> Hi all, Hi Jarkko > So, is it so that norm conserving way is the only one for g thus far? > If is, why and would it be doable with ultrasoft?
Yes, norm conserving is the only way within 'linear response' method implemented so far. Big part of it is there actually, just needs a careful check on equations and probably addition of overlap operator 'velocities'..It is high priority in our todo list. If you want to contribute in implementation, it would be great. Otherwise, stay tuned! emine kucukbenli, postdoc, theos, epfl, switzerland ps: i believe there is an almost ready to be released "converse approach" version for g tensor with USPPs. For more on converse, check the references here: http://code.google.com/p/converse-nmr/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://pwscf.org/pipermail/pw_forum/attachments/20130122/789a6de4/attachment.html
