COLVAR output of a restarting job of metadynamics

338 views
Skip to first unread message

Guanna Li

unread,
Oct 28, 2015, 10:45:47 AM10/28/15
to cp2k
Hi,

I am a new user of CP2K doing metadynamic simulation. 

Step1: submit a meta job with two defined CVs and it finished after 100000 steps
Step2: continue my meta job by using the output file of XXX.restart  

Then I found the data of COLVAR.metadynlog of step2 was different from the step 1. Can anybody tell me what the meaning of columns 4-9 in the COLVAR.metadynlog? 

COLVAR.metadynlog  Step1:
         Time                 CV1            CV2          
      0.00000000      0.48665      0.01772      0.00000      0.00000      0.00000      0.00000      0.00000      0.00000
      0.50000000      0.50674      0.01802      0.00000      0.00000      0.00000      0.00000      0.00000      0.00000
      1.00000000      0.52609      0.01830      0.00000      0.00000      0.00000      0.00000      0.00000      0.00000 

Question 1: what are the 4-9 columns?

COLVAR.metadynlog  Step2:
         Time                
   5000.00000000      0.00294      0.02489     -0.03561     -0.28327      1.41194      0.00000      0.02782      0.00498
   5000.50000000      0.00308      0.02640     -0.03240     -0.23322      1.38499      0.00000      0.02850      0.00480
   5001.00000000      0.00321      0.02791     -0.02962     -0.18213      1.35876      0.00000      0.02911      0.00462

 Question 2: what are the 2-9 columns? I assumed that the 2nd and 3rd columns would be CV1 and CV2, and from the 4th column they should also be zero as the output of step1. But apparently they are not. 
 




Marcella Iannuzzi

unread,
Oct 28, 2015, 12:17:31 PM10/28/15
to cp2k


Hi 

If MTD is running, the columns from 4 to 9 should not be zero.
Since you use two CVs, the columns in the *COLVAR* file are
Time    CV1    CV2    MTD_Force on CV1       MTD_F on CV2      WallPot_F on CV1        WallPot_F on CV2    MTDPot       WallPot

where the forces and the potentials are intended at the actual position in the phase space where the trajectory is at that specific time,
hence, considering the number of MTD hills already spawned.

Regards
Marcella

Guanna Li

unread,
Oct 28, 2015, 12:51:17 PM10/28/15
to cp...@googlegroups.com
Marcellar, thank you for your explanations.
You point to a new question: I confirmed that my MTD was running smoothly and I can view the reaction was "taking place" along the CVs. 
Then the question would be: why I had zero for columns 4-9 in the first step? I am wondering that maybe columns 4-9 are controlled by the PRINT section and the default settings for them was "OFF"? However, i didn't find any key words related to them in the manual.

Guanna

--
You received this message because you are subscribed to a topic in the Google Groups "cp2k" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/cp2k/DN3CFGly2m0/unsubscribe.
To unsubscribe from this group and all its topics, send an email to cp2k+uns...@googlegroups.com.
To post to this group, send email to cp...@googlegroups.com.
Visit this group at http://groups.google.com/group/cp2k.
For more options, visit https://groups.google.com/d/optout.



--
Guanna from gmail

Guanna Li

unread,
Nov 2, 2015, 7:36:09 AM11/2/15
to cp2k
Can anyone help me out form this question?
To unsubscribe from this group and all its topics, send an email to cp2k+unsubscribe@googlegroups.com.

To post to this group, send email to cp...@googlegroups.com.
Visit this group at http://groups.google.com/group/cp2k.
For more options, visit https://groups.google.com/d/optout.



--
Guanna from gmail

Marcella Iannuzzi

unread,
Nov 2, 2015, 9:51:09 AM11/2/15
to cp2k
Are you sure that those columns are all zeros?
If MTD is running this makes no sense.
Marcella

Guanna Li

unread,
Nov 2, 2015, 10:16:29 AM11/2/15
to cp2k
Hi Marcella,

Yes, you are right! Thank you very much!
It was my mistake. Those columns are not all zeros. I am sorry for bothering you with this.

Guanna
Reply all
Reply to author
Forward
0 new messages