[Wien] Number of charge concentration in the system

2017-10-31 Thread halim said
Dear Wien2k  users community, I would like to ask you if we can get the results of  total number of concentration of the system from wien2k codeLooking forward your answer. Halim Said ___ Wien mailing list

Re: [Wien] Querry in a resultant structure

2017-10-31 Thread Abderrahmane Reggad
Hello again I have repeated the calculation using the low symmetry structure and i found a close results to those of the monoclinic one. I have checked the wyckoof positions and the symmetry operations of the orthorhombic and the monoclinic structures and I have found that the later is a

Re: [Wien] problems with gfortran when unit 6 is associated to file

2017-10-31 Thread Pavel Ondračka
Od: Peter Blaha Komu: A Mailing list for WIEN2k users Datum: 31. 10. 2017 15:27:42 Předmět: Re: [Wien] problems with gfortran when unit 6 is associated to file "I can reproduce this. Apparently, some "clever informatics-guy" has

Re: [Wien] configuring static linking with siteconfig

2017-10-31 Thread Laurence Marks
I agree that you cannot link Scalapack and different versions of mpi (e.g. openmpi versus impi), the blacs are different. However, one can (I have several times) do a static link of Scalapack and then use different versions of impi, different infiniband drivers etc on different clusters. I have

Re: [Wien] configuring static linking with siteconfig

2017-10-31 Thread Peter Blaha
SCALAPACK and mpi are interconnected !!! You cannot statically link scalapack and then dynamically add different versions of mpi. fftw: it is so trivial to copy the fftw libraries into eg. $HOME/libs and link them from there And a home-directory is hopefully available on every cluster

Re: [Wien] problems with gfortran when unit 6 is associated to file

2017-10-31 Thread Peter Blaha
I can reproduce this. Apparently, some "clever informatics-guy" has changed the default behaviour, namely that write(*,...prints to stdout while write(6,...prints to the connected file, and ONLY if it has not been connected explicitly via an OPEN statement, it would print to

[Wien] CPU advice

2017-10-31 Thread Laurence Marks
I am going to buy a few more nodes for my own small cluster. From what I can see the Skylark 6130 looks to be an appropriate CPU/cost compromise, taking into account that I have 4 IB connections available so 32 cores/node is appropriate. I am undecided as yet about 4Gb/core or 6Gb/core -- the

Re: [Wien] Methods to change spin states

2017-10-31 Thread Laurence Marks
Dear Pavel, Thanks. Probably this and other small utilities should go into a directory somewhere that is on the unsupported page. On Thu, Oct 26, 2017 at 8:15 AM, nov...@fzu.cz wrote: > Dear Laurence, > > when we calculated the exchange integrals by subtracting total energy of >

Re: [Wien] Methods to change spin states

2017-10-31 Thread novakp
Dear Laurence, when we calculated the exchange integrals by subtracting total energy of different spin configurations we first converged single configuration. Starting .clmup(dn) and .vorbup(dn) for next configuration were obtained by fliping the selected spin. To this end simple programs were