[gmx-users] LINCS warnings with timestep of 5 fs (but not 4 fs)
Chris Neale
chris.neale at utoronto.ca
Mon Jul 27 20:32:57 CEST 2009
Hello,
I am recently experimenting with a 5 fs timestep using virtual hydrogens
and LINCS. I am getting sporadic LINCS warnings, although the system
appears stable and does not blow up during the 100ns in which I get such
warnings. If I reduce the timestep to 4 fs from 5 fs (and increase
nstlist to 5 from 4), then I do not get any such LINCS warnings. It
occurs to me that in a larger timestep there could be a larger rotation
and that this warning may be entirely benign. I am not sure about that,
however, and would welcome any advice here, especially on the way that
the 30 deg threshold for a warning message is derived.
Thank you,
Chris.
Here is a snippit from my stderr output during a run:
...
Step 1387503, time 6937.51 (ps) LINCS WARNING
relative constraint deviation after LINCS:
rms 0.000156, max 0.001672 (between atoms 17832 and 17831)
bonds that rotated more than 30 degrees:
atom 1 atom 2 angle previous, current, constraint length
17832 17831 30.9 0.0946 0.0947 0.0945
Step 1387503, time 6937.51 (ps) LINCS WARNING
relative constraint deviation after LINCS:
rms 0.000139, max 0.001582 (between atoms 17832 and 17828)
bonds that rotated more than 30 degrees:
atom 1 atom 2 angle previous, current, constraint length
17832 17831 31.1 0.0946 0.0946 0.0945
Step 3601434, time 18007.2 (ps) LINCS WARNING
relative constraint deviation after LINCS:
rms 0.000035, max 0.000308 (between atoms 18168 and 18165)
bonds that rotated more than 30 degrees:
atom 1 atom 2 angle previous, current, constraint length
18168 18167 30.1 0.0945 0.0945 0.0945
Step 4391822, time 21959.1 (ps) LINCS WARNING
relative constraint deviation after LINCS:
rms 0.000121, max 0.002041 (between atoms 14190 and 14189)
bonds that rotated more than 30 degrees:
atom 1 atom 2 angle previous, current, constraint length
14190 14189 30.5 0.0945 0.0947 0.0945
Step 4796357, time 23981.8 (ps) LINCS WARNING
relative constraint deviation after LINCS:
rms 0.000128, max 0.001443 (between atoms 15532 and 15531)
bonds that rotated more than 30 degrees:
atom 1 atom 2 angle previous, current, constraint length
15532 15531 31.5 0.0945 0.0946 0.0945
Step 4796357, time 23981.8 (ps) LINCS WARNING
relative constraint deviation after LINCS:
rms 0.000111, max 0.001185 (between atoms 15532 and 15531)
bonds that rotated more than 30 degrees:
atom 1 atom 2 angle previous, current, constraint length
15532 15531 31.8 0.0945 0.0946 0.0945
Step 4796358, time 23981.8 (ps) LINCS WARNING
relative constraint deviation after LINCS:
rms 0.000221, max 0.002694 (between atoms 15532 and 15531)
bonds that rotated more than 30 degrees:
atom 1 atom 2 angle previous, current, constraint length
15532 15531 34.0 0.0946 0.0948 0.0945
...
########################################
With a .mdp file that looks like this:
integrator = sd
nsteps = 20000000
tinit = 0
dt = 0.005
comm_mode = linear
nstcomm = 4
comm_grps = System
nstxout = 20000000
nstvout = 20000000
nstfout = 20000000
nstlog = 0
nstlist = 4
nstenergy = 20000
nstxtcout = 20000
ns_type = grid
pbc = xyz
coulombtype = PME
rcoulomb = 0.9
fourierspacing = 0.12
pme_order = 4
vdwtype = cut-off
rvdw_switch = 0
rvdw = 1.4
rlist = 0.9
DispCorr = no
Pcoupl = Berendsen
pcoupltype = semiisotropic
compressibility = 4.5e-5 4.5e-5
ref_p = 1. 1.
tau_p = 4.0 4.0
tcoupl = Berendsen
tc_grps = System
tau_t = 1.0
ref_t = 310.
ld-seed = -1
annealing = no
gen_vel = yes
unconstrained-start = no
gen_temp = 310.
gen_seed = -1
constraints = all-bonds
constraint_algorithm= lincs
lincs-iter = 1
lincs-order = 6
;;;EOF
More information about the gromacs.org_gmx-users
mailing list