Date: Tue May 29 2018 - 18:19:31 CDT
Good catch - no, the thermostat was switched off (but I was minimising for 10ps before beginning REMD). I've just done a short run with the thermostat on and it's looking a lot more like it should now.
Thanks very much!
Jazmin Ozsvar | PhD Candidate
Weiss Laboratory | School of Life and Environmental Sciences | Faculty of Science
THE UNIVERSITY OF SYDNEY
Level 4 East, Charles Perkins Centre | The University of Sydney | NSW | 2006
From: Norman Geist <norman.geist_at_uni-greifswald.de>
Sent: Tuesday, 29 May 2018 5:12:11 PM
To: namd-l_at_ks.uiuc.edu; jozs2019_at_uni.sydney.edu.au
Subject: AW: namd-l: Difference between REMD replica temperature and actual temperature
Do you have the langevin thermostat enabled?
Von: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] Im Auftrag von jozs2019_at_uni.sydney.edu.au
Gesendet: Dienstag, 29. Mai 2018 04:03
Betreff: namd-l: Difference between REMD replica temperature and actual temperature
I've been trying to set up some T-REMD runs and have noticed that the actual temperature reported for replicas appears to be very different to the theoretical replica temperature. For example, I get history files that look like this:
2000 10 356.95 1.571019497206337 -27830.44422562166
4000 11 362.02 1.657781183829707 -27833.337255755807
6000 12 367.17 1.634859540746946 -27831.33867217014
8000 13 372.38 1.605235843972867 -27829.948596293
As far as I know, the third column is the theoretical temperature, whereas the fourth column is the actual temperature - if this is the case, then the difference is huge! I've tried running replicas for a few ns but the results don't really change.
I was wondering whether anyone has seen something similar and could give me some direction for troubleshooting. Or if I'm totally wrong about what the number in the fourth column means, please let me know!
This archive was generated by hypermail 2.1.6 : Thu Dec 12 2019 - 23:19:51 CST