Re: different result from different cpu

From: Xiaoqing Wang (kitty_wxq_at_yahoo.com.cn)
Date: Tue May 10 2005 - 19:59:45 CDT

Jerome,

My system is protein+membrane+water.
The config files attached.

Any tips would be greatly appreciated!

Sincerely,
Xiaoqing

> Xiaoqing,
>
> What system are you simulating?
> Can you send the config and output files?
>
> Jerome
>
> On Tuesday 10 May 2005 14:31, Xiaoqing Wang wrote:
> > Dear all,
> >
> > We have recently started to use NAMD to run macromolecular simulations.
> > NAMD is a powerful and user-friendly software, and we really like it.
> >
> > I have compiled charm++/namd on amd64.
> > "./build charm++ mpi-linux opteron -no-build-shared" when build charm
> >
> > I downloaded binary namd "NAMD_2.5_Linux-i686" on my own intel P4 CPU.
> >
> > But when I run the same program with the same input files on the
> > different cpus, the results are different.
> >
> > The attachment file "equil.namd" is input file.
> > The attachment file "equil_out2.xst" is from intel cpu and
> > "equil_out2_amd64.xst" is from amd64 when I use 1 cpu.
> >
> > My system become larger when equilibrate the system on amd64.
> >
> > Any more files needed?
> >
> >
> > Any tips would be greatly appreciated!
> >
> > Sincerely,
> > Xiaoqing

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:39:26 CST