[Chimera-users] Smooth transitions in movies

[Tom Goddard]

Hi Mitalee,

   I added the "movie crossfade" command to Chimera last Friday.  It 
allows you to smoothly interpolate between two frames in a movie.  
Here's a video showing how to use it.

     http://www.cgl.ucsf.edu/chimera/videodoc/crossfade/

You need a recent Chimera daily build to use this.

  In short the way it works is you simply add a command like

     movie crossfade frames 25

to your movie recording Chimera command script where you want the 
transition to occur.  It doesn't show the transition when recording but 
it does appear in the movie file.

     Tom


> Hi Mitalee
>
>     This is a Chimera limitation.  Two or transparent models will not be
> shown correctly.  Transparency works correctly only when at most one
> model is transparent.  You have the volume data displayed in "solid"
> style which is a kind of transparent rendering, and you have a surface
> model which is transparent.  How defective the transparency looks
> depends on the order the models were opened in Chimera (or maybe it
> depends on which has higher model id number, don't recall).  In any
> case, both orders look seriously wrong.  If you reverse the order of
> opening your two models I think your surface will remain totally
> invisible until it becomes 0% transparent when it will suddenly appear.
>
>     This limitation is a rather fundamental design problem in Chimera.
> Chimera draws the models one at a time.  A model that is drawn after
> another model can only appear in front of the first model.  Chimera
> first draws all opaque models, then draws all transparent models.
>
>     If you look at the termite gut video I made
>
>       http://www.cgl.ucsf.edu/chimera/animations/movies/termitegut.mov
>
> where surfaces fade in you'll see at the start of the fade in, a black
> blob suddenly appears representing the object.  This is the same bad
> behavior you see in your case, but your surface is white (the background
> color) and is in front of the volume data making it look especially bad.
>
>     This problem has arisen many times in, especially in movie making
> where transparency is used to fade objects in and out.  I have an idea
> for a solution although it is not yet implemented.  But we just hired
> Daren Weber to work on animation improvements in Chimera and perhaps he
> can try this.  The idea is to enhance the movie command to add a
> "crossfade" option.  You would display the volume, then use a command
> like "movie crossfade 25" then show the opaque surface.  The command
> would take the image right before the command was issued and the next
> image and linearly blend them over 25 frames.  This would be useful for
> many fade scenarios.  It is limited in that it won't be able to have
> models moving and a fade occurring at the same time since that is not
> simply a linearly blending of two images.
>
>     I'll see if Darren is interested in trying to add this if you are
> interested.
>
>     If you do not want to wait for this, another option is to record your
> movie without the fade -- just have the surface suddenly appear.  Then
> drop the movie into an video editing program like Final Cut and add the
> crossfade with that software.
>
>       Tom
>    
>> Hi Tom,
>>
>> I was using the surft command, trying to make a segment appear inside of a
>> cmap file. Basically, I have a cmap file open and visible, and I want to
>> show the segmented cell appear slowly. My problem is that when I set the
>> transparency of the segmented cell to 100 (i.e. surft 100 #0), it doesn't
>> exactly become transparent, or what I thought would be transparent. When I
>> enter this command, it makes the entire image transparent at that point
>> (so the shape of the segmented cell is preserved and matches the
>> background. I've attached an image of what I got as a result. Is there a
>> way to make preserve the cmap image and just make the segmented cell
>> overlay onto it without completely "removing" the section where the
>> segmented cell will appear?
>>
>> Thanks,
>> Mitalee
>>      
> _______________________________________________
> Chimera-users mailing list
> Chimera-users at cgl.ucsf.edu
> http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
>