On Nov 3, 2004, at 8:19 PM, Miguel wrote:
I noticed that when I use the viewer.evalString to show a different frame from within the AtomSetViewer I get the 'Script completed' message in the ScriptViewer.
Other people asked for it because it is 'chime compatible'
I don't really like that and was wondering whether I could make a more direct call (or via a [new] method in the JmolViewer) that will display a certain frame and start/stop a vibration,
That seems quite reasonable. If I don't send you something tomorrow then
send me a reminder.
Great.
As I mentioned earlier I was hoping to do a animation over only a subset of AtomSets (e.g., only the z-matrix ones from a geometry optimization). So I was planning to do my own type of animation by calling the newly created setFrame(atomSetIndex) method to show a frame, and loop that through the AtomSetIndex values that are part of the list selected. This would make it tricky for me to do the timing (fps type thing) that you allow one to set, so if it were possible to have a JmolViewer method, e.g., animateFrame(int atomSetIndex[]), to loop through the values in the atomSetIndex array, as opposed to over all the atomsets?
and possible change the amount of displacement in the vibration (they do go a bit far). I actually had somewhat expected the displacement to be reduced if I choose Display:Vector:Scale 0.5 (the vectors do indeed get smaller, but the vibration is equally large).
Well, the two used to be tied together. But people wanted to control the
displayed vectors and the vibration separately.
So there are now two separate script commands:
vector scale .5 vibration scale .5
The command that you chose from the menu is incorrect ... it is only scaling the vector without scaling the vibration ... that is a bug.
I assume it only requires one to put in the Jmol-resources.properties in the vectorScale*Script
vector scale *; vibration scale *
I hope that down the line the AtomSetChooser could grow into something that is similar to both the Vibrate... and Animate... Extras in v9, where I notice that no 'Script completed' is echoed...
Scripting was not used for this in v9.
On a side note. I noticed that when you select Start Vibration in the
Tools menu it always goes to the first frame, which may not have a
vibration in it. I would propose tonot go to frame 1 but only start the
vibration. I found where I would do that. (in the
Jmol-resources.properties).
Also the script 'First frequency' goes to frame 1, which does not make
sense. So maybe get rid of that too.
We need to think about this.
The changes/enhancements that you have made go well beyond the current design. We need to figure out the right way to do it without breaking existing applications and other file formats.
I realize that my observations are only from a very limited standpoint and not all kinds of files that people deal with. I initially interpreted the 'Start Vibration' to mean start vibrating the molecule I am currently looking at, so I was surprised when it changed the model before it turned on the vibration. In my case it was very noticeable since my first frame was not one with vectors on it. It is OK with me how it is now, so don't worry. I just thought I'd share my observation and suprise.
On the other hand, one could reason that it should be possible to go through each of the AtomSets and start at the first one that has vibrations.
Yes, for vibrations that probably makes sense.
The question is how that would be implemented, since the menu commands seem to mainly be defined as scripts (not all, I realize) so I assume we'd need to make a new 'action' in the Jmol.java. When you expose the methods in the viewer to do this, I could try to implement that, if you want me to. Also the next vibration may not necessary be in the next atomset/frame in the list, so the 'next vibration' may have to determine the real one to display...
But those frame methods are not mirrored in the JmolViewer itself. Should they be? (Notice how careful I have gotten about drilling further down using not explicitly public declared fields in JmolViewer? :-).
It would be helpful for you to think about what these methods would look
like.
Q: Do we only need a mechanism to inquire whether or not each Model/AtomSet has vibration vectors?
As far as I can tell that is indeed the case.
Ren�
------------------------------------------------------- This SF.Net email is sponsored by: Sybase ASE Linux Express Edition - download now for FREE LinuxWorld Reader's Choice Award Winner for best database on Linux. http://ads.osdn.com/?ad_idU88&alloc_id065&op=click _______________________________________________ Jmol-developers mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jmol-developers
