> Jmol-b6 has to option to draw more than one unit cell... but say, e.g.,
> a  2x2x2 unit cell... would that be difficult to implement?

I am not sure.
The graphics portion certainly would not be difficult.
You said that it was difficult for you to understand Fabian's code. Since
I don't understand the complexities of the geometry, if it was difficult
for you then it will be almost impossible for me.

As I understand it, the atoms in the other cells are reflections and
translations of the atoms in the unitcell. (I think that Peter previously
said that there were 243(?) types of transformations.) In any case, for a
given unitcell we should be able to construct transform matrices to
generate the siblings.

Q: Egon, can you do the calculations (or explain how to do the
calculations) to transform the atoms from the unit cell into the
neighboring cells?


I wonder what is the right way to represent these extra cells/atoms in
memory.

Two thoughts come to mind:

1. Do not create new atoms for the other cells
----------------------------------------------
We should be able to render the other cells by simply applying these
different transformation matrices to the unitcell.

This has several benefits:
 - no additional data structures to represent atoms.
 - any coloring/animation/vibration of the unitcell atoms
   would automatically be propogated to the 7 sibling cells.
 - we could easily turn the other cells on/off with just a checkbox
 - it would be very cool

Major disadvantage:
 - less control over the rendering in the other cells.

2. Actually create new atoms for the other cells
------------------------------------------------
The other option is to apply these matrices to the unitcell atoms and
actually generate atoms for the other cells.

The major disadvantage is that we would have to duplicate all the atom
data structures. These things are small, so that is not really a problem.

I think the bigger problem is that script writers would have to manage all
those atoms.

But that is also an advantage. A script writer could control the rendering
in each cell independently. For example, the atoms in each cell could be
rendered in a different color ... to help with visualization.

If we did this, I think that we would say that each cell was a 'model',
using the .pdb terminology ... because there is already good scripting
support for selecting and displaying individual models.



In writing this, I have almost convinced myself that option #2 is better.


What do you folks think?


Miguel





-------------------------------------------------------
The SF.Net email is sponsored by EclipseCon 2004
Premiere Conference on Open Tools Development and Integration
See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
http://www.eclipsecon.org/osdn
_______________________________________________
Jmol-users mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/jmol-users

Reply via email to