I have used TINKER for teaching purposes and recently I asked students to install it on their own computers to work on assignments outside of class. The problem that Mark Fleharty mentions (Molecule loads but does not display on screen) came up a few times in these various installations.
I have mentioned it to Jay Ponder (the creator of TINKER) but haven’t heard anything back yet.
One thing that it ISN’T due to is the colour of the molecules – that much was clear after looking at the various machines.
Some possible sources of the problem:
1) Do you have the Java3D extensions properly installed and does TINKER know where to find them?
2) On Windows machines there should not be spaces in the fully qualified path names for .xyz, .key or other files that FFE/TINKER must read from and write to. This caused various other problems with TINKER and I am not sure that it is relevant to this particular problem. Note that I said “fully qualified path names” – your file called structure.xyz might actually be C:\My Documents\Molecular Models\TINKER\structure.xyz or something like that. FFE reads these names properly but when it passes them back to the executable modules it uses a simple command line and that will NOT accept spaces in file names.
3) We found that the “invisible molecule” problem sometimes seemed to be associated with a particular .xyz file, while other files would read and display properly. We were never able to identify what was different about these files but possibly there are embedded but non-displaying characters in some files from various text editors that are making FFE unhappy.
4) Again assuming you are on a PC, try switching off any fancy hardware acceleration you might have on your graphics card. I doubt that FFE is written to make use of these features, but it is quite possible that some of them might be incompatible with the JAVA graphics code in FFE. If FFE displays correctly with dumbed-down graphics, then try switching the features back on one at a time and testing FFE in between. Perhaps you will find the problem this way – we did not have time to try this with the students’ PCs but it is quite plausible as an explanation.
Mark asked whether anyone else uses TINKER – yes it is fairly widely distributed but not the most common package for dynamics and modeling.
Dr. Philip G. Hultin
Professor of Chemistry,
On 3/11/07, John W Daily john.daily_-_colorado.edu <owner-chemistry,ccl.net> wrote: