社长,威武, |
zako 发表于 2024-9-5 20:04 You should add option for xtb code like that you mentioned. |
zako 发表于 2024-9-5 20:04 It should be the latter. During the calculation process, Gaussian only acts as an optimizer, and all energy-related calculations are performed by the external program pointed to by the external keyword, so the energy-related settings are also external. |
Dear Professor, If we want to consider the solvent, what should we do ? add the solvent using Gaussview "ex scrf=(solvent=thf)", or add solvent to xtb code using --gbsa ? what's best |
maximos 发表于 2024-9-4 18:51 The chk file produced by this Gaussian+xtb running may be used to provide geometry for subsequent DFT calculation, but this running cannot provide wavefunction as initial guess for subsequent DFT calculation. |
Is it possible to use the generated checkpoint file from optimisation (mol.chk) to run a DFT calculation (at B3LYP for ex) ? (I mean read geometry and wave function), because i tried and didn't work. |
WOOOOWOOOO 发表于 2023-11-27 10:38 *** Configuration 4 *** Current date: 2023-11-27 Time: 10:13:57 Loading geometry 4 from the inputted trajectory file Wall clock time elapsed for calculating this configuration: 6 s Wall clock time elapsed for calculating this configuration: 5 s *** Configuration 3 *** Generating xtb.xyz file... *** Configuration 4 *** 诸如以上信息都是错乱的,明显并行方式不当,肯定是同时运行了多个molclus进程,输出信息搅合在一起,结果能正常就怪了,运行效率不巨低就怪了 |
本帖最后由 WOOOOWOOOO 于 2023-11-27 10:50 编辑 sobereva 发表于 2023-11-27 02:14 xtb。您翻到s开头的附件后面有很多的 Error: Cannot find .xtboptok file! The task is failed! 文本末尾显示srun: error: g0503: tasks 0-11,13-23: Killed 我在论坛里搜了一下这个问题,看了环境配置都是对的,xtb.out也没看出来啥问题。我跑了您在一个文章中的6个Li的团簇搜索,完全没问题。麻烦您看看我这个是啥问题。 |
5ab871b01f6dedc2e9e3d5360ece942.png (23.98 KB, 下载次数 Times of downloads: 30)
33.55 KB, 下载次数 Times of downloads: 1
WOOOOWOOOO 发表于 2023-11-26 22:34 说清楚什么程序崩溃 molclus调用xtb优化,在优化过程中不收敛,不可能导致molclus崩溃 |
sobereva 发表于 2023-11-26 20:12 好的,老师,谢谢您。还有个问题我想请教一下您,我是在做单个四氧化三铁团簇的构象搜索,用的是Fe.xyz和O.xyz,分别为3和4个。但是genmer产生的结构质量很差,用molclus调用xtb去优化由于结构过于离谱优化100来个程序就崩溃了。请问在用genmer产生初始的结构时有什么建议吗?要做一些什么约束呢? |
WOOOOWOOOO 发表于 2023-11-26 11:26 看卢老师博文《使用Molclus结合xtb做的动力学模拟对瑞德西韦(Remdesivir)做构象搜索》 技巧:利用crest程序节约xtb批量优化大量结构阶段的耗时 |
WOOOOWOOOO 发表于 2023-11-26 11:26 xtb又不会自动给你优化里面每一帧的结构,当然得用molclus调用xtb |
请问我用genmer产生了200个团簇构象,再用xtb优化,但是只优化一个结构就结束了怎么回事呀? 命令:xtb 200_Fe3O4.xyz --gfn 1 --opt tight -c 0 -u 5 |
sobereva 发表于 2023-7-23 23:48 xtb mol.xyz --chrg $charge --uhf $uhf --hess --grad > xtbout 2>&1 进行了如下修改,程序运行后在屏幕上不显示了。谢谢sob老师 |
PESPES 发表于 2023-7-23 16:06 自己改脚本,把xtb显示的这种信息重定向到/dev/null里避免显示 |
手机版 Mobile version|北京科音自然科学研究中心 Beijing Kein Research Center for Natural Sciences|京公网安备 11010502035419号|计算化学公社 — 北京科音旗下高水平计算化学交流论坛 ( 京ICP备14038949号-1 )|网站地图
GMT+8, 2024-11-23 20:42 , Processed in 0.218925 second(s), 27 queries , Gzip On.