[gmx-users] [Fwd: questions of afm pulling code]

[Berk Hess]

>From: David van der Spoel <spoel at xray.bmc.uu.se>
>Reply-To: Discussion list for GROMACS users <gmx-users at gromacs.org>
>To: gmx-users at gromacs.org
>Subject: [gmx-users] [Fwd: questions of afm pulling code]
>Date: Thu, 09 Dec 2004 15:17:23 +0100
>
>-------- Forwarded Message --------
>From: Jian Zou <zouj01 at mails.tsinghua.edu.cn>
>To: spoel at xray.bmc.uu.se
>Subject: questions of afm pulling code
>Date: Thu, 9 Dec 2004 22:02:28 +0800
>Hi Mr. David van der Spoel,
>
>I tried to use "afm_rate1=0.001" but found the pulling group moved too 
>fast.
>
>I checked the manual but found that the unit for afm_rate is changed from
>nm/timestep to nm/ps.
>I'm just wondering if nm/ps for afm_rate is a typo in manual 3.2, and some
>one in the mailing list also agrees with me.
>
>I checked gmx-3.2.1/src/mdlib/pull.c, and fould
>pull->grp[i].spring[m] = pull->grp[i].AfmInit[m] + pull->ref.x_unc[m] +
>    pull->grp[i].AfmVec[m]*pull->grp[i].AfmRate*step*dt;
>It seems that nm/ps should be correct for gmx 3.2.1.
>
>Therefore I am confused.

It is indeed nm/ps.
Why do you think the pulling is too fast?

>
>
>Another problem I'm facing is how to set ref_group to be the absolute
>coordinate. The manual says that just leave it blank. I did so but error
>occured. Is this a bug or I missed something?

I have fixed a bug with absolute coordinates for umbrella sampling.
This fix probably also solves your problem.
Please try the CVS version.

Berk.

_________________________________________________________________
Play online games with your friends with MSN Messenger 
http://messenger.msn.nl/