乐平 发表于 2022-11-2 19:32
做计算并不是精度越高就越好,通常情况下是在精度和耗时之间取平衡。
先问一下自己为什么一定要用 EDIFF ...
Mind: In most cases, the convergence speed is exponential, so often, the cost for the few additional iterations is small. For high precision calculations, we recommend decreasing EDIFF to 1E-6. For finite difference calculations (e.g. phonons), even EDIFF = 1E-7 might be required in order to obtain very accurate results.
https://www.vasp.at/wiki/index.php/EDIFF
inner7peace 发表于 2022-11-10 11:06
精度越高确实代表误差越小,当然这意味着耗时也更久。
你可以把NELM调大一点,多跑跑,如果始终降不到1E ...
| 欢迎光临 计算化学公社 (http://bbs.keinsci.com/) | Powered by Discuz! X3.3 |