Same input give different output- transition metal surface

[zhouych]

Dear user and developers,

I am facing a problem with calculating a transition metal oxide surface: use the same input files, but the output energy and magnetic moment differ each time.

I attached my input files and two OSZICAR calculated at different time.

People said that maybe different calculations arrive at different local minimal states. Here, even I set an initial magnetic moment, calculations still give different final states. Is there any good method to find the lowest energy state?

All comments and suggestions are welcome! Thank you very much!

All the best,
Yecheng

[zhouych]

Dear all,
I have uploaded OUTCARs for 3 runs. Thank you!
Cheers,
Yecheng