Re: margin value

From: Peter Freddolino (petefred_at_ks.uiuc.edu)
Date: Thu Jun 14 2007 - 09:58:01 CDT

Hi Madhu,
sorry to add more doubts now, but I just noticed that you're doing TMD.
Unlike margin (which won't have a physical effect), there have been some
issues with parallel TMD runs recently that we're looking into (see the
thread rooted at
http://www-s.ks.uiuc.edu/Research/namd/mailing_list/namd-l/5924.html),
so you may want to try a one processor calculation and see if the same
thing happens.
Best,
Peter

M. Madhu wrote:
> Hi Peter,
>
> Thank you very much for your reply. You cleared all my doubts.
>
> Madhu
>
> ----------------------------------------
>
>> Date: Tue, 12 Jun 2007 15:28:48 -0500
>> From: petefred_at_ks.uiuc.edu
>> To: madhu7_at_hotmail.com
>> CC: namd-l_at_ks.uiuc.edu
>> Subject: Re: namd-l: margin value
>>
>> Hi Madhu,
>> the margin shouldn't have any physical effect on the simulation -- all
>> that it does is change how namd splits up the system into patches for
>> the simulation. However, if you're doing an NPT simulation, you should
>> not expect to get identical results between runs -- please see the
>> discussion of the seed keyword at
>> http://www.ks.uiuc.edu/Research/namd/2.6/ug/node26.html.
>> The margin keyword itself is discussed at
>> http://www.ks.uiuc.edu/Research/namd/2.6/ug/node26.html#5124
>> Peter
>>
>> M. Madhu wrote:
>>
>>> Hi NAMD users,
>>>
>>> During my TMD simulation (NPT) in a lipid bilayer, I get the following fatal error.
>>> ================================
>>> FATAL ERROR: Bad global exclusion count!
>>>
>>> FATAL ERROR: See http://www.ks.uiuc.edu/Research/namd/bugreport.html
>>> ------------- Processor 0 Exiting: Called CmiAbort ------------
>>> Reason: FATAL ERROR: Bad global exclusion count!
>>>
>>> FATAL ERROR: See http://www.ks.uiuc.edu/Research/namd/bugreport.html
>>> ================================
>>>
>>> However, It works without any warning If use a margin value greater than 50.
>>> I tried with margin 50 and margin100. According to NAMD user's guide margin
>>> parameter will not change the physical results of the simulation. But the final
>>> energies obtained with margin 50 and 100 are different.
>>>
>>> Cany anyone tell me what exactly this parameter is doing in the simulation?
>>> How this parameter retains the physical results with different margin values,
>>> though the final energies are different?
>>> Whether a large margin value (eg: 50) causes any adverse effects in the simulation.
>>>
>>> Thanks,
>>> Madhu
>>>
>>> _________________________________________________________________
>>> Catch the cricket action with MSN!
>>> http://content.msn.co.in/Sports/Cricket/Default.aspx
>>>
>>>
>
> _________________________________________________________________
> The idiot box is no longer passe!
> http://content.msn.co.in/Entertainment/TV/Default.aspx
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:44:50 CST