Re: Pressure Questions

From: Jérôme Hénin (
Date: Wed Feb 10 2010 - 11:52:30 CST

Hi Rick,

On 10 February 2010 15:16, Rick Calloway <> wrote:
> Hi community,
> Can someone help me understand the relationship between the 3 by 3 pressure
> tensor ouptut on the PRESSURE: line and the single number printed under the
> PRESSURE column heading on the ENERGY: line?
> I assume they are both in bar so is there a tensor matrix operation that is
> done on the 3x3 matrix that results in a single magnitude?

The scalar pressure is one-third the trace of the stress tensor (which
really means the average of the diagonal elements).

> Second, when profile pressures are output the result is only a single 3
> column vector.  When looking for the instantaneous pressure along the
> profile (like in Justin Gullingsrud's paper on Lipid Bilayer pressure
> profiles) what does the 3 col vector represent

That's in the user's guide, see the pressureProfile keyword.

> and how are the normal and
> lateral pressures calculated?

Should be clear when you have the answer to the previous question.

> Lastly,  the documentation between 2.6, 2.7b1 and 2.7b2 is a little
> confusing as to what is included in the runtime output.  When doing PME is
> the profile pressure output complete or does the PME contribution to
> pressure still have to be calculated from the dcd file later and added in?

Yes, the two-step process is the only way, unfortunately. Only
post-processing the trajectory will give you the long-range PME


This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:53:45 CST