AMBER Archive (2009)

Subject: Re: [AMBER] Negative "number of measurements" in PTRAJ

From: Thomas Cheatham (tec3_at_utah.edu)
Date: Tue Apr 07 2009 - 11:57:28 CDT

> I am trying to use PTRAJ to calculate some radial distribution functions
> from my simulations. I noticed that the number of measurements is negative.
> I checked the source code and it appears that the number of measurements is,
> indeed, stored in an integer variable. I am only processing some 20000
> frames. Is this a bug?

20,000 frames * lot's of water * lot's of solute atoms could in
principle blow the integer format (> ~2.1 billion measurements). This is
probably a bug that I haven't encountered yet... I/we should update the
radialInfo->measurements counter in actions.h to a float.
info->measurements++ may also need to be altered to be:

        info->measurements+= 1.0;

I'll look into this.

--tec3

_______________________________________________
AMBER mailing list
AMBER_at_ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber