"waggy" <hon### [at] handbasketorg> wrote:
> Nice work!
>
> I've recently been doing some POV-Ray animation, as well. I use a different .pov
> file for each scene, and derive scene clocks from the global clock. I've also
> found it helpful to define a "transition" clock between scenes. It runs [-1,1],
> and both scene files on either side of the transition key frame (when the
> transition clock is zero) are loaded during the transition time period. The
> scene transition clock is used to make scene "transition in" and "transition
> out" effects.
>

Can't say I understand from your video just what is going on.

> I like to upload a lossless video. I don't know if mencoder supports it, but
> this seems to work well using ffmpeg (Linux version). The compression seems
> pretty good. It compressed 322MiB of gzipped PNG source frames to a 41 MiB
> lossless video. Youtube seems to handle this lossless encoding fine.
>
> ffmpeg -b 1000k -r 30 -f image2 -i FrameFolder/FrameName%04d.png \
>  -vcodec libx264 -pix_fmt yuv420p -preset veryslow \
>  -tune animation -qp 0 AnimName.mkv
>
> Unfortunately, few players can deal with the lossless H.264 encoding, so I
> experimented until I found a high-quality lossy combination that seems to play
> ok under stock Win7, Mac, and Linux. (Compression is similar to above.)
>
> ffmpeg -b 1000k -r 30 -f image2 -i FrameFolder/FrameName%04d.png \
>  -vcodec libx264 -pix_fmt yuv420p -preset veryslow \
>  -profile baseline -tune animation -crf 1 AnimName.mp4
>
> Here's what I'm working on. It currently has five short scenes.
>
> http://www.youtube.com/watch?v=eX3i1ioAvBA
>
> I still have a lot of work to do on it.

I encoded from png frames using mencoder.  The codec I used was x264, and the
command line was somewhat altered from a 2-step process where the second step
was merging audio with another tool that I couldn't find.  Fortunately, I wasn't
using audio, so it wasn't necessary.

my macros currently handle motion in four ways:
1) acceleration given Delta-V
2) acceleration given a constant of acceleration
3) steady motion given speed
4) steady motion given distance

They're probably not as elegant as Chris Colefax's spline macros.  It would be
easier, for instance, to orient a spacecraft along a spline which defines it's
path, and use another spline to define the velocity at which it traverses that
path, but I'm not good with splines.

Regards,
A.D.B.

Post a reply to this message