Re: [AMBER-Developers] ntb vs ifbox?

From: Ben Roberts <ben.roberts.geek.nz>
Date: Tue, 6 Dec 2011 10:08:18 -0500

On 4/12/2011, at 9:13 p.m., case wrote:

> On Sun, Dec 04, 2011, Yong Duan wrote:
>>
>> It's so easy to mess up now. If one accidentally sets wrong ntp, you'd
>> have to restart.
>
> Did you mean "ntb" here? I don't know how to prevent people from having
> to set the correct value for ntp.
>
> I think the "ntb" variable is redundant, since its proper value can be
> determined by examining IFBOX and ntp. Keeping in mind backwards
> compatability the best approach I see is this:
>
> 1. Calculate ntb from IFBOX and ntp;
> 2. Exit with an error message if the user's input ntb is different from
> that calculated.
> 3. Update as many of the example scripts and tutorials as we can to avoid
> using ntb at all.

I would agree with the general statement about the redundancy of ntb; I've never understood why it's been necessary to set both ntb and ntp.

I also agree with Jason's earlier comments: forcing people to edit the prmtop if they want a particular outcome is the wrong way to go about it. I suggest it would be easier to foul up a simulation by an almost-but-not-quite-correct prmtop file than an obviously wrong input file.

It seems to me that the question is whether we're better off making people edit the code itself if they want to use a non-standard value of ntb, or if we let them set ntb and just print out a warning or something in the log file.

Cheers,
Ben


_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Tue Dec 06 2011 - 07:30:03 PST
Custom Search