From: JC Gumbart (gumbart_at_ks.uiuc.edu)
Date: Tue Oct 14 2014 - 22:27:07 CDT
I’ve run a number of REMD simulations with colvars but don’t believe I’ve seen this before. We’re using the alpha colvar and the five windows are at 0.1, 0.3, 0.5, 0.7, and 0.9. In the log file of replica 0, I see output at 3000 step intervals (why 3000? I have no idea, but it should be 1000)
colvars: # centers = { 0 }
colvars: # centers = { 0.1 }
colvars: # centers = { 0.3 }
colvars: # centers = { 0.3 }
colvars: # centers = { 0.3 } and so on
However, according the job output, the first exchange doesn’t happen until 20,000 steps "EXCHANGE_ACCEPT 3 4 RUN 20”. Replica 0, in fact, never exchanges, and the colvars traj output also supports this, with all numbers around 0.1.
So why does the log for replica 0 disagree with what we observe?
Thanks!
JC
This archive was generated by hypermail 2.1.6 : Wed Dec 31 2014 - 23:22:55 CST