[Jmol-developers] [ jmol-Bugs-3156530 ] isosurface display

[SourceForge.net <no...@so...>]

Bugs item #3156530, was opened at 2011-01-12 14:00
Message generated for change (Comment added) made by hansonr
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=379133&aid=3156530&group_id=23629

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Applet
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Bob Hanson (hansonr)
Summary: isosurface display

Initial Comment:
Hello,
I'm having problems displaying the homo and lumo from .cube files on the JMol applet.  I have links below the applet that updates the displayed molecule so that the homo or lumo can be shown.  Sometimes it works correctly.  Other times only the positive or negative surface is displayed.  Sometimes no surface is displayed.  I am attaching screenshots of all three cases from the same molecule.  I am also attaching a cube file that I use to get the isosurface info for the homo.  Here is what the script looks like that updates the display of the molecule:

<script type="text/javascript">
          jmolInitialize("${jmol_url}");
          jmolApplet(700, "load ${homo_cube_url}");
          jmolSetLinkCssClass("jmol-link");
          jmolBr();
          jmolLink("isosurface off", "Clear");
          jmolLink("isosurface off; isosurface pos color red cutoff ${doc.results.homo_contour} ${homo_cube_url}; isosurface neg color blue cutoff -${doc.results.homo_conto
ur} ${homo_cube_url}", "HOMO");
          jmolLink("isosurface off; isosurface pos color red cutoff ${doc.results.lumo_contour} ${lumo_cube_url}; isosurface neg color blue cutoff -${doc.results.lumo_conto
ur} ${lumo_cube_url}", "LUMO");
        </script>


Craig


----------------------------------------------------------------------

>Comment By: Bob Hanson (hansonr)
Date: 2011-01-14 15:02

Message:
First, if you can figure out how to open the Java console (not the Jmol
console) you might find some sort of error there. 

Second -- could be the cutoffs. 

Third -- the randomness of this makes no sense. Not sure you mean
sometimes with one molecule, sometimes with another, or sometimes with the
same orbital and sometimes not, but with the same orbital. I can't imagine
what is going on there, unless it's something about memory -- that you are
not selecting out the specific subset of atoms or something.

If you could point me to the web site, that might be best. 

----------------------------------------------------------------------

Comment By: https://www.google.com/accounts ()
Date: 2011-01-14 09:48

Message:
Thanks for pointing out the SIGN keyword.  After switching to:

jmolLink("isosurface off; isosurface SIGN color red blue cutoff
${doc.results.homo_contour} ${homo_cube_url}", "HOMO");

The behavior changed.  Now I never get just a positive or negative
surface, instead it is sometimes displayed correctly, or not at all.  Just
a rough estimate, but hitting the HOMO link gives me about a 1 in 3 chance
of seeing the isosurface.  When I watch the console, it will indicate that
an isosurface was created even when it does not appear on the molecule:

isosurface1 created with cutoff=0.017 min=-0.116008 max=0.11920901; number
of isosurfaces = 1
isosurface full data range -1.0 to 1.0 with color scheme spanning -1.0 to
1.0



----------------------------------------------------------------------

Comment By: Bob Hanson (hansonr)
Date: 2011-01-12 17:54

Message:
No need to use two surfaces -- just use the SIGN keyword. This applies your
cutoff to both positive and negative cutoffs and gives you surfaces for
both:

isosurface SIGN yellow green cutoff 0.01 data/benzene-homo.cub.gz

----------------------------------------------------------------------

Comment By: https://www.google.com/accounts ()
Date: 2011-01-12 14:03

Message:
Oops.  I didn't get the other files attached that I meant to.  I'm not sure
how to upload the images that show the surface incorrectly displayed.  If
anyone needs them please let me know how to get them attached.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=379133&aid=3156530&group_id=23629