d-d hoppings between different sites not currently removed

[Sohrab Ismail-Beigi]

At present, d-d hoppings between different sites are not removed when reading in the hopping data in inputhrbin(). This is because the current SVO.bin test case actually has large imaginary hoppings and also some large d-d hoppings (~0.15 eV or ~0.1 eV) which is strange. The code to remove d-d hoppings is inside inputhrbin() and has been checked but is commented for now.

Resolution: either add the code back and make a new default test (with better SVO.bin?) or just don't remove these things and forget the issue.

[Sohrab Ismail-Beigi]

Code that removes d-d hoppings now added (uncommented) from inputhrbin.m

[Sohrab Ismail-Beigi]

This issue is back: if the d-d hopping are removed, of course the band structure at U=Up=J=0 (non interacting) starts differing from the DFT one. In some cases differences can be large. So what is needed is to create a flag on whether to do this or not and inputhrbin.m needs to get the flag. And the flag needs to be documented.... (and the setup_system.m file need updating in examples 1 and 3 to have the flag). Easy stuff. Sohrab will try to do this ASAP.

[Sohrab Ismail-Beigi]

Still needs fixing

[Sohrab Ismail-Beigi]

I edited inputhrbin() to take a new flag zeroddflag which turns on the d-d intersite hopping on or off (1 or 0). Then had to edit setup_system.m to add the flag, the calls to intputhrbin() to use it, and the documentation README.pdf to tell the user what this is. All setup_system.m files needed to be update as well.

[Sohrab Ismail-Beigi]

Issue has now been resolved.