Could it be this: http://www.mail-archive.com/[email protected]/msg35107.html
I've posted a few times on the fact that this tool is broken for constant pressure simulations.
I'm not sure why both even and odd would be under the overall average, but it seems possible.
Chris. -- original message -- Hi all. I've noticed an unexpected behaviour using g_density. I have a trajectory (with time step 2) and it is split into two subsets (using trjconv) with, lets say, even and odd steps respectively. According the the algorithm I expected that the density obtained with the original trajectory would be the average of the density with each generated trajectory. However, that was not the case, and for each slice, while density calculated from the subsets was practically the same, the density calculated with the original was different (greater or lower). My system is a membrane, the original trajectory is 50ns long (25001 steps) and I calculate the density along the bilayer normal (z). I am using GROMACS4.5.3 and a I've issued the following commands: *To get the subsets: trjconv -f bilayer.xtc -s bilayer.trp -o bilayer_odd -dt 4 trjconv -f bilayer.xtc -s bilayer.trp -o bilayer_even -dt 4 -b 2 The generated trajectories have 12501 and 12500 steps respectively. *To calculate the densities: g_density -f bilayer.xtc -s bilayer.tpr -o dens_all g_density -f bilayer_odd.xtc -s bilayer.tpr -o dens_odd g_density -f bilayer_even.xtc -s bilayer.tpr -o dens_even The result is attached. dens_odd and dens_even are very similar, but surprisingly (at least for me), dens_all is different (it actually integrates more atoms than the other two). I haven't seen anything in the code that could explain that result (maybe when removing pbc??). Is it a bug or a mistake from me? Thank you! Javier -------------- next part -------------- -- gmx-users mailing list [email protected] http://lists.gromacs.org/mailman/listinfo/gmx-users Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/Search before posting! Please don't post (un)subscribe requests to the list. Use the www interface or send it to [email protected]. Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
