[Chimera-users] Using Per-Frame commands to recolor structures with attribute data is very slow

[David Slochower]

Thanks Eric. I was able to define all the attributes at once, as you suggested, and this does make things a lot faster, although recoloring still slows down playback. Thanks for the tip!

> On Nov 27, 2018, at 10:43 AM, Eric Pettersen <pett at cgl.ucsf.edu> wrote:
> 
> Hi David,
> 	An approach to make the playback faster is to define all your frame-attr data at once by make one big defattr file that is basically all your attribute files concatenated together, but with the attribute names changed to append the corresponding frame number, i.e. a file that defines myattr1, myattr2, etc.  Then in your perframe command, use $1 to add the frame number to the end of the attr name.
> 	So while the defattr part of your process will be considerably slower, the playback will be faster.  This approach has the additional advantage that if you save a session, all the attribute definitions will be preserved and you won’t need to redefine them.
> 
> —Eric
> 
> 	Eric Pettersen
> 	UCSF Computer Graphics Lab
> 
> 
> On Nov 26, 2018, at 8:49 PM, David Slochower <slochower at gmail.com <mailto:slochower at gmail.com>> wrote:
>> 
>> Hi,
>> 
>> As a follow up to this thread (http://plato.cgl.ucsf.edu/pipermail/chimera-users/2009-March/003712.html <http://plato.cgl.ucsf.edu/pipermail/chimera-users/2009-March/003712.html>) from 2009, I’d like to color a model by a custom attribute that changes from frame to frame. Using the workaround mentioned in that thread, we are able to use Per-Frame commands to load a new attribute file each frame, and then color the structure by the attribute, but this awfully slow (probably over 1 second per frame). 
>> 
>> Is there a way to cache the attribute data? Or a way to improve the performance of defining the attributes and re-coloring the structure? I know that I could use a Python script to load each coordinate frame with its associated attribute data, save images, and then stitch them together for a movie, but the current goal is for data exploration.
>> _______________________________________________
>> Chimera-users mailing list: Chimera-users at cgl.ucsf.edu <mailto:Chimera-users at cgl.ucsf.edu>
>> Manage subscription: http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users <http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users>
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://plato.cgl.ucsf.edu/pipermail/chimera-users/attachments/20181127/b07222c8/attachment.html>