Csfit

Dear Axel,

I have a problem when using the csfit command. I set three directions,namely (100) (110) (111),in the dir.in file and then csfit -d. By default it generates 15 structures
(stretch_100_110_111_0_0,stretch_100_110_111_0_1,stretch_100_110_111_0_2,stretch_100_110_111_1_0,stretch_100_110_111_1_1,stretch_100_110_111_1_2,stretch_100_110_111_2_0,stretch_100_110_111_2_1,stretch_100_110_111_2_2,stretch_100_110_111_3_0,stretch_100_110_111_3_1,stretch_100_110_111_3_2,stretch_100_110_111_4_0,stretch_100_110_111_4_1,stretch_100_110_111_4_2)
in directory 0 and 1, respectively. After the calculations of the structural energies are completed. It displays that
"Structure generation completed. Calculations done, fitting the data. Not enough directions specified - need at least 3.
I only have the output file "csdebug.out", and there are no cs.in and cs.log files.
I tried to add more directions , 5 in total, (100) (110) (111) (210) (311),in the dir.in file. It seems that the dir.in only can read the first three directions, and generates the same structures as the dir.in which containing 3 directions do.

Regards,
wzchen

Sorry but I’d need you input and ouput files (excluding large vasp files) to see what is happening.

Dear Axel,
Thanks for your time. Here you will find the input and output files. I should note here that I did not use the command "pollmach runstruct_vasp -w csvasp.wrap &" to automatically control the jobs. Intead, I went to the subdirectories and used "runstruct_vasp -nr &", and then copy the generated files (INCAR, KPOINTS, POSCAR and POTCAR) to do the vasp calculations in another cluster machine. I copy the resulting files (CONTCAR, OSZICAR and OUTCAR) to the subdirectories and did "runstruct_vasp -ex &".

Regards,

wzchen

Your dir.in file should contain

1 0 0
1 1 0
1 1 1
2 1 0

Note the space between the numbers.
Sorry the help file was not very clear on this point. The next version with include the instruction:
Each index must be separated by a space and
each direction must be on a separate line.
I’m afraid you’ll have to rerun your vasp calculations.

:slight_smile:

wzchen