[CP2K-user] KS matrix in real space

[Peng Gao]

Sorry, miss-spelling, DFT%PRINT%S_CSR_WRITE%REAL_SPACE is 
DFT%PRINT%KS_CSR_WRITE%REAL_SPACE.

On Wednesday, June 23, 2021 at 9:16:39 AM UTC+8 Peng Gao wrote:

> Hello, everyone!
> I am trying to output KS matrix in real_space, but the matrixes' 
> difference is very small with and without DFT%PRINT%S_CSR_WRITE%REAL_SPACE 
> as follows:
>
> K_space             Real_space          Difference
> 1  1  -9.50E-01    1  1  -9.50E-01    -7.50E-06
> 1  2  6.72E-01     1  2  6.72E-01      3.77E-06
> 1  3  1.03E-02     1  3  1.03E-02     -4.20E-07
> 1  4  7.04E-02     1  4  7.04E-02      6.81E-07
> 1  5  -6.85E-02    1  5  -6.85E-02    -4.62E-07
> 1  6  -7.22E-03    1  6  -7.22E-03     5.91E-07
> 1  7  -6.82E-02    1  7  -6.82E-02    -9.12E-07
> 1  8  6.49E-02     1  8  6.49E-02      5.68E-07
> 1  9  -5.21E-02    1  9  -5.21E-02    -3.84E-08
> 1  10  -6.97E-02   1  10  -6.97E-02   -7.87E-07
>
> I tried in CP2k 8.2 and trunk versions and got the same result. Did I miss 
> something to get the correct KS matrix in real space? 
> BTW, are the basis in the spherical molden file and KS matrix in the same 
> order? Both are generated in CP2K.
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20210622/bed98b8c/attachment.htm>