Hi,
> During a simulation with ntt=1 and tautp
> the velocities are scaled proportionally per time step from v to lambda*v
> where
> lambda=1 + (delta)t/2*tautp(T0/T-1) and tautp (from the namelist is
> taken in this equation by sander).
...from AMBER7/runmd.f:
scaltp = SQRT(1.0d0+DTTP*(EKIN0/EKGP - 1.0d0))
....It seems there's a sqrt missing or 'lambda' is a truncated taylor
series of 'scaltp' (or I misunderstand something).
>
> Please let me know if my assumptions were correct and if this makes
> sense...I know it might be trash but somehow I need to run NAMD on The
> cluster that I am using now because there is a problem with Amber and I
> a have a "time pressure problem" and I am trying to use the same
> protocol for the 2 programs.
....This might not help directly, but:
I don't know what type of simulation you're doing.
If it's "plain" md, production runs should rather not use temperature
coupling
(no matter whether it's done via velocity rescaling, Langevin dynamics
or whatever). Anyway, I believe if the results depend on what type of heat
bath was used, there seems to be a general problem.
good luck
Andreas
--
)))))
(((((
( O O )
-------oOOO--(_)--OOOo-----------------------------------------------------
o Wolfgang Andreas Svrcek-Seiler
o (godzilla)
svrci.tbi.univie.ac.at
.oooO Tel.:01-4277-52733
( ) Oooo.
-------\ (----( )--------------------------------------------------------
\_) ) /
(_/
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Mon Jul 28 2003 - 17:53:01 PDT
