Initial SCF Convergence extreamly slow in QMMM Calculations
권경표
Hello,
I am currently running QMMM calculations with EGFP, but I'm facing a problem where the initial SCF convergence is taking an extremely long time.
I've attached my input (inp) and log files for reference.
Could you please help me identify what might be causing this issue?
Thank you for your assistance.
Jürg Hutter
your output file doesn't show any SCF iterations at all. I assume your QM system
takes too long for you to start the SCF iterations, so you killed the job?
It seems that you don't have a proper restart file
READ RESTART : WARNING : DIFFERENT # AOs 371 217
This would be the first thing to fix.
Next, you are using MOLOPT basis sets with a hybrid functional. I would advice to
use the ADMM method in this case.
A good starting point (e.g. to get a good restart file) is to start with a simple GGA functional
and test if all the settings are working. After that you can restart with the hybrid functional.
regards
JH
________________________________________
From: cp...@googlegroups.com <cp...@googlegroups.com> on behalf of 권경표 <kkp...@gmail.com>
Sent: Thursday, August 22, 2024 7:31 AM
To: cp2k
Subject: [CP2K:20596] Initial SCF Convergence extreamly slow in QMMM Calculations
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+uns...@googlegroups.com<mailto:cp2k+uns...@googlegroups.com>.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/051819e1-fe11-4638-b19a-4fc91004d5e3n%40googlegroups.com<https://groups.google.com/d/msgid/cp2k/051819e1-fe11-4638-b19a-4fc91004d5e3n%40googlegroups.com?utm_medium=email&utm_source=footer>.
권경표
Hello,
Thank you very much for your prompt response.
I wanted to inform you that I had to kill the job after the SCF iteration process continued for about 10 hours without progress.
Upon investigating the issue, I found that the error message you mentioned, "READ RESTART: WARNING: DIFFERENT # AOs 371 217," was indeed the cause of the problem.
It appears that this issue arose because the initial RESTART file was generated using PBE and with the DZVP-MOLOPT-GTH basis set instead of the TZV2P-GTH, leading to extremely slow SCF iteractions.
After generating the restart file using the PBE functional and the TZV2P-GTH basis set, the calculations proceeded at a normal speed.
I appreciate your advice, and I wish you a wonderful day.
Thank you again!
Kyoung Pyo Kwon