[gmx-users] NVE simulations

[Anton Feenstra]

"David L. Bostick" wrote:
> 
> I understand that Sid is going for the largest timestep possible as Anton
> has published.  I would still recommend starting with a small timestep
> until equilibrium is reached.  It really shouldn't take long to do since it
> seems to be a simple, perhaps small system.  It should then be easier to
> take the final configuration from the equilibrium ensemble as an initial
> configuration for the simulations with a larger timestep...  Anyway, it
> will give a baseline for comparison.  .. but I'm no expert.

Non-equilibrium situations are actually an additional possible source
for energy drift, although in principle it shouldn't affect the total
energy.

Oh, by the way I forgot to mention that also single precision
calculation
can be a considerable source of energy drift, once all other sources
have
been eliminated. So if you *rrreeeaaaaallllyyyy* need accurate drifts,
you
will also need to run in double precision.

-- 
Groetjes,

Anton
 ________ ___________________________________________________________
|        | Anton Feenstra                                            |
| .      | Dept. of Pharmacochemistry - Vrije Universiteit Amsterdam |
| |----  | De Boelelaan 1083 - 1081 HV Amsterdam - The Netherlands   |
| |----  | Tel: +31 20 44 47608 - Fax: +31 20 44 47610               |
| ' __   | Feenstra at chem.vu.nl - http://www.chem.vu.nl/afdelingen/FAR|
|  /  \  |-----------------------------------------------------------|
| (    ) | Dept. of Biophysical Chemistry - University of Groningen  |
|  \__/  | Nijenborgh 4 - 9747 AG Groningen - The Netherlands        |
|   __   | Tel +31 50 363 4327 - Fax +31 50 363 4800                 |
|  /  \  | K.A.Feenstra at chem.rug.nl - http://md.chem.rug.nl/~anton   |
| (    ) |-----------------------------------------------------------|
|  \__/  | "Beat On the Brat With a Baseball Bat" (The Ramones)      |
|________|___________________________________________________________|