                star projectors the same question and see what they say.
              Your message has been successfully submitted and would be delivered to recipients shortly.