This is an old revision of the document!
it call l-bfgs-b routine till: - the convergence criteria defined in cg.optional are reached (f(i,j)<fmax for all i,j) - the algorithm is not able to find the smaller energy in spite of f>fmax - the maximum number of steps exceeds
iquench = -5 dt = 0.5 !/only if FIRE minimization is used - last parameter in cgopt.optional >0/
This routine uses the same input file as Conjiged Gradients (iquench = -4). It does not use all parameters, but they must be there.
0.1 ! max displacement - NOT USED in BFGS 0.1 ! dummy argument - NOT USED in BFGS 0.00000001 ! total energy tolerance - NOT USED in BFGS 0.001 ! fmax - force tolerance - end of minimization if fij<fmax 10000 ! Max number of opt steps 0 !number of restarts in case ftot>fmax (it seems it nave no effect) 0 !if >0 then dt must be defined in fireball.in. Number of FIRE downhill steps after BFGS minimization "fails"
iquench = -6 dt = 0.5 !timestep in MD
This routine uses the same input file as Conjiged Gradients (iquench = -4). It does not use all parameters, but they must be there.
0.1 ! max displacement 0.5 ! NOT USED in FIRE 0.0 ! total energy tolerance - the walker can make step uphill if Enew-Emin<Etol, Etol=0 => only downhill steps allowed 0.001 ! fmax - force tolerance - end of minimization if fij<fmax 10000 ! Max number of opt steps 0 ! NOT USED in FIRE minimization