how to run MD in 20000 steps continuously

[Andre]

The situation has greatly improved when we use compiled 2.6 version 
(CP2K-3.0 before) and could  simulate about 3700 steps once. Anyway, I will 
take your suggestion, too.

在 2016年9月8日星期四 UTC+8下午5:25:06，Samuel Andermatt写道：
>
> Well, you can always change the source code as CP2K is open source, but 
> changing your input file might be easier. In general what I see with your 
> input file is that you have set EPS_PGF_ORB to 10^-32 which is really low. 
> In general you should probably try o make changes to the input file to 
> reduce the memory footprint. It is also strange that you had enough memory 
> for 381 steps, as the memory footprint should not change too much over the 
> curse of a simulation.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20160916/36f8db73/attachment.htm>