[lammps-users] fix ave/spatial standard deviation


I'm trying to calculate the standard deviation in the stress values for fix ave
spatial command. I am shocking a 25x25x200 block of copper. I used the data in
the dump files to compute the std deviation (<x*x> - <x>*<x>)^0.5 for a plane of
atoms of thickness specified in the fix command. However, the value I've
computed for the average stress in a certain plane is different to the value
outputed for that plane by fix avespatial. I'm positive that my code is correct.
How does the fix ave spatial command compute the average value of stress within
a thinkness delta?


Brennig Williams

Look in fix_ave_spatial.cpp. If you set the time parameters appropriately
you can get the spatial average for a single snapshot in each layer. This
would be averaging the same values for that snapshot that would
be in a dump file of per-atom stress. So you should be able to match
them up.