Note to Self: Calculations Edition

Not sure whether I've mentioned this, but the kind of (solid state) calculations I run at work typically converges quickly, on the order of seconds to minutes. Unfortunately, this isn't always true.

The main reason the calculations go so quickly is that the theory and basis sets used are very...um...simple. Kind of like VSEPR - the amazing thing is that it works as well as it does.

For "busy" cells with lots of atoms, though, the calculations still take a fair bit of time, even if the system is high symmetry. Calculating over more k-points adds to the problem. Right now my record is seven hours and counting. Thankfully, this time I'm not running calculations on a computer that restarts every 24 hours, so at least I can leave this overnight.

So far the calculations are relatively straightforward, because the symmetry is the same in all cases. Let's hope that the more problematic ones work as smoothly as this set, though I am not optimistic in that regard.

[Edited]
An energy calculation is diverging. Crap. I have to let the current calculation finish before trying something else, though.

[Edited again]
Previously mentioned calculation eventually converged with no prodding from me whatsoever. Whew. Now I'm starting the more problematic calculations; it's probably time to pay a visit to The Program That Shall Not be Named. If it's smart, the program might even do some subgroup operations for me.

Incidentally: The rigid band model is a freaking cop-out. Psh.