selective dynamics

[osharia]

I get the following error when I try to use selective dynamics
========
POSCAR found : 1 types and 2 ions
LDA part: xc-table for Ceperly-Alder, standard interpolation
No initial positions read in
=================




My poscar file looks like
==============
Si
5.38
0.0 0.5 0.5
0.5 0.0 0.5
0.5 0.5 0.0
2

Selective dynamics
Cartesian
0.0 0.0 0.0 T T T
0.29 0.25 0.25 F F F
=============

Without selective dynamics everything works fine. What could be the problem?

[forsdan]

Remove the empty line between the "2" and the "Selective dynamics" and you will find everything to work out just fine.

cheers,
/Dan

[osharia]

I removed the empty line but did't resolve the problem.

[forsdan]

Well, Your POSCAR without the empty line is working for me with some POTCAR plus some simple INCAR, KPOINTS files. So then something else not related to your POSCAR is off as well (unless you have a hidden character not showing up in the editor that you are using). Please post your INCAR file as well and please doublecheck that your input-files are correct.

<span class='smallblacktext'>[ Edited Wed Mar 31 2010, 06:27PM ]</span>

[alex]

Are you sure cartesian and selective dynamics works in combination?
Workaround to check: remove "selective dynamics" and all the T afterwards, start vasp (only a few seconds) until the direct coordinates appear and put them into your POSCAR _with_ selective dynamics.

cheers,

alex

[osharia]

alex

I tried both cartesian and direct. I tried everything with POSCAR, yelled it, begged it but didn't help. This is my INCAR file
PREC = high
NELMIN = 4
NELM = 70
ENCUT = 600
EDIFFG = -2E-02
NSW = 150 number of steps for IOM
IBRION = 2 ionic relax: 0-MD 1-quasi-New 2 CG
ISIF = 1 determines which DOF are changed(ions, cell vol,cell shape)
LVTOT=T
ISMEAR = 0

I tried to run on different machines but everywhere the problem is the same. I tried vasp 4 and vasp 5 but problem is still there. All of the vasps are compiled by me. Could it be because I did something wrong during compilation?

[alex]

In my case the input works fine, if you remove the empty line as Dan suggested above.

Is this your very first vasp calculation? Have you tried some example from some place else with your binary? There might be also pre-compiled binaries on the VASP ftp-server for cross-check.
Please try that before we go for a deep dive.

alex

[admin]

-- if there is one blank line between the line giving the number of atoms and the 'Selective Dynamics' line, the error you report
shows up,
-- if this blank line is removed, the error vanishes if the logicals saying whether an atom has to relax or not are given.
-- a combination of giving the 'selective dynamics' line without keeping the T T T ..., causes the same crash as the version with the blank line.
===> please first of all check whether you deleted the logicals in the line of the atoms' positions by error
If you have not, please check whether the line 'Selective Dynamics' starts without a leading blank or hidden character
('strings POSCAR' will show every character, even if it cannot be seen with your standard editor)

[debosruti]

Hi everybody,
I am a new user of VASP. I am trying to run a Selective Dynamics calculation and my POSCAR file reads like this:

System = Ni C
1.00000000
16.8481480000 0.00000000000 0.0000000000
0.00000000000 14.5908000000 0.0000000000
0.0000000000 0.0000000000 18.00000000
1 1
Selective Dynamics
Direct
0.4040000000 0.0000000000 0.3957000000 F F F
0.0000000000 0.4040000 0.3857000000 T T T

However, every time I run the calculation it recognises only 1 type and 1 ion in the output file.
I am using the POTCAR file with Ni and C in it, in that order. However, is there something to be mentioned in the INCAR file in order to indicate the 2 different types of ions? Please help.

Dev

[alex]

Are you sure you have to 2 entries in POTCAR?

try "grep TIT POTCAR" without the hyphenation

alex