You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
setupCheckpointing is called only at the start of the simulation, limiting checkpointing to only the initial fields. However, potentially a new field may be written at a later timestep. Maybe a turbulence model is turned on after initializing with a laminar flow, maybe a function object, etc.
There was a bug associated with a temporary field 'grad(U)' popping up during a step on the 'kOmegaSST' turbulence model calculation, see pull request #344. But this was the opposite case so to say, because a field was present at initialization but not at later timesteps.
I still need to think about the implementation and if this is really needed. Perhaps it's needed if users report problems related to this.
The text was updated successfully, but these errors were encountered:
setupCheckpointing is called only at the start of the simulation, limiting checkpointing to only the initial fields. However, potentially a new field may be written at a later timestep. Maybe a turbulence model is turned on after initializing with a laminar flow, maybe a function object, etc.
There was a bug associated with a temporary field 'grad(U)' popping up during a step on the 'kOmegaSST' turbulence model calculation, see pull request #344. But this was the opposite case so to say, because a field was present at initialization but not at later timesteps.
I still need to think about the implementation and if this is really needed. Perhaps it's needed if users report problems related to this.
The text was updated successfully, but these errors were encountered: