Re: reinitatoms

From: Brian Radak (brian.radak_at_gmail.com)
Date: Fri Feb 08 2019 - 07:51:25 CST

Hi Laura,

I'm still a bit confused by your sequence of commands - you want to load a
psf and coordinates, but randomly initialize velocities. Then, before
running, you use reinitatoms to load in a new coor/vel/xsc file?

It would make more sense to me (but perhaps I'm completely missing the
point) to load the files you want and then use reinitvels to randomize
velocities at a later point rather than reloading them from file. I believe
you can "stash" the initial velocities, so to speak, with the checkpoint
command (although I'll admit I haven't used this much). That way you can
get the velocities from file at startup and still have the ability to
randomize things.

HTH,
Brian

On Fri, Feb 8, 2019 at 4:39 AM Laura Lopes <laurajoanalopes_at_gmail.com>
wrote:

> Dear all,
>
> I just encountered the same problem discussed in this e-mail again, so I
> decided to reopen the thread, since I got no explanation to my problem at
> that time.
> To answer Brian, I have a simulation where I need to start from a lot of
> different points, decided during the simulation. Because I need to make a
> decision before even running my first trajectory, I provide to NAMD a
> temperature parameter, otherwise it gives me an error.
> Of course I solved this issue by simply giving namdbin files, that are
> useless to me, instead of the temperature command. But I really just wanted
> to understand what is the origin of this error !
> Best,
>
> Laura
>
> On Fri, Jan 12, 2018 at 9:20 PM Brian Radak <brian.radak_at_gmail.com> wrote:
>
>> Why do you need to use reinitatoms here? I would have though that this
>> would only be useful after having run dynamics for some period of time.
>>
>> Brian
>>
>> On Thu, Jan 11, 2018 at 11:36 AM, Laura Joana <laurajoanalopes_at_gmail.com>
>> wrote:
>>
>>> Dear all again,
>>>
>>> Sorry for the two e-mails, but I decided to explore the issue a bit more.
>>> Looking at the NAMD 2.12 release features I guess the problem is in this
>>> sentence:
>>> "The patch grid and other simulation paramters are *not* changed, so
>>> the dimensions of the new and old structures must be compatible."
>>> (under Ability to reload molecular structure without restarting).
>>> Making a diff with my two logfiles I get this:
>>>
>>> < Info: PATCH GRID IS 3 (PERIODIC) BY 3 (PERIODIC) BY 1 (PERIODIC)
>>> < Info: PATCH GRID IS 2-AWAY BY 2-AWAY BY 1-AWAY
>>> < Info: Reading from binary file
>>> /home/silva-ll/cyclodextrine/ic/cycloA_A.vel
>>> < Info: REMOVING COM VELOCITY -0.0432567 0.00241296 0.0339374
>>> < Info: LARGEST PATCH (0) HAS 417 ATOMS
>>> ---
>>> > Info: PATCH GRID IS 4 (PERIODIC) BY 2 (PERIODIC) BY 2 (PERIODIC)
>>> > Info: PATCH GRID IS 2-AWAY BY 1-AWAY BY 1-AWAY
>>> > Info: REMOVING COM VELOCITY 0.087986 -0.125532 0.0988174
>>> > Info: LARGEST PATCH (3) HAS 237 ATOMS
>>>
>>> Does anyone know if I'm in the good road to find out what is the problem
>>> ?
>>> Best regards,
>>>
>>> Laura
>>>
>>>
>>> On Thu, Jan 11, 2018 at 3:26 PM, Laura Joana <laurajoanalopes_at_gmail.com>
>>> wrote:
>>>
>>>> Dear all,
>>>>
>>>> I'm having a problem involving the use of the command *reinitatoms
>>>> file*.
>>>> I run two simulations, almost the same configuration file, the only
>>>> difference is that for the first one I call *bincoordinates,
>>>> binvelocities and extendedsystem* with some checkpoint and the other I
>>>> use the *temperature* command. For both I call *reinitatoms file* (some
>>>> other checkpoint) before any run command. I expected no difference between
>>>> those simulations. The problem is that the first one runs normally and the
>>>> second give me an error and do not even run ! The error is "Periodic cell
>>>> has become too small for original patch grid".
>>>> So the question is: what is the difference in using *reinitatoms file*
>>>> after *bincoordinates, binvelocities and extendedsystem* and after the
>>>> *temperature* command ? Shouldn't NAMD just read *file.coor, file.vel
>>>> and file.xsc* and ignore the other previous commands ?
>>>> Best regards,
>>>>
>>>> Laura
>>>>
>>>
>>>
>>

This archive was generated by hypermail 2.1.6 : Sat Dec 07 2019 - 23:20:25 CST