Question about compute msd

LAMMPS LAMMPS General Discussion
1

Hello! I was confused about the ‘compute msd’. I used this command to get msd of particles but the value was too small to cooperate with the dump file I have output. Below is my code:

#####
reset_timestep 0
fix shear first addforce 0.082 0 0
dump final_one all custom 5 dump.finalone id type x y z fx fy fz vx vy vz omegax omegay omegaz
dump_modify final_one sort id
compute 1 all msd com yes
variable msdx equal c_1[1]
variable msdy equal c_1[2]
variable msd equal c_1[4]
variable istep equal step
fix 9 all print 5 "${istep} ${msdx} ${msdy} ${msd}" screen no file msd1.dat
run 200000000
undump final_one
unfix 9
#####

The question is in my dumpfile the displacement of particles is on the order of 1 while in MSD file the msd is just on the order of 1e-5

I guess it is because that the value of compute msd is between each step but I am not sure.

Any help will be appreciated! Thanks advance!

2

That statement makes no sense. In research it is almost always a bad idea to guess. You can provide a hypothesis, but that needs to be at the very least be supported by a statement from the documentation or by the source code.

How carefully have you read the documentation of compute msd?
What does it say about the com yes keyword? Why did you choose it?
When you compare the to the dump output, did you factor in its effect, too?

3

Thanks for your suggestion. The point is the com yes keyword! Thank you!