Dear Prof. van de Walle,
Thanks very much for your development of the atat, it actually help me a lot on my Phd research work. Now, I have confused by a problem when I tired to fit a Cluster Expansionof the Al-Mg-Zn alloy.
With mmaps i runing, I bulided four structures by using command-"touch ready" and separately calculated the energy of these structures by typing command "runstruct_vasp & " , and then ,an energy file with nothing and an error file contains "An error occured during this vasp run. Check vasp.out.static and vasp.out.relax for more information." were generated.
The contnt of vasp_out.static was shown below.
[guest2@node14 0]$ cat vasp.out.static
running on 1 total cores
distrk: each k-point on 1 cores, 1 groups
distr: one band on 1 cores, 1 groups
using from now: INCAR
vasp.5.3.3 18Dez12 (build Mar 15 2014 11:08:31) complex
POSCAR found : 0 types and 0 ions
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image PC Routine Line Source
vasp 000000000052055E Unknown Unknown Unknown
vasp 000000000040EA46 Unknown Unknown Unknown
vasp 000000000040A72C Unknown Unknown Unknown
libc.so.6 000000367F21ECDD Unknown Unknown Unknown
vasp 000000000040A629 Unknown Unknown Unknown
Hi,dear Axel
Thank very much for your reply and Sorry for not telling you I have resolved that problem insteaded of the lattice constant parameter 111 90 90 90 with 4.095 4.095 4.095 90 90 90 in lat.in. what stupid I am.
But now I have another difficult thing. I have been calculating the Cluster Expansion of the Al-Mg-Zn ternary alloy about 15,000 mins using the atat with paralleling 4 core on mmaps and 12 cores on vasp, and it has not converged yet.
Here I give you the input files lat.in and vasp.wrap below.
I see nothing really wrong with your input (other that the KPPRA value being on the small side for a metallic system).
But I can’t tell you what is wrong since you don’t tell me in which way your results are not converged… (new group states? cv score too large?)
Maps version 3.18
The internal database of structures extends up to 7 atoms/unit cell, see predstr.out
Among structures of known energy, true ground states differ from fitted ground states
New ground states with at most 7 atoms/unit cell predicted , see predstr.out
Crossvalidation score: 0.00855673
This is content of maps.log, neither the groud state nor the CV have converged. what can I do?