[gmx-users] Continuing inturepted REMD simulation

[Nizar Masbukhin]

Dear gromacs users,
I have finished an REMD simulation, tough it had interupted (by electrical
instabilty that lead computer shutdown) several times.  I continue the
interupted simulation using this command:
mpirun -np 8 mdrun_mpi -v -multidir md[01234567] -replex 2000 -cpt 3 -cpi
state.cpt

however, the remd statistics show unexpected exchange probability.
Replica exchange statistics
Repl  3365 attempts, 1683 odd, 1682 even
Repl  average probabilities:
Repl     0    1    2    3    4    5    6    7
Repl
Repl  number of exchanges:
Repl     0    1    2    3    4    5    6    7
Repl        0    0    0    0    0    0    0
Repl  average number of exchanges:
Repl     0    1    2    3    4    5    6    7
Repl      .00  .00  .00  .00  .00  .00  .00


Repl                                Empirical Transition Matrix
Repl       1       2       3       4       5       6       7       8
Repl  1.0000  0.0000  0.0000  0.0000  0.0000  0.0000  0.0000  0.0000  0
Repl  0.0000  1.0000  0.0000  0.0000  0.0000  0.0000  0.0000  0.0000  1
Repl  0.0000  0.0000  1.0000  0.0000  0.0000  0.0000  0.0000  0.0000  2
Repl  0.0000  0.0000  0.0000  1.0000  0.0000  0.0000  0.0000  0.0000  3
Repl  0.0000  0.0000  0.0000  0.0000  1.0000  0.0000  0.0000  0.0000  4
Repl  0.0000  0.0000  0.0000  0.0000  0.0000  1.0000  0.0000  0.0000  5
Repl  0.0000  0.0000  0.0000  0.0000  0.0000  0.0000  1.0000  0.0000  6
Repl  0.0000  0.0000  0.0000  0.0000  0.0000  0.0000  0.0000  1.0000  7

That statistics was very different when i try remd simulation with short
timesteps, and uninterupted one. the exchange probabilty was 0.3 as i
expected.
Could anyone tell me whats wrong?
--