Page 1 of 2

Quantum ESPRESSO 6.4 and EPW 5.1

Posted: Tue Mar 19, 2019 2:06 pm
by sponce
Dear all,

The new version of QE 6.4 and EPW 5.1 have been released. You can find information on how to install it at http://epw.org.uk/Main/DownloadAndInstall

New development in EPW 5.1 includes:
* Support for USPP
* Improvements to the cumulant spectral functions (computed with both convolution and FFT)
* New Wigner-Seitz cell construction
* Speed and memory optimizations for mobility calculations
* k-point symmetry support for homogeneous BTE
* Linear mixing with BTE
* Introduction of a "qwindow" to reduce the required number of q-points to be calculated
* Load balancing on active k-points among cores.
* Indirect absorption corrections while reading external eigenvalues
* Optimizations in the case of ultra-dense grids
* Exact velocity calculations with vme (read .bvec from new w90)
* Various modularization of the code.
* Removed input variables:
- time_max
* New input variables:
- use_ws
- vme variable now working

Important notice: the examples in EPW/examples/ refer to the examples for the tutorials (see EPW website). Those examples are not always up to date as they are not tested automatically.
Therefore I strongly recommand to use the examples provide in the QE/test-suite/epw_*
Those examples are automatically tested and should therefore always be working with the current version and a benchmark reference is provided.

Known issue: There is a problem with the EPW/bin/pp.py to convert the dyn files from text to .xml.
Therefore if you make a phonon calculation with SOC, make sure to put .xml in the input of the phonon code like:
fildyn = 'si.dyn.xml',

This will force the phonon code to produce .xml files.

Happy EPWing !

Re: Quantum ESPRESSO 6.4 and EPW 5.1

Posted: Fri Mar 22, 2019 3:35 pm
by ilias-serifi
Hello Sir
i get a problem on the install of the newest version

Code: Select all

All done. ERROR: only 72 out of 73 tests passed.
Failed test in:
   /home/lphe-ms/qe_release_6.4/test-suite/epw_mob_ibte_sym/
Makefile:71: la recette pour la cible run-tests-epw-parallel a choue
make: *** [run-tests-epw-parallel] Erreur 1



and when i run the epw i get

Code: Select all

Parallel version (MPI), running on     4 processors

     MPI processes distributed on     1 nodes
     K-points division:     npool     =       4

 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     Error in routine epw_readin (5010):
     reading input_epw namelist
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

     stopping ...
application called MPI_Abort(MPI_COMM_WORLD, 1) - process 0



such i can run the same input on the old version and i didn't have any problem


Best Regard

Re: Quantum ESPRESSO 6.4 and EPW 5.1

Posted: Fri Mar 22, 2019 7:41 pm
by hpaudya1
Hi,

This looks like a problem in running the code. Your input file is not correct for the new version of the code. If you have used the same input (which runs fine) from the old version, you may have a input flag that exists in the old version and which may not exist in the new version. For example, If I remember correctly, parallel_q does not exist in the new version.

Best,
Hari Paudyal
SUNY-Binghamton

Re: Quantum ESPRESSO 6.4 and EPW 5.1

Posted: Tue Mar 26, 2019 11:26 am
by ilias-serifi
Hi


Thanks a lot it's working well now

Best Regards

Re: Quantum ESPRESSO 6.4 and EPW 5.1

Posted: Mon Nov 25, 2019 9:45 am
by Muhammad Talha
Hi ,
after

Code: Select all

tar -xvf qe-6.4_release_pack.tgz && cd qe_release_6.4 && ./configure

i got this error
gzip: stdin: unexpected end of file
tar: Unexpected EOF in archive
tar: Unexpected EOF in archive
tar: Error is not recoverable: exiting now

regards
muhammad talha

Re: Quantum ESPRESSO 6.4 and EPW 5.1

Posted: Fri Nov 20, 2020 3:40 pm
by Sylvester10
I am trying to compile epw but I get this error when i run $make, Kindly assist

=====================================================
gzip: ../archive/v3.1.0: unexpected end of file
tar: Unexpected EOF in archive
tar: Unexpected EOF in archive
tar: Error is not recoverable: exiting now
*** Unable to download https://codeload.github.com/wannier-dev ... .gz/v3.1.0.
*** Verify that the url is correct.
make[3]: *** [plugins_makefile:96: uncompress-w90] Error 1
make[3]: Leaving directory '/home/syl2/Applications/q-e-qe-6.6/install'
make[2]: *** [Makefile:250: w90] Error 1
make[2]: Leaving directory '/home/syl2/Applications/q-e-qe-6.6'
make[2]: Entering directory '/home/syl2/Applications/q-e-qe-6.6/wannier90-3.1.0'
mkdir: cannot create directory ‘src/obj’: No such file or directory
make[2]: *** [Makefile:211: objdir] Error 1
make[2]: Leaving directory '/home/syl2/Applications/q-e-qe-6.6/wannier90-3.1.0'
make[1]: *** [Makefile:108: wannier] Error 2
make[1]: Leaving directory '/home/syl2/Applications/q-e-qe-6.6/EPW/src'
make: *** [Makefile:7: epw] Error 2

Re: Quantum ESPRESSO 6.4 and EPW 5.1

Posted: Fri Nov 20, 2020 3:59 pm
by hlee
Dear Sylvester10:

First, I would like to kindly ask you in the future to post the question separately in the corresponding board instead of posting it as replies to the announcements.

Returning to your issue, as the error message says, for unknown reasons you can't download the Wannier90 code from https://codeload.github.com/wannier-dev ... .gz/v3.1.0 .

Sincerely,

H. Lee

Re: Quantum ESPRESSO 6.4 and EPW 5.1

Posted: Mon Nov 23, 2020 8:24 am
by qiaoqiao66
Dear H. Lee,
I am trying to use the EPW 5.1 or 5.2 with QE version 6.4 or 6.5. I've managed to compile QE with EPW successfully (There were no errors reported during the installation), but when I run the epw.x, the program stopped and the following error appear in epw.out.

= BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES
= PID 11747 RUNNING AT cu45
= EXIT CODE: 6
= CLEANING UP REMAINING PROCESSES
= YOU CAN IGNORE THE BELOW CLEANUP MESSAGES

Under the same cluster account, use the 6.3 version to run the test suite/epw examples and my calculation system can successfully finish calculate, but use the 6.4 version to run the test suite/epw examples, there will be running crashes. Besides, testing the examples in suite/epw with version 6.5 worked fine, but running my calculation system the run crashed with the same error. So it should not be the cause of insufficient memory.
This problem has puzzle me for a long, long time, and I am looking forward to your reply!

With kind regards,
Zhi Yuan

Re: Quantum ESPRESSO 6.4 and EPW 5.1

Posted: Mon Nov 23, 2020 8:32 am
by qiaoqiao66
Dear H. Lee,
I am trying to use the EPW 5.1 or 5.2 with QE version 6.4 or 6.5. I've managed to compile QE with EPW successfully (There were no errors reported during the installation), but when I run the epw.x, the program stopped and the following error appear in epw.out.

= BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES
= PID 11747 RUNNING AT cu45
= EXIT CODE: 6
= CLEANING UP REMAINING PROCESSES
= YOU CAN IGNORE THE BELOW CLEANUP MESSAGES

Under the same cluster account, use the 6.3 version to run the test suite/epw examples and my calculation system can successfully finish calculate, but use the 6.4 version to run the test suite/epw examples, there will be running crashes. Besides, testing the examples in suite/epw with version 6.5 worked fine, but running my calculation system the run crashed with the same error. So it should not be the cause of insufficient memory.
This problem has puzzle me for a long, long time, and I am looking forward to your reply!

With kind regards,
Zhi Yuan

Re: Quantum ESPRESSO 6.4 and EPW 5.1

Posted: Mon Nov 23, 2020 10:16 pm
by hlee
Dear Zhi Yuan:
Dear Sylvester10:

First, I would like to kindly ask you in the future to post the question separately in the corresponding board instead of posting it as replies to the announcements.
As I already mentioned in the thread above of this post, you should post the question separately in the corresponding board instead of posting it as replies to the announcements.

Also I can't give you the clear answer to your question unless you provide with me enough information; I would suggest you to use the most recent version of EPW with QE and start from the scratch.

Sincerely,

H. Lee