[gmx-users] question regarding g_energy -aver output in gromacs 4.5.4

[Jacob Alan Spooner]

Dear gmx_users,

I have a question regarding the 4.5.4 version of gromacs, and the g_energy program.
In the past I had used g_energy with the -aver option, which would give me partial
sums along with my instantaneous energy values at each output step.  This allowed
me to take the difference between two successive partial sums divided by the my
nstenergy value to get the average energy over that output period.  Now when I use
the -aver option my output contains my instantaneous value, and what is said to be
the exact average.

I have the feeling that this exact average is what I used to calculate by hand, i.e
the average of the energy over each output period.  I tried to test this by calculating
the average by hand using gmxdump and my ener.edr file and my numbers match up
 to the fourth decimal place, but aren't exactly the same.  Is this difference simply
because gmxdump is truncating the energy values and giving them to me at lower
precision? Am I correct on my assumption of what this -aver option is giving me?
Any guidance would be appreciated as I'm not so good at understanding the source
code.

Thank you in advance

Jake Spooner
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20110505/9b7b3060/attachment.html>