Multiwfn forum

Multiwfn official website: http://sobereva.com/multiwfn. Multiwfn forum in Chinese: http://bbs.keinsci.com/wfn

You are not logged in.

#1 2023-04-29 21:42:25

ucb-mgcf
Member
Registered: 2023-04-29
Posts: 2

reading in .wfn files produced by q-chem

Hi,

First, I want to say that I work at a computational facility at UC-Berkeley and several of our users use Multiwfn; it's a great program, and thank you for creating it.

We came across a problem with using .wfn files produced by q-chem.  I've isolated one example that works and one that doesn't, and I've determined that the problem is in parsing the 2nd line of the .wfn file.  Here are the 2nd lines of each of the files:

GAUSSIAN             34 MOL ORBITALS    176 PRIMITIVES        4 NUCLEI
GAUSSIAN             88 MOL ORBITALS    1410 PRIMITIVES       40 NUCLEI

I observed that the number of primitives begins at exactly the same character in each case.  I tried deleting one space to the left of the 1410 in the 2nd file, so that now all the words in the line are exactly aligned, and now both files are correctly read in to multiwfn.  So evidently multiwfn is expecting a fixed width input format for the components of this line.  I'm posting this in the forum so that this can either be fixed in multiwfn, or if you think the problem is instead that q-chem should be writing this line in fixed width format, let me know, and I'll pass that information on to q-chem.

thanks

Offline

#2 2023-04-30 03:45:49

sobereva
Tian Lu (Multiwfn developer)
From: Beijing
Registered: 2017-09-11
Posts: 1,830
Website

Re: reading in .wfn files produced by q-chem

I've updated Multiwfn today on Multiwfn website, now it reads this line in free format.

Offline

#3 2023-04-30 21:51:47

ucb-mgcf
Member
Registered: 2023-04-29
Posts: 2

Re: reading in .wfn files produced by q-chem

Thanks!  It works now for both wfn files.  And as a happy side note, I wanted to let you know that I proceeded to use the same option choices to check the AIM charges for the larger complex using the new version and the version we had previously been using (3.7), and the result for the new version is more numerically accurate:

3.8dev:  Sum of above integrals:        236.00022435
3.7:  Sum of above integrals:        235.91120789

best regards

Offline

#4 2023-05-01 08:57:17

sobereva
Tian Lu (Multiwfn developer)
From: Beijing
Registered: 2017-09-11
Posts: 1,830
Website

Re: reading in .wfn files produced by q-chem

Yes, 3.8(dev) fixed a relevant problem in 3.7, you can find corresponding information in update history page:

•[2021-Sep-1] (IMPORTANT) The result of AIM basin integration in basin analysis module is slightly inaccurate, this bug was introduced since Multiwfn 3.5.

Offline

Board footer

Powered by FluxBB