Skip to content

Fix eam/alloy potential parsing#1463

Merged
brucefan1983 merged 1 commit intobrucefan1983:masterfrom
mushroomfire:master
Apr 24, 2026
Merged

Fix eam/alloy potential parsing#1463
brucefan1983 merged 1 commit intobrucefan1983:masterfrom
mushroomfire:master

Conversation

@mushroomfire
Copy link
Copy Markdown
Contributor

This PR fixes a bug in the eam_alloy potential file parsing. To ensure closer consistency with LAMMPS, the spline interpolation method has also been updated to match its implementation.

Validation has been performed by comparing energy, force, and virial against LAMMPS results. The differences are consistently within the 1e-4 level or better, which is likely limited by the use of float32 precision. Below are the test results for a CrCoNi eam/alloy system:
total E: lmp=-1332158.100253 gpumd=-1332158.097856
energy: max|diff|=1.751200e-05 rms=5.421647e-07
force : max|diff|=1.555827e-04 rms=5.918578e-06
virial: max|diff|=1.924350e-04 rms=5.169926e-06

Overall, this update improves compatibility and numerical consistency with LAMMPS for eam/alloy potentials.

@brucefan1983
Copy link
Copy Markdown
Owner

This is good for people caring about the consistency between codes, although it does not mean our previous version is wrong or not good.

@brucefan1983 brucefan1983 merged commit aea5584 into brucefan1983:master Apr 24, 2026
2 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants