Some questions about 'CONTINUE_RUN'

[ganbaranaito]

Hello everyone, I have some problems about 'CONTINUE_RUN'.

My work is crashed due to some reasons, then I change 'CONTINUE_RUN' from 'FALSE' to 'TRUE' in order to restart. So, I have a question. If my task is not crashed and it will continue running, are these results same with the results produced by restart run?

Thanks in advance!

 

[fischer]

A "CONTINUE_RUN" will give you the same results as a restart run. The "CONTINUE_RUN" is automatically doing a restart run when the previous run successfully finishes.

Chris

 

[ganbaranaito]

A "CONTINUE_RUN" will give you the same results as a restart run. The "CONTINUE_RUN" is automatically doing a restart run when the previous run successfully finishes.

Chris

Thank you for replying!

 

[CGL]

Why I set the CONTINUE_RUN = TRUE, but the model output the initial time as 0001-01? It seems like the model is restart but not continue run.
Shoud I set the REF_TIME? or other parameter？@fischer

 

[slevis]

You should use
CONTINUE_RUN = TRUE
after first starting a simulation as a startup, or branch, or hybrid.
When the first part of the simulation generates restart files and rpointer files, then you can use
CONTINUE_RUN = TRUE
to restart. The model will ignore the REF variables in this case.

 

[CGL]

You should use
CONTINUE_RUN = TRUE
after first starting a simulation as a startup, or branch, or hybrid.
When the first part of the simulation generates restart files and rpointer files, then you can use
CONTINUE_RUN = TRUE
to restart. The model will ignore the REF variables in this case.

Yes. I change the CONTINUE_RUN to FASLE. The modle run as I wish and continue run by previous timestamp.