write restart or write data file

I read the documents and find this:
" Because a data file is in text format, if you use a data file written out by this command to restart a simulation, the initial state of the new run will be slightly different than the final state of the old run (when the file was written) which was represented internally by LAMMPS in binary format. A new simulation which reads the data file will thus typically diverge from a simulation that continued in the original input script. "

my simulation is in the equilibrium stage (npt run for get to equilibrium ) and is not the final simulation stage.
do you think it is ok or I should do my simulation from scratch.

I like to use write_data because I can easily manipulate the system if necessary. If you’re not sure, you could always use both like Eliar says. If you’re asking if you should restart your simulation because you’ve been using read_data, I would say it depends on what you’re doing. For my research, it doesn’t matter if simulations diverge due to slightly different starting conditions (it’s actually better if they diverge since I usually run multiple samples of the same system).

Also, there is command-lline option -r2data for LAMMPS that
converts a restart file into a data file. See section 4.2 of
the manual. So you can write restart files, then later
convert any of them you want to data files.