| Re: leads ( No.1 )|
- Date: 2017/03/13 13:22
- Name: T. Ozaki
I wonder that such a small difference may not change results, and
that, thereby, you can simply enforce -.00001 to zero and regenerate
internal coordinates from fractional to cartesian coordinates.
| Re: leads ( No.2 )|
- Date: 2017/03/13 19:34
- Name: Artem Pulkin
- "i can not change -.00001 to zero because i obtained these unit vectors after variable cell optimization."
1e-5 is likely below the error bar, at least for default convergence parameters.