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

Re: Klein bottle animation

Expand Messages
  • vortexswirling
    Hi Adrian, ... This is a very cool animation. I found a web page which shows LiveGraphics3D animated. It shows shome kaleidocycles.
    Message 1 of 8 , Jun 1, 2007
    View Source
    • 0 Attachment
      Hi Adrian,

      > I have just added the correct wrapping for the equations
      > that I use for the "usual" Klein bottle surface. Here is an
      > animation (1.1Mb)
      >
      > http://www.freewebtown.com/adrian/misc/anim_klein2_6363.gif

      This is a very cool animation.

      I found a web page which shows LiveGraphics3D animated. It shows shome
      kaleidocycles.

      http://www.ac-noumea.nc/maths/amc/polyhedr/kaleido1_.htm

      It would be interesting to see how they were constructed (I'm
      suspecting mathematica). LG3D then lets the object to be rotated while
      it is being animated.

      Roger
    • Adrian Rossiter
      Hi Roger ... I like these animations too. Also, it is nice to be able to add in the transparency now as it really helps show the surface. ... It is a good
      Message 2 of 8 , Jun 1, 2007
      View Source
      • 0 Attachment
        Hi Roger

        On Fri, 1 Jun 2007, vortexswirling wrote:
        >> http://www.freewebtown.com/adrian/misc/anim_klein2_6363.gif
        >
        > This is a very cool animation.

        I like these animations too. Also, it is nice to be able to add
        in the transparency now as it really helps show the surface.


        > I found a web page which shows LiveGraphics3D animated. It shows shome
        > kaleidocycles.
        >
        > http://www.ac-noumea.nc/maths/amc/polyhedr/kaleido1_.htm

        It is a good site. (I have written to Maurice before as he had
        a link to some of my jitterbug animations.)

        I don't know how the kaleidocycle animations were created, but
        I would guess it was Mathematica too.

        One of my main motivations for working on the Antiprism GUI
        is to have a framework for animated models like these. I often
        wish I could turn the GIF animations around. I dug out my
        proof-of-concept GUI code the other day and got it running
        so I might start adding to that bit at a time.

        Adrian.
        --
        Adrian Rossiter
        adrian@...
        Home: http://antiprism.com/adrian
      • vortexswirling
        Hi Adrian, ... I ve looked at the .m files and they are there is a lot of calculations going on. Each one would have to be custom made, if not from scratch,
        Message 3 of 8 , Jun 1, 2007
        View Source
        • 0 Attachment
          Hi Adrian,

          > > http://www.ac-noumea.nc/maths/amc/polyhedr/kaleido1_.htm
          >
          > It is a good site. (I have written to Maurice before as he had
          > a link to some of my jitterbug animations.)
          >
          > I don't know how the kaleidocycle animations were created, but
          > I would guess it was Mathematica too.

          I've looked at the .m files and they are there is a lot of
          calculations going on. Each one would have to be custom made, if not
          from scratch, from some generator.

          > One of my main motivations for working on the Antiprism GUI
          > is to have a framework for animated models like these. I often
          > wish I could turn the GIF animations around. I dug out my
          > proof-of-concept GUI code the other day and got it running
          > so I might start adding to that bit at a time.

          I can look at this problem unbiased by any kind of experience! Could
          not such a viewer be similar to antiview? It would read in a sequence
          of generated OFF files as frames.

          The animation loop would display each geom in a loop. Any rotation,
          translation, scale would be applied to all the geoms. If that was too
          computationally intensive, then it could apply the delta of the
          current position to the next displayed frame.

          Does this sound reasonable?

          Roger
        • Adrian Rossiter
          Hi Roger ... It could do that, but initially I would be considering animations where it is feasible to generate the model stages as needed. This is what I did
          Message 4 of 8 , Jun 2, 2007
          View Source
          • 0 Attachment
            Hi Roger

            On Sat, 2 Jun 2007, vortexswirling wrote:
            >> One of my main motivations for working on the Antiprism GUI
            >> is to have a framework for animated models like these. I often
            ...
            > I can look at this problem unbiased by any kind of experience! Could
            > not such a viewer be similar to antiview? It would read in a sequence
            > of generated OFF files as frames.

            It could do that, but initially I would be considering animations
            where it is feasible to generate the model stages as needed. This
            is what I did for Jitterbug GL

            http://www.freewebtown.com/adrian/opengl/jitterbug_gl.html

            It checks the time, works out how far through a cycle the time
            corresponds to, and then creates the model for that stage of
            the cycle.

            It would also be nice to save a sequence of models, in various
            formats, made in equal steps to automate making frames for
            animations or movies. (I had a record button on an earlier
            program that saved every single image that opengl rendered,
            but that was a bit of a disk filler!) This could also involve
            saving the frames as OFF files for later replaying as you
            suggest.

            Adrian.
            --
            Adrian Rossiter
            adrian@...
            Home: http://antiprism.com/adrian
          • vortexswirling
            Hi Adrian, I have the Jitterbug app and it is cool! With edges on there seem to be an anomoly once in a while. ... I have a sequence of 20 frames which
            Message 5 of 8 , Jun 2, 2007
            View Source
            • 0 Attachment
              Hi Adrian,

              I have the Jitterbug app and it is cool! With edges on there seem to be
              an anomoly once in a while.

              > It would also be nice to save a sequence of models, in various
              > formats, made in equal steps to automate making frames for
              > animations or movies. (I had a record button on an earlier
              > program that saved every single image that opengl rendered,
              > but that was a bit of a disk filler!) This could also involve
              > saving the frames as OFF files for later replaying as you
              > suggest.

              I have a sequence of 20 frames which generates a GIF of 165K. If I were
              to use 20 OFf files it would be 20 x 4.5K. If those 20 OFF files were
              in a zip file they'd take up practically no room! Sounds like a new
              file format to me!

              For some reason those 20 frames seem to be smooth enough for me. I
              suppose if I slowed it down it wouldn't be.

              This sounds like something where there'd be two threads of execution.
              There would be the frames that would just loop. The other would take
              the commands. I would just assume that all the frames were evenly
              spaced time-wise. Have you found situtations where that doesn't look
              right?

              Would this kind of thing look as good as those coming from povray?

              Roger
            • Adrian Rossiter
              Hi Roger ... I don t remember seeing any here. Was it edges joined at one vertex but pointing in the opposite direction of the other vertex? ... I currently do
              Message 6 of 8 , Jun 3, 2007
              View Source
              • 0 Attachment
                Hi Roger

                On Sun, 3 Jun 2007, vortexswirling wrote:
                > I have the Jitterbug app and it is cool! With edges on there seem to be
                > an anomoly once in a while.

                I don't remember seeing any here. Was it edges joined at one
                vertex but pointing in the opposite direction of the other
                vertex?


                > This sounds like something where there'd be two threads of execution.
                > There would be the frames that would just loop. The other would take
                > the commands.

                I currently do rendering as a function called from an idle
                loop, so it still runs while the controls are being used.
                The same thing could be done for "rendering" to a file.


                > I would just assume that all the frames were evenly
                > spaced time-wise. Have you found situtations where that doesn't look
                > right?

                I would think it was reasonable to assume. Although, with GIF
                animations the time that each frame is visible can be set
                individually.


                > Would this kind of thing look as good as those coming from povray?

                For making image frames the rendering stage could be to pass the
                current geometry/camera settings to off2pov and then run the output
                through POV-Ray. Otherwise, it could just save the OpenGL rendering
                which would look like it does on the screen. Another possibily is
                to render with OpenGL offscreen at a larger size and then reduce
                it to the image size, which may look a bit smoother.

                Adrian.
                --
                Adrian Rossiter
                adrian@...
                Home: http://antiprism.com/adrian
              • vortexswirling
                Hi Adrian, ... to be ... This is hard to describe. Whenever the model passes through an octahedron, there are a couple of free edges that flash on for an
                Message 7 of 8 , Jun 3, 2007
                View Source
                • 0 Attachment
                  Hi Adrian,

                  > > I have the Jitterbug app and it is cool! With edges on there seem
                  to be
                  > > an anomoly once in a while.
                  >
                  > I don't remember seeing any here. Was it edges joined at one
                  > vertex but pointing in the opposite direction of the other
                  > vertex?

                  This is hard to describe. Whenever the model passes through an
                  octahedron, there are a couple of free edges that flash on for an
                  instant at about 11 o'clock and 5 o'clock. I tried to pause it so I
                  could get a snapshot of it. But once I pause and restart it these edges
                  stop flashing! Who knows what that could be. I wouldn't worry about it!

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