[lammps-users] fix msd problem

Dear users,

I am observing different values in the fix msd dump files depending on the number of runs (repeating in a loop) for a constant total number of runs.

The time the fix is issued is exactly the same for all examples, however the dump values differ up to 20 %.

Simplified script:

Loop 6/9/18 times

{

Fix msd

Run 600/400/200

}

If I compare the msd files issued at the same time, the values are different, but why?

I use the latest version of lammps.

here is the complete, essential part of my input script:

a=runs

variable c equal 9

variable a loop $c

label loopa

group CH4_diff_$a region diff

b=counts_diff

variable b equal count(CH4_diff_$a)

if $b > 0 then “include fix.txt”

run 400

if $a == $c then “jump in.txt end”

next a

jump in.txt loopa

label end

This is the fix file

fix 5_$a CH4_diff_$a msd 100 diff_$a.txt com no

thanks for your help,

Thomas

Dear users,

thomas,

I am observing different values in the fix msd dump files depending on
the number of runs (repeating in a loop) for a constant total number
of runs.

this is no big surprise.

MSD values can take a very long time to converge and they can
converge on different time scales. i've looked into that a loooong
time ago. you can see some slides from a talk on the subject
in pages 10 to 13 at:
http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/files/talk-trieste2004-water.pdf

cheers,
   axel.