> Jan wrote:
>
>> I want to compare matching surfaces and as a sandbox I used the three
>> MODEL DNA binding protein structure 1lcd
> [snip]
>
>> but not as expected, the dots surface persist while the rest of the
>> rendering changes with frame.
>
> That sounds like a bug. The dots rendering code is not respecting the
> currently displayed model.
The dot surface code now works properly with multiple models(frames) in
Jmol 10pre11n.
Previously this code was not differentiating between atoms in different
models(frames). This bug also affected the within({distance}, {set}) atom
expression operator.
Performance tech note
---------------------
While working on this I observed some performance problems when working
with macromolecules. Jan specifically asked about 1LCD. This has 3 models,
each of which includes 1100 atoms (not particularly large).
on a 2.8 Ghz Pentium ...
load samples/pdb/1LCD.pdb
select *
set solvent off; dots on # VDW surface takes 1.2 seconds
dots off
set solvent on; dots on # Solvent Accessible Surface takes 12.2 seconds
In addition, the time to render a single frame to the screen was about
doubled when the SAS was displayed. This is because the current code is
doing AxisAngle calculations during the rendering process to paint the
saddle shapes formed between two adjacent spheres.
I have no plans to work on this performance issue until we address true
surfaces ... after the official v10 release.
Miguel
calculating the VDW surface (solvent off) takes 1.1 seconds
calculate
-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=4721&alloc_id=10040&op=click
_______________________________________________
Jmol-developers mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/jmol-developers
