I believe we are still using 6dx for the forecast time step in our workflow and experiments.
I suggest we try 8dx, in the cycling and longer-range forecasts.
This will make experiment run faster and save a not small % of core hours.
Once we confirm this larger time step works well for you, we should set this as the default in the workflow.
I believe we are still using 6dx for the forecast time step in our workflow and experiments. I suggest we try 8dx, in the cycling and longer-range forecasts. This will make experiment run faster and save a not small % of core hours. Once we confirm this larger time step works well for you, we should set this as the default in the workflow.