[SIESTA-L] Siesta 1.3p abortion

Hi, When running either CG or FC calculations on SGI Origin3000, after a few (20 or so) cycles, the program simply stopped. There were no error messages in output files. The error from the system reads: MPI:cpuset name = 9211-0 Assertion failed: (context>>18) == 0, file comm.c, line 107 IOT Tra

Re: [SIESTA-L] FixSpin

Dear Pablo,   Thanks a lot for your kind reply.   Best Xin Pablo Ordejon <[EMAIL PROTECTED]> wrote: Dear Xin,There was a bug in 1.3 causing this problem. This bug was fixedsome time ago. You can download the patches for this and otherfixes to the 1.3 release in the Siesta web page (in Distribution

Re: [SIESTA-L] FixSpin

Dear Xin, There was a bug in 1.3 causing this problem. This bug was fixed some time ago. You can download the patches for this and other fixes to the 1.3 release in the Siesta web page (in Distribution/Patches). Sincerely, Pablo Ordejon According to stargmoon: --0-1997243756-1086200477=:57

[SIESTA-L] FixSpin

Dear Siesta-1.3p developers,   I am testing the spin polarized calculation by using FixSpin option. I always got the WARNING: Qtot and Tr[D*S] are not equal, in fact they are very different.  However, with the same *.fdf and pseudopotential, I went back to version0.15, there are no such problems.

[SIESTA-L] problem for fcc band structure

Content-Type: multipart/alternative; boundary="--=_1086187721-11935-110" Content-Transfer-Encoding: binary MIME-Version: 1.0 X-Mailer: MIME-tools 5.411 (Entity 5.404) Hi, all: I found that I cannot get correctely band structure for fcc metals such as Ag. I cannot fix if it is potenti

[SIESTA-L] exchange-correlation potential and energy

Dear All, hi, I wanted to calculate the exchange-correlation (XC) potential in each point of my structure. In Siesta, there is no command for calculating the XC potential. But I think there is one way to calculate XC potential. In siesta we can get the Total potential and the Electrostatic po