By using iwrap, you mean set iwrap=1 in your input file? If you use
iwrap, this problem should not occur, since particles are not allowed
to diffuse to such far locations... The only problem I can see with
this is if you don't have periodic boundary conditions (i.e. ntb=0),
and if you have free particles (like ions, for instance) that are
diffusing away from the rest of your system (since even with implicit
solvent/gas phase, the COM translation is eliminated every X number of
steps according to nscm.
All the best,
Jason
2010/4/8 Yunfen Wang <wangfen041237.163.com>:
> Jason，
> I have tried that before ,but when running to the same ''ns'', it met the same trouble ! And also , I have used ''iwrap'' in my input file
> Thank you
> Wang
>
>
>
>
> 在2010-04-08 12:03:31，"Jason Swails" <jason.swails.gmail.com> 写道：
>>2010/4/7 Yunfen Wang <wangfen041237.163.com>:
>>> Dear Jason,
>>> Thank you for your last instructions . Now , I have found the **** in the restart file ,and deleted the following lines from there . I re-run simulation , a new problem has appeared . In the Output there has another error:
>>> | Flags:
>>> getting new box info from bottom of inpcrd
>>> | INFO: Old style inpcrd file read
>>> | peek_ewald_inpcrd: Should not be here
>>> What's wrong? Should I delete content in other files while delete the **** in the restart file ?
>>
>>You can't simply delete the lines with **** in your restart file. The
>>existence of a **** means that one of the coordinates for one of the
>>atoms is simply unknown, and you can't use the restart file at all.
>>You have to go back to a restart file that has no ***** in place of
>>coordinates (a different file!). (When you delete ****, the program
>>finds fewer coordinates than it is expecting to).
>>
>>All the best,
>>Jason
>>
>>> Expect your further instructions !
>>> Thank you
>>> Wang
>>>
>>>
>>>
>>> 在2010-04-07 22:00:05，"Jason Swails" <jason.swails.gmail.com> 写道：
>>>>Hello,
>>>>
>>>>Check your restart file for *****. This indicates that one of the
>>>>coordinates has become too large for the fixed-format form of the
>>>>restart file. If this is the case, you'll have to use your last
>>>>working restart file and re-run simulations from there. To avoid this
>>>>problem, use 'iwrap' in your input file. Details about this variable
>>>>can be found extensively in the archives as well as detailed in the
>>>>manual.
>>>>
>>>>Good luck!
>>>>Jason
>>>>
>>>>On Wed, Apr 7, 2010 at 9:56 AM, Yunfen Wang <wangfen041237.163.com> wrote:
>>>>> Dear users,
>>>>> When I run MD simulations, I often meet the follow Error and the simulation is stopped.
>>>>> The error : Sander bomb in subroutine nonbond_list
>>>>> volume of uncell too big ,too many subcells list grid memory needs to be reallocated ,restart sander
>>>>> when I meet it , I restart the input file and the running is continued But this time , I restart , it reminds the follow error and break the simulation
>>>>> The error :
>>>>> At line 2441 of file -ew -setup.f (unit9 "***.rst")
>>>>> Traceback: not available , complile with
>>>>> - ftrace = frame or - ftrace = full
>>>>> Fortran runtime error :bad value duiring floating point read
>>>>> I didn't know why . Is there anybody meeting the similar cituation ? Can give me advice ?
>>>>> Thank you in advance
>>>>> Reminding that I have checked the software and maked test , there was nothing wrong .
>>>>> Looking forward for your suggestions and instruction.
>>>>> sincerly,
>>>>> Wang
>>>>>
>>>>> _______________________________________________
>>>>> AMBER mailing list
>>>>> AMBER.ambermd.org
>>>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>>>
>>>>
>>>>
>>>>
>>>>--
>>>>---------------------------------------
>>>>Jason M. Swails
>>>>Quantum Theory Project,
>>>>University of Florida
>>>>Ph.D. Graduate Student
>>>>352-392-4032
>>>>
>>>>_______________________________________________
>>>>AMBER mailing list
>>>>AMBER.ambermd.org
>>>>http://lists.ambermd.org/mailman/listinfo/amber
>>>
>>> _______________________________________________
>>> AMBER mailing list
>>> AMBER.ambermd.org
>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>
>>>
>>
>>
>>
>>--
>>---------------------------------------
>>Jason M. Swails
>>Quantum Theory Project,
>>University of Florida
>>Ph.D. Graduate Student
>>352-392-4032
>>
>>_______________________________________________
>>AMBER mailing list
>>AMBER.ambermd.org
>>http://lists.ambermd.org/mailman/listinfo/amber
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
>
--
---------------------------------------
Jason M. Swails
Quantum Theory Project,
University of Florida
Ph.D. Graduate Student
352-392-4032
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Apr 08 2010 - 06:00:04 PDT
