喵星大佬 发表于 2024-5-15 20:21
抓个虫,ORCA/xtb不能用周期性边界条件不妨碍NVT系综
如果生物体系模拟xtb/DFTB+的精度绝对是不足够的,尤 ...
ZetaFunction 发表于 2024-5-15 20:54
ORCA的MD只能通过定义“反弹边界”的方式实现体积控制,而且只能是方形球形等特定形状,感觉和真正的NVT ...
喵星大佬 发表于 2024-5-16 10:21
在IRC上做US完全可以每个窗口独立做,比如去商业超算上一次开几十台机器其实也用不了多久
再就是开边 ...
The MPI parallelization of sander/xtb only improves the performance of (1) the MM calculation, and (2) pre-SCF calculation of the QM/MM Ewald planewaves. As you correctly note, the XTB software does not support MPI parallelization. In my experience, the parallelization of XTB is not that great for systems with ~150 atoms; however, the XTB performance is similar to other programs such as DFTB+ and SQM.
欢迎光临 计算化学公社 (http://bbs.keinsci.com/) | Powered by Discuz! X3.3 |