[CP2K:1935] Re: Lagrange multipliers

[Laino Teodoro]

This is an option to consider.. maybe in the future.
Thanks for your suggestions.
cheers,
Teo

On 23 Mar 2009, at 07:53, ivaylo wrote:

>
> Hi Teo,
> Thanks for the quick reply. Scripting may help but eventually it may
> be best to provide the option to selectively turn off the printing
> from cp2k. Since there is no compression the file size can quickly get
> out of hand. For instance five steps on a small (~70,000 atom QMMM
> system) produces a ~10M file.  A typical 6ps run at this rate would
> create a 24G file (and 20-50ps runs would be quite unmanagable).
>
> Cheers,
> Ivaylo
>
> On Mar 22, 11:08 pm, Laino Teodoro <teodor... at gmail.com> wrote:
>> Hi,
>>
>> the way it is now, you get always the Lagrange multipliers of all
>> constraints.
>> The order in the file is: first all internal (collective, 3x3, 4x6)
>> then all external (collective,
>> 3x3, 4x6).
>> WIth some scripting you should be able to get the numbers you want.
>>   cheers,
>> Teo
>>
>> On 23 Mar 2009, at 06:26, ivaylo wrote:
>>
>>
>>
>>
>>
>>> Hello all,
>>> I encountered the following problem while trying out constrained MD
>>> with cp2k (QMMM).
>>
>>>   &CONSTRAINT
>>>     &G3X3
>>>       EXCLUDE_QM
>>>       DISTANCES 1.8896447 1.8896447 3.0862864
>>>       MOLNAME MOL6  # these constraints need to be enforced to  
>>> keep MM
>>> water rigid
>>>       ATOMS 1 2 3
>>>     &END G3X3
>>>     CONSTRAINT_INIT T
>>>     &COLLECTIVE
>>>      COLVAR 1  # this is the colvar for which I want to print the
>>> lagrange multiplier
>>>      INTERMOLECULAR T
>>>      TARGET [angstrom] 0.0
>>>     &END COLLECTIVE
>>>     &LAGRANGE_MULTIPLIERS ON
>>>      ADD_LAST NUMERIC
>>>      COMMON_ITERATION_LEVELS 26000
>>>      FILENAME ./CONSTRAINT
>>>     &END LAGRANGE_MULTIPLIERS
>>>   &END
>>
>>> The problem is that when  LAGRANGE_MULTIPLIERS  is set to ON all
>>> values (MM water included) are dumped into the ./CONSTRAINT file. Is
>>> there a way to print only the lagrange multiplier for the collective
>>> variable of interest? Any ideas would be appreciated.
>>
>>> Cheers,
>>> Ivaylo- Hide quoted text -
>>
>> - Show quoted text -
> >