[gmx-users] Engery not converge on GROMACS~

[F JH]

Dear GMX users:

I use double precision GMX to minimize enegry,but I find it alway show that 
"conjugate gradient not converged in XXX steps"

How should I improve my job?or I should use sd(steepest descent) instead of 
cg(conjugate gradients)
Why does GROMACS stop the energy minimization even if it knows the energy 
is still high?

Here is my mdp file

title                    = 0108
cpp                      = /lib/cpp
include                  = -I../top
define                   =
integrator               = cg
dt                       = 0.002
nsteps                   = 50000
emstep                   = 0.0001

_________________________________________________________________
現在就上 MSN 會員目錄：在線上結交新朋友，找到興趣相投的夥伴 
http://members.msn.com?pgmarket=zh-tw