Super-large coordinate values in bader analysis output in 6.5.1

Problems running VASP: crashes, internal errors, "wrong" results.


Moderators: Global Moderator, Moderator

Locked
Message
Author
yujia_teng
Newbie
Newbie
Posts: 49
Joined: Thu May 25, 2023 6:24 pm

Super-large coordinate values in bader analysis output in 6.5.1

#1 Post by yujia_teng » Wed Dec 17, 2025 2:54 am

I found that the coordinate in ACF.dat file from bader charge analysis is super-large in VASP 6.5.1. So then I checked with previous versions, 5.4.4 and 6.4.1. There's no problem there.

So I doubt it's because something is changed in v6.5.1 caused this problem. The input/output are attached. The inputs are exactly same but just computed in different version.

The header of the file from VASP 6.5.1 result.

Code: Select all

 #         X           Y           Z       CHARGE      MIN DIST   ATOMIC VOL
 --------------------------------------------------------------------------------
    1  -61.546380  108.970358 2250.743269   12.029834     1.126935    22.828261
    2 -141.556694  249.921189  976.694269   11.913088     1.149931    18.854765
You do not have the required permissions to view the files attached to this post.

alexey.tal
Global Moderator
Global Moderator
Posts: 510
Joined: Mon Sep 13, 2021 12:45 pm

Re: Super-large coordinate values in bader analysis output in 6.5.1

#2 Post by alexey.tal » Fri Dec 19, 2025 9:55 am

Dear yujia_teng,

Indeed, I was able to reproduce this issue. The problem is caused by a slight change in the formatting of the CHGCAR files in VASP 6.5.0: an extra space was added before the atomic positions. As a result, the issue can be easily avoided by removing this extra space in front of the atomic positions (see the attached example).

In principle, we do not develop the Bader code, and therefore we cannot guarantee its compatibility with every VASP version. I inspected the CHGCAR parsing in the Bader code, and it indeed assumes a very specific file format and does not allow for much flexibility. Given this, I am not surprised that the compatibility turned out to be rather fragile.

If you would like, I can provide a patch for the Bader code to make it compatible with VASP 6.5.0 and later. However, in practice it is probably much simpler to manually edit the CHGCAR file.

I will also discuss with my colleagues whether this compatibility issue should be addressed on our end.

Best regards,
Alexey

You do not have the required permissions to view the files attached to this post.

yujia_teng
Newbie
Newbie
Posts: 49
Joined: Thu May 25, 2023 6:24 pm

Re: Super-large coordinate values in bader analysis output in 6.5.1

#3 Post by yujia_teng » Mon Dec 22, 2025 4:36 am

Dear Alexey,
Thanks for identifying the issue. I deleted the extra space and it works. From my side, I can just remove the extra space manually or write a simple script to automatically do this each time. So I think a patch is not needed for this.

I also want to mention another format change here. There's no problem, and it doesn't affect anything. I just noticed it. The spacing between atom names in CONTCAR are larger in 6.5.1, probably this also starts from 6.5.0.

Code: Select all

 6.5.1: In              Se              C
 5.4.4/6.4.1:  In   Se   C

Best,
Yujia

Last edited by yujia_teng on Mon Dec 22, 2025 4:37 am, edited 1 time in total.

Locked