[Chimera-users] Movie Question: EM Densities about robbon structures

[Thomas Goddard]

Hi Jared,

  If you use Chimera commands that specify how many frames to rotate,
shift, scale, fade over, ... then it does not matter how fast or slow
your machine is.  Each frame is recorded only after it is drawn, and
none will be skipped.  Only when you move a model by hand with the mouse
will it matter how fast the machine is.

  Here are some problems with smooth transitions I have observed.
Fading out a PDB model in stick mode produced all kinds of weird
flickering of the bonds.  This I believe is a bug in Chimera
transparency of bonds and I plan on submitting a bug report for that.
Smoothly changing the contour level does not produce a very smooth
change in surface appearance.  Little spots pop up or disappear at
discrete contour levels.  This is because the contouring algorithm
(marching cubes) changes the surface triangulation in discrete ways
when the contour level passes through a grid point data value.  I
don't know of any remedy although using surface smoothing (volume
dialog under surface and mesh options) may help.

  Let me know what problems you encounter.  I am interested in
improving the Chimera movie capabilities.

	Tom


> Date: Mon, 18 Sep 2006 13:11:39 -0500
> From: "Jared Godar" <jared.godar at vanderbilt.edu>
> Sender: jared.godar at gmail.com
> To: "Thomas Goddard" <goddard at cgl.ucsf.edu>
> Subject: Re: [Chimera-users] Movie Question: EM Densities about robbon structures
> 
> I did, thank you much...  I have successfully installed the additional
> extensions and am attempting to incorporate them into a .CMD script for my
> movie.  So far that is going fairly smoothly, with the exception of frames
> involving fading in/out of structures are taking significantly longer than
> others resulting in a jumpy movie.  I'm going to attempt to try it on a
> fastest machine and see if that helps.
> 
> The independent rotation is working, the fade in/out of specific models is
> good, and I'm going to try the contour transition next.
> 
> Thanks again,
> 
> Jared