When loading log files from Orca or Gaussian, which contain a frequency calculation, into Avogadro2 a window with frequencies pops up. Once a frequency is selected, an animation of the corresponding vibration can be visualized. That’s awesome.
But sometimes, i.e. in multi step jobs or during a geometry optimization with periodically recalculated Hessians, there is more than one frequency calculation in a log file.
The expected behaviour would be that one is able to access / viszalize all the frequency calculations, with the last one to be displayed by default.
The actual behaviour is that only one frequency calculation is displayed without information about which one is shown (first ?, last ?).
By the addition of two buttons (previous and next Hessian) at the top of the dialouge, one should be able to select the desired frequency calculation to inspect, like it is possible to view the steps of a geometry optimization in the animation window.
That’s going to take some time, since it will require adding support for storing Hessians for each coordinate set. The UI you describe is reasonable, but there’s some “plumbing” including in CJSON that’s needed.
Also ran into this in my workflow. I’m running calculations where the Hessian is recomputed every 15 steps to deal with some pesky small imaginary frequencies. I have my own code that processes the outputs correctly but when I use Avogadro to visualize I only see the modes from the first job.
Not sure how many people would need this functionality / if it is worth the effort. I can live without it but I’ll add myself to the “I’d be nice to have it” tally.
Yes, there’s definitely a need to track a variety of properties through multiple steps. We’re getting into some transition state / reaction path calculations, and tracking the vibrational modes through the IRC / NEB is very helpful. (This would also be a good teaching exercise.)
I have some ideas on this and I’m penciling in early summer for “more properties across steps” work (e.g. bond lengths / geometry, forces, orbitals, etc.)
And hopefully the NSF proposal we submitted back in December comes through because it’s very much along these lines.