ERROR : no valid trial for current configuration

nayan · January 3, 2025, 9:13am

Hello,
my .gin file is starting with
"montecarlo bulk_noopt
cell
11.54000 11.53000 11.54000 90.000000 90.000000 120.000000
cartesian
…

temperature 300

#mcmove 0.0
mcswap only 1.0 1 Ni Mn

mctrial 100
mcsample 10 mc_traj_accepted_100_101

species
Na 1
Ni 2.09090909
Mn 4
O -2

buckingham grad
Na O 1271.504 0.3000 0.00 0.0 15.0 #ucl GULP library
Mn O 1397.63 0.3211 0.00 0.0 12.0
O O 22764.3 0.149 44.53 2.0 15.0
Ni O 1582.500 0.28820 0.00 0.0 12.0
dump every 1 example60.res
"
i found the output fill as
" Monte Carlo *

Number of trial MC steps = 100

Output frequency in steps = 100

Sample frequency in steps = 10

Probability of creation = 0.000000
Probability of destruction = 0.000000
Probability of translation = 0.500000
Probability of rotation = 0.000000
Probability of swap = 0.500000
Probability of strain = 0.000000

Maximum Cartesian shift = 0.050000 Angstroms
Maximum rotational shift = 180.000000 degrees
Maximum strain shift = 0.100000
Number of possible swaps = 1
Pairs for swap 1 = 1

File for MC samples = mc_traj_accepted_100_10.gmc

!!!
!! ERROR : no valid trial for current configuration
!!!

!!!
!! ERROR : no valid trial for current configuration
!!!
"
at the end in output file i am getting 3 trial structures. Can you help me to resolve the issue?

julian · January 4, 2025, 12:34am

Without having your whole input (including coordinates) it’s not possible to say for certain. However, the most likely thing, given that you only have a swap between Ni and Mn, is that your structure doesn’t have any of the atom types specified left to swap.

nayan · January 4, 2025, 12:00pm

100.got (47.0 KB)
mc_traj_accepted_100.gmc (17.8 KB)
the output files

nayan · January 4, 2025, 12:01pm

nnmo.gin (6.0 KB)

julian · January 6, 2025, 12:39am

Thanks for providing your input which makes it much easier to spot the problem. You can see from the output that translation is included in the MC by default as a move, but you have specified “bulk_noopt” which means that no atoms are allowed to move. Hence the error that there are no valid trial moves for translation. The solution is either to put the probability for translation to zero (i.e. turn this off) or to replace bulk_noopt with “conv” to enable translation for some atoms.