Re: [AMBER] Should ig value be set explicitly?

From: Adrian Roitberg <>
Date: Mon, 13 Feb 2017 11:42:18 -0500


Be careful about which version of the manual and which version of amber
you are using, that is one of the things that have changed over the years.

I am afraid I have bad news for you. If you are continuing your
simulations after a restart with the SAME value of ig, then trouble
might show up. Since you never know if and when if will show up, I would
not trust the results. The effects are not readily visible (i.e. the
system does not blow up, it just behaves in a way that looks ok, but it
is not). You can look at some papers from my group on this issue.


On 2/13/17 11:17 AM, Azade Yazdan Yar wrote:
> Dear all,
> In my simulations I am using 'ntt=3'.
> I noticed that the ig value is better to be set to '-1'.
> I forgot to do this and now I have a set of restarted simulations in which
> ig parameter is not set. Although in the manual it says that the default
> value for this parameter is '-1', but when I look in the mdout files of
> different restart runs, all of them have the exact same ig value (71277)
> (actually ig has this value in any simulation which I have done so far with
> amber!).
> My question is that: if the default value is -1 in the case where I have
> not defined this parameter explicitly, why all the mdouts have the same ig?
> and should I rerun my simulations with 'ig=-1'?
> Thanks a lot for your help and concern.
> Best regards,
> Azade
> _______________________________________________
> AMBER mailing list

Dr. Adrian E. Roitberg
University of Florida Research Foundation Professor.
Department of Chemistry
University of Florida
AMBER mailing list
Received on Mon Feb 13 2017 - 09:00:02 PST
Custom Search