Re: [Wien] Install Wien2k using oneAPI

[Peter Blaha]

I just installed oneAPI today  (it is free !!!). You need the Base and 
the HPC (high performance computing) toolkit.


It was trivial to install, but the setting of the environment is 
slightly different.



Once installed, WIEN2k compiles with the standard option (ifort+mkl) 
without any changes.



Am 11.03.2021 um 16:52 schrieb Laurence Marks:
Since nobody has responded, I will. I have oneAPI running without any 
issue for some months. Of course sometimes the installation is tricky. 
I suggest going ahead, then coming back to the list if there are any 
problems.


On Wed, Mar 10, 2021 at 12:38 AM Riyajul Islam > wrote:


Dear Wien2k users,
I have used Intel parallel studio for intel fortran compiler to
install Wien2k 19.2 but the license has expired. As they have
recently transitioned from Intel Parallel Studio XE to Intel
oneAPI, so is there any way that I can install Wien2k using Intel
oneAPI?

Kind regards
-- 
Riyajul Islam

National Institute of Technology Nagaland

___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at


https://urldefense.com/v3/__http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien__;!!Dq0X2DkFhyF93HkjWTBQKhk!FKkETL-5jLbLm0qQ0ojSzxyA70Chd3yLLZAlh9Yi1WoJ-qZ4aVzufokABoC7A10Jr6LDlw$



SEARCH the MAILING-LIST at:

https://urldefense.com/v3/__http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html__;!!Dq0X2DkFhyF93HkjWTBQKhk!FKkETL-5jLbLm0qQ0ojSzxyA70Chd3yLLZAlh9Yi1WoJ-qZ4aVzufokABoC7A12B24_idA$






--
Professor Laurence Marks
Department of Materials Science and Engineering
Northwestern University
www.numis.northwestern.edu 
"Research is to see what everybody else has seen, and to think what 
nobody else has thought" Albert Szent-Györgyi


___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html


--
---
Peter Blaha,  Inst. f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-158801165300
Email: peter.bl...@tuwien.ac.at
WWW:   http://www.imc.tuwien.ac.at  WIEN2k: http://www.wien2k.at
-

___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] Install Wien2k using oneAPI

[Laurence Marks]

Since nobody has responded, I will. I have oneAPI running without any issue
for some months. Of course sometimes the installation is tricky. I suggest
going ahead, then coming back to the list if there are any problems.

On Wed, Mar 10, 2021 at 12:38 AM Riyajul Islam  wrote:

> Dear Wien2k users,
> I have used Intel parallel studio for intel fortran compiler to install
> Wien2k 19.2 but the license has expired. As they have recently transitioned
> from Intel Parallel Studio XE to Intel oneAPI, so is there any way that I
> can install Wien2k using Intel oneAPI?
>
> Kind regards
> --
> Riyajul Islam
> National Institute of Technology Nagaland
>
> ___
> Wien mailing list
> Wien@zeus.theochem.tuwien.ac.at
>
> https://urldefense.com/v3/__http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien__;!!Dq0X2DkFhyF93HkjWTBQKhk!FKkETL-5jLbLm0qQ0ojSzxyA70Chd3yLLZAlh9Yi1WoJ-qZ4aVzufokABoC7A10Jr6LDlw$
> SEARCH the MAILING-LIST at:
> https://urldefense.com/v3/__http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html__;!!Dq0X2DkFhyF93HkjWTBQKhk!FKkETL-5jLbLm0qQ0ojSzxyA70Chd3yLLZAlh9Yi1WoJ-qZ4aVzufokABoC7A12B24_idA$
>


-- 
Professor Laurence Marks
Department of Materials Science and Engineering
Northwestern University
www.numis.northwestern.edu
"Research is to see what everybody else has seen, and to think what nobody
else has thought" Albert Szent-Györgyi
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] opticcpara crashed for mstar

[Dr. K. C. Bhamu]

Dear Prof. Oleg
Thank you for your patience for my query.
Here is my script to calculate mstar [1] from a successfully finished mstar
calculation.

This is for Si only. For other systems, some parameters need to be tuned.
I am not a good programmer, so the script may not look in good format.

[1]. https://we.tl/t-SYldxNNner


Thank you
Bhamu


On Thu, Mar 11, 2021 at 4:26 AM Rubel, Oleg  wrote:

> I am glad Si worked out. Sorry, Bhamu. You need to figure it out even if
> it takes more time. When searching through eigenvalues, keep in mind the
> number of digits. For instance, 0.383706 from case.scf can appear as
> 0.3837058 in the eigenvalue file.
>
> All the best
> Oleg
>
> 
> From: Wien  on behalf of Dr. K.
> C. Bhamu 
> Sent: Wednesday, March 10, 2021 11:50
> To: A Mailing list for WIEN2k users
> Subject: Re: [Wien] opticcpara crashed for mstar
>
> Dear Prof. Oleg
> I tried to follow your advice and it worked for Si case. I found VBM at
> #KP1 and CBM at #KP15. The band index edges were matched with the
> particular KP in case.energyso.
>
> But In my case, I could not find the VBM/CBM edges either in case.eneryso
> or in case.bands.agr/spagetti_ene.
>
> I finished two different calculations (with shifted k-mesh and unshifted
> k-mesh) and both can be downloaded from [1].
> [1] https://we.tl/t-Mhv6nv4Zlf
>
> I would be grateful to you, if you can suggest me #KP's with respect to
> VBM/CBM.
> CBM is not exactly on high-symmetry  k-point.
>
>
> Thank you
> Bhamu
>
>
> On Tue, Mar 9, 2021 at 11:13 PM Rubel, Oleg  rub...@mcmaster.ca>> wrote:
> Dear Bhamu,
>
> to get a VBM k-point index, you would need to look for "0.383706" in your
> case.energyso and see which k-point this eigenvalue belongs to. For CBE,
> look for "0.401998" (it can be a different k-point).
>
> In Si VBM is at Gamma, and CBM is far away from Gamma. Thus we have two
> different k-points. After some trials, I found that 7x7x7 unshifted k-mesh
> "hits" CBM of Si quite precisely.
>
> Of course you need to be confident that the selected k-mesh "hits" the
> relevant extremum (extrema) of your band structure. For instance if you
> need Gamma, do not select a shifted mesh. When the extremum is away from
> the high-symmetry points and coordinates are known, it might be easier to
> use case.klist_band and target the point(s) of interest.
>
> I hope it helps
> Oleg
>
> 
> From: Wien  wien-boun...@zeus.theochem.tuwien.ac.at>> on behalf of Dr. K. C. Bhamu <
> kcbham...@gmail.com>
> Sent: Tuesday, March 9, 2021 08:15
> To: A Mailing list for WIEN2k users
> Subject: Re: [Wien] opticcpara crashed for mstar
>
> Dear Prof. Oleg
> Sorry for the late reply. As we have the opposite time zone so I am
> replying according to your office time.
>
> My previous information was from pbe.spaghetti_ene.
>
> Here is the relevant part from case.scf.
>
>  Bandranges (emin - emax) and occupancy:
> :BAN00066:  660.2189000.284902  1.
> :BAN00067:  670.2817550.322256  1.
> :BAN00068:  680.2817550.322256  1.
> :BAN00069:  690.2879140.335582  1.
> :BAN00070:  700.2879140.335582  1.
> :BAN00071:  710.2941570.351734  1.
> :BAN00072:  720.2941570.351734  1.
> :BAN00073:  730.3010710.368987  1.
> :BAN00074:  740.3010710.368987  1.
> :BAN00075:  750.3400590.383706  1.
> :BAN00076:  760.3400590.383706  1.  VBM
> :BAN00077:  770.4019980.551266  0.  CBM
> :BAN00078:  780.4019980.551266  0.
> :BAN00079:  790.4728340.603663  0.
> :BAN00080:  800.4728340.603663  0.
> :BAN00081:  810.5567370.634231  0.
> Energy to separate low and high energystates:   -0.57035
>
> Just above this part, I see
> :KPT   :  NUMBER OF K-POINTS: 40
>0.0   0.0 angle (M,z), angle (M,x) deg
>
> SPIN-ORBIT EIGENVALUES:
>  K=   0.14286   0.14286   0.14286 1  (In case of "Si" it is 0 0 0
> so you took KP =1?)
>   MATRIX SIZE=  552   WEIGHT= 8.00
>
> Does it mean that I should look for #KP 1? and then of course, bandinxed
> 75,76 for VBM and CBM, respectively.
> But in case of Si, you took #KP =1 for VBM and #KP=15 for CBM.
> So now the band index is clear to me.
>
> My next query is, how can I know #KP for VBM and CBM?
>
> Sorry for the long conversation.
>
> Thank you
> Bhamu
>
> On Mon, Mar 8, 2021 at 8:48 PM Rubel, Oleg  rub...@mcmaster.ca>>>
> wrote:
> Dear Bhamu,
>
> it seems you have k-point index=38, band index=653. To be on a safe side,
> I would look for band ranges (":BANXXX") in case.scf (last iteration). The
> occupancies are written down in the same table. If you have questions about
> interpretation of :BANXXX, it will be better if you list