Dumping out vector defined through fix property atom

Dead LAMMPS users,

I have a simple problem related to dump custom command. I have defined a per-atom vector which has 12 elements. I perform some calculations on this vector and want to dump it out, however I get the following error.

“Dump modify fix ID does not compute per-atom vector”

This is the input script commands I use:

fix LB all property/atom LB vector yes no no -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1
dump dmp all custom {nDump} dump.{OutName} id type f_LB[1] …f_LB[12]

The manual says I should be able to do this. I can’t understand what is wrong.
Thanks for the help in advance.

Sincerely,
Liza

Dead LAMMPS users,

I have a simple problem related to dump custom command. I have defined a per-atom vector which has 12 elements. I perform some calculations on this vector and want to dump it out, however I get the following error.

“Dump modify fix ID does not compute per-atom vector”

This is the input script commands I use:

fix LB all property/atom LB vector yes no no -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1
dump dmp all custom {nDump} dump.{OutName} id type f_LB[1] …f_LB[12]

The manual says I should be able to do this. I can’t understand what is wrong.
Thanks for the help in advance.

Your fix syntax disagrees with the (current) manual which also states that the fix offers no output options. Please explain why you think otherwise and provide confirmation.

Axel

This fix is seems to work fine in order to define a per-atom vector, since I am able to manipulate the values in the *.cpp files.
The dump custom command works when I use a scalar instead of a vector when I use f_name. So if I put

dump dmp all custom {nDump} dump.{OutName} id type f_LB

and LB is a scalar it doesn’t spit out any errors, and outputs the quantity I want.
Would I have to use a compute property/atom and then use dump custom by using c_LB[1]…c_LB[12]?

Thanks

This fix is seems to work fine in order to define a per-atom vector, since I am able to manipulate the values in the *.cpp files.
The dump custom command works when I use a scalar instead of a vector when I use f_name. So if I put

dump dmp all custom {nDump} dump.{OutName} id type f_LB

and LB is a scalar it doesn’t spit out any errors, and outputs the quantity I want.
Would I have to use a compute property/atom and then use dump custom by using c_LB[1]…c_LB[12]?

You don’t explain how you can use a feature that doesn’t exist and is not documented. How should anybody give advice on this?

More specifically, this command:

fix LB all property/atom LB vector yes no no -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1

is not correct syntax in current LAMMPS. It gives an error:
ERROR: Illegal fix property/atom command (…/fix_property_atom.cpp:97)

What version of LAMMPS are you using, and where did you get the
code for fix_property_atom.cpp? I don’t recall the command ever had the
syntax you are using.

Steve