Can't recognize real k-points in Crystal converter

[JaeHyung Yu]

Hi,

I am currently using the CRYSTAL converter of recent package, and I found that crystal converter cannot recognize real k-points in the output file.

However, when I use converter of qwalk 0.97.1, it could recognize real k-points that I have in the file.

So I looked into the crystal2qmc.cpp file, the code is not really changed.

Did anyone have experience on this thing?

Thank you!

[Lucas Wagner]

Hi Jae,

It works for me. Could you post the output, an explanation of what exactly you did, and the output of crystal2qmc?

Lucas

--
You received this message because you are subscribed to the Google Groups "QWalk user's group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to qwalk-users...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[JaeHyung Yu]

Hi Lucas, 

Unfortunately, because output file size is huge (around 1.7 Gb), I cannot post the output. 

What I am doing is ZnO with N defect calculation, with Trail-Needs ECP. 

My converting procedure is :

(1) Because of large exponents in Trail-Needs ECP, the ECP information in CRYSTAL output is messed up - So I manually change the ECP informations.

(2) Then I tried to convert my output file.

    i) When I use latest version crystal2qmc converter, I only get 2 outputs (output.basis & output.jast2)

       And I get this message:

found basis    ATOM   X(AU)   Y(AU)   Z(AU)  N. TYPE  EXPONENT  S COEF   P COEF   D/F/G COEF

first line    1 ZN  -1.740  -3.013  -0.052

Found ZN : effective charge 12

Found O : effective charge 6

Found N : effective charge 5

Detected a ROHF or UHF calculation.

N_up-N_down is found to be 1

Found 0 k-points with real eigenvectors 

minimum exponent 0.1402

cutoff length 11.4625

Writing real orbitals ...  

      ii) When I used converter in qwalk 0.97.1 package, I got this message : 

Detected a ROHF or UHF calculation.  What is the spin state?(1=singlet, 2=doublet...)

2

Found 6 k-points with real eigenvectors 

0 :    1 (  0  0  0)   position 130485

1 :    3 (  2  0  0)   position 65195043

2 :    7 (  0  2  0)   position 217012333

3 :   18 (  0  0  2)   position 672464185

4 :   20 (  2  0  2)   position 737528743

5 :   24 (  0  2  2)   position 889346033

Please choose a point[0-5]:

And I confirmed that those points are corresponding real k-points in my system.

I tried to get output for all kpoints mentioned above, and converter gave me all files : *basis, *jast2, *orb, *slater, and *sys files.

2015년 12월 2일 수요일 오후 4시 7분 59초 UTC-6, Lucas 님의 말:

[Lucas Wagner]

Hi Jae,

It's very difficult to debug problems without more information than this. Try to think about it from my perspective; I don't know really anything about what you're doing.

Questions:

 0) Can you really not at least give me the input files?

 1) Are you using properties or crystal to print out the orbitals? 

 2) Try a smaller system. Does the same problem occur?

 3) If you just print out the gamma point does the problem occur?

 4) How many k-points are you printing out?

 5) Does the new version crash? Is it a segfault?

Best,

Lucas