[gmx-users] simulating in double precision but writing in singleprecision

Berk Hess gmx3 at hotmail.com
Tue Jul 20 16:53:43 CEST 2004




>From: Milton Taidi Sonoda <msonoda at iqm.unicamp.br>
>Reply-To: Discussion list for GROMACS users <gmx-users at gromacs.org>
>To: gmx-users at gromacs.org
>Subject: [gmx-users] simulating in double precision but writing in 
>singleprecision
>Date: Tue, 20 Jul 2004 11:18:02 -0300
>
>Hi all
>
>I'm using gromacs version 3.2.1.
>I'm simulating a system under NVE ensemble, and in order to
>conserve energy I have to do that in double precision.
>However in the analisys I'm performing, wave length dependent
>dielectric response, the double precision configuration is not
>necessary, and also the OS (freeBSD 4.7) I'm
>using do not support the big trr files produced.
>Moreover, sadly vmd (at least the version I have, which is 1.7.1)
>do not read trr in double precision.
>
>Is it possible to simulate in double precision but writing the
>trajectory in single precision?
>
>If not, anyboby has an advice/hint on thise issues?

If you only need coordinates you can use xtc files,
you can set the precision, but default is probably enough.
xtc files are much smaller than trr.

http://www.gromacs.org/documentation/reference_3.2/online/mdp_opt.html#out

Berk.

_________________________________________________________________
Protect your PC - get McAfee.com VirusScan Online 
http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963




More information about the gromacs.org_gmx-users mailing list