[gmx-users] Nose-hoover T-coupling in REMD

[ms]

>> On 25/12/10 02:12, Qin Qiao wrote:
>>
>>> Thanks a lot!
>>>
>>> It says Nose-hoover coupling generates the correct canonical ensemble, and
>>> that's the reason why I used it. I wonder whether v-rescale can also get
>>> the
>>> correct ensemble. Could you tell me?
>>>
>>
>> As far as I know v-rescale should generate a correct ensemble.

On 25/12/10 12:40, Qin Qiao wrote:
 > I'm not sure, since actually kinetic energy should have fluctuation, 
and the
 > v-rescaling may suppress the fluctuation..
 >

(PLEASE don't top post!!)

Well, the thermostat author, who I happen to know, says it generates the 
right ensemble. :) I'm not an expert so I can't say if it's true, but if 
you like you can read the original papers, linked here:

http://sites.google.com/site/giovannibussi/Research/algorithms#TOC-A-stochastic-velocity-rescaling-the

cheers,
m.

-- 
Massimo Sandal, Ph.D.
http://devicerandom.org