[CP2K-user] [CP2K:11082] comparing CP2K-DFTB (5.1) and DFTB+ (18.2)

[Wei Lai]

I tried the trunk version and results were updated.  Of course efforts have 
to made to converge Ewald energy but it appears that bugs have been fixed!

core energy: -14.7017837002 (dftb+), -14.68632501975257 (cp2k, alpha=0.5), 
-14.70135116830477 (cp2k, alpha=1)
scc energy:  1.2109415266 (dftb+), 1.23879325339532 (cp2k, alpha=0.5), 
1.21153669321739 (cp2k, alpha=1)

Thanks, Wei

On Friday, January 11, 2019 at 10:13:03 AM UTC-5, Wei Lai wrote:
>
> Dear Juerg,
>
> Thanks for the effort in looking into the problems.  In addition, I found 
> that the real-space part of 1/R contribution to gmcharge was done with a 
> neighbor list, while the gamma_matrix contribution to gmcharge was not.  In 
> my opinion they should be treated in the same way (perhaps both a neighbor 
> list).  I'd like to bring this up to your attention.  I look forward to the 
> fixes!
>
> Thanks, Wei
>
> On Friday, January 11, 2019 at 3:37:36 AM UTC-5, jgh wrote:
>>
>> Hi 
>>
>> I looked into these problems and I see that this behavior is 
>> from a combination of bugs and hidden constraints on variables. 
>>
>> The bugs were introduced when the MIC was dropped after the 
>> implementation 
>> of the k-point code. The Ewald parameters are partly set from the 
>> input and partly set by using default values within the code. 
>>
>> To be on the save side I would suggest to only do DFTB calculations 
>> with large enough computational cells. You can use 
>> the MULTIPLE_UNIT_CELL keyword if you have a small cell. 
>>    &TOPOLOGY 
>>       MULTIPLE_UNIT_CELL ${NREP} ${NREP} ${NREP} 
>>     &END 
>>     &CELL 
>>       ABC 4.0 4.0 4.0 
>>       MULTIPLE_UNIT_CELL ${NREP} ${NREP} ${NREP} 
>>     &END CELL 
>>
>> I guess a cell with 10 Angstrom box length should be ok. 
>>
>> Use an ALPHA value that is 1 or larger and adjust GMAX for 
>> your box size to get converged values. 
>>
>> I am working on fixes, they should be ready in the TRUNK version soon 
>> and will be in the next release. 
>>
>> best regards 
>>
>> Juerg 
>>
>> -------------------------------------------------------------- 
>> Juerg Hutter                         Phone : ++41 44 635 4491 
>> Institut für Chemie C                FAX   : ++41 44 635 6838 
>> Universität Zürich                   E-mail: hut... at chem.uzh.ch 
>> Winterthurerstrasse 190 
>> CH-8057 Zürich, Switzerland 
>> --------------------------------------------------------------- 
>>
>> -----cp... at googlegroups.com wrote: ----- 
>> To: "cp2k" <cp... at googlegroups.com> 
>> From: "Maxime Van den Bossche" 
>> Sent by: cp... at googlegroups.com 
>> Date: 12/22/2018 10:45PM 
>> Subject: Re: [CP2K:11082] comparing CP2K-DFTB (5.1) and DFTB+ (18.2) 
>>
>> Hi all, 
>>
>> I also did a quick comparison now between CP2K and DFTB+ 
>> for a simple rocksalt MgO structure (8 atoms in total, 
>> using the OT minimizer in CP2K). The input and output 
>> files for both codes are attached. 
>>
>> What I found so far: 
>>
>> * Without periodicity, the two codes agree quite well 
>>   in terms of the H0 and SCC energies (up to ~1e-4 Ha) 
>>   as well as the Mulliken charges. 
>>
>> * In the periodic case, so far I haven't managed to 
>>   get comparable results with the two codes. The CP2K 
>>   results indeed appear sensitive to the choice of the 
>>   Ewald alpha parameter, similar to what Wei observed. 
>>   For each of the three ALPHA values I tried, the resulting 
>>   energies and SCC charges are quite different from DFTB+. 
>>
>> * In contrast to Wei, though, I also get different results 
>>   for both codes in non-self-consistent calculations 
>>   (but at least here the CP2K results indeed no longer 
>>   depend on ALPHA). 
>>
>> It would be great if you could check whether my CP2K 
>> inputs have been appropriate. If so, I hope this example will 
>> help to get to the bottom of this. 
>>
>> Best, 
>> Maxime   
>>   -- 
>>  You received this message because you are subscribed to the Google 
>> Groups "cp2k" group. 
>>  To unsubscribe from this group and stop receiving emails from it, send 
>> an email to cp2k+... at googlegroups.com. 
>>  To post to this group, send email to cp... at googlegroups.com. 
>>  Visit this group at https://groups.google.com/group/cp2k. 
>>  For more options, visit https://groups.google.com/d/optout. 
>>   
>>
>> [attachment "files.zip" removed by Jürg Hutter/at/UZH] 
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20190114/f450da9e/attachment.htm>