AMBER Archive (2009)
Subject: RE: [AMBER] production phase
From: Ross Walker (ross_at_rosswalker.co.uk)
> In AMBER tutorials, I found the author used ntb=1 with NTT=3, ntb=2
This is not as simple a question as you think it is. It depends on what you want to simulate, what you want to learn etc. Generally I recommend heating with ntb=1, ntt=3 then equilibrating the pressure with ntb=2,ntt=3 and then switching to constant volume (ntb=1) for production at which point you can either keep a langevin thermostat, if you want langevin dynamics, switch to ntt=0 and run NVE=0 or use a very week berendsen coupling ntt=1.
Be aware that Langevin is more expensive in parallel since it does not scale as well. NTT=3 necessarily corrupts the dynamics so if you want to measure things like correlation functions you will not be able to do this with ntt=3. So it really is a question of what you are actually looking to simulate.
All the best
Note: Electronic Mail is not secure, has no guarantee of delivery, may not be read every day, and should not be used for urgent or sensitive issues.