[gmx-users] neighborsearching in semiisotropic systems

[Hannes Loeffler]

On Tue, 2008-06-03 at 17:26 +0200, Till Rudack wrote:
> Dear Justin,
> the system is well minimized. I performed a steepest descent 
> minimization followed by a conjugate gradient minimization and then 
> minimize it again with steepest descent. The problem occurred after 3ns 
> or in the second run after 5ns so I do not think it is a problem of  
> minimization .

I did not check single vs. double precision but I have problems on
certain multi-processor systems.  When I move my jobs to a different
hardware they typically run without further problems.