Re: Incorrect Atom Count

From: Charles Zhao (
Date: Wed Dec 02 2009 - 17:58:03 CST

binCoordinates->restart.coor file
coordinates->pdb file
That's how it's set, which is as far as I know correct.

From: BIN ZHANG <>
To: Charles Zhao <>
Sent: Wed, December 2, 2009 2:31:20 PM
Subject: Re: namd-l: Incorrect Atom Count

Hi, Charles:

Normally, this kind of error occurs because you have mixed the usage of binary file with normal pdb file, i.e. use pdb file for "binCoordinates" or vice versa. Have you checked this?


On Dec 2, 2009, at 1:53 PM, Charles Zhao wrote:

After equilibrating with a periodic cell for 68000 steps, my simulation terminated, complaining of the patches growing too small. I have seen online that the easiest and most useful approach to dealing with this is simply to restart the simulation from saved vel, coor, and xsc, which will cause the patches to reset and the simulation to proceed. However, attempting to do this, I find that the simulation terminates upon reading the coor file, complaining of:
>FATAL ERROR: Incorrect atom count in binary file
>This makes no sense to me. WrapWater and WrapAll are turned on, so no molecules could escape, and it seems strange that the atom count would somehow change. I have not found a satisfactory solution so far, and using the old version of the coor/vel/xsc gives the same result.


This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:53:33 CST