Re: Size of Timestep

From: Roman Petrenko (rpetrenko_at_gmail.com)
Date: Mon Apr 05 2010 - 11:20:26 CDT

hint: integration time step is bound by the fastest bond vibration.

On Mon, Apr 5, 2010 at 8:54 AM, Axel Kohlmeyer <akohlmey_at_gmail.com> wrote:
> On Mon, Apr 5, 2010 at 7:33 AM, Aditya Ranganathan <aditya.sia_at_gmail.com> wrote:
>> Hi,
>> I`m running an all atom simulation (NVT ensemble) to study the interpeptide
>> interaction between ACTH and Beta Endorphin chains. I would like to know
>> about the value of the timestep which would be suitable for the same. I
>> generally use timestep 2fs for all my simulation. However, I would like to
>> know if I can use larger values of timestep, say 3fs. Will doing so cause
>> instability or lead to a decrease in accuracy.
>>
>> I would like to try a timestep of 3fs as it directly gives a speedup in the
>> simulation and allows me to access longer simulation times. What dictates
>> the length of the timestep in any simulation? Is any timestep more than 2fs
>> not recommended?
>
> if you don't know the answer to this question yourself, you should
> not run MD simulations. please pick up a text book on MD and/or
> numerical integration and learn about these issues.  many even
> have detailed error analysis.
>
> cheers,
>    axel.
>
>>
>> Thanks ,
>>
>> Srivastav Ranganathan
>> Junior Research Assistant,
>> IIT Bombay
>>
>>
>
>
>
> --
> Dr. Axel Kohlmeyer    akohlmey_at_gmail.com
> http://sites.google.com/site/akohlmey/
>
> Institute for Computational Molecular Science
> Temple University, Philadelphia PA, USA.
>
>

-- 
Thanks,
Roman Petrenko

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:53:59 CST