open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello!

I have a problem with my res file.
I cant open it with bluekenue or fudaa.
I have an error message.
My file is quite heavy 4 Go.



Does anybody know what the is problem?
Thank you!

Hi Elsa,

It doesn't seem to be a problem of size of your file, but of its format: the header it's not correct. How this file has been created ? if it's a result from a simulation, does the simulation end correctly ?

Regards,
Laurent

This is one of my sortie file. It seems to end correctly...
Do you need another file to check?

Do you think if I change the header it will work?

Thanks

Laurent,

so sorry I didnt send you the right sortie file.
This is the good one :



-Ive done as you told me and change the number of maximum iterations to 300 instead of 100,
-This simulation is not for 3minutes but 35h. that's why it is heavy. As I have difficulties with the weir file, I put elements in the mesh at the weir level to simulate the wier with the mesh and without the weir file.
-Ive simulated with 48 processors so in parallel.

You ll find the cas of this simulation there.

Ok.

- I'm not sure to really understand what you did with the elements of the mesh but I think it can be the problem, as BlueKenue complains about elements. Can you tell us more about it? if you modified the geometry file, can you read it after your modification ?

- Can you read your result file in sequential mode? no need to run the whole simulation, just 1-2 time steps.

Laurent

Elsa,

5 things which can help you:

1) the log file seems not to be complete. Are you sure the computation well finished? It misses some lines at the end, something like that:
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
... handling result files
+> cas
moving: res.res

My work is done



2) as I told you yesterday, you have warning messages that tell you that the solver doesn't succeed to converge to reach the requested precision (1e-4 by default). Either you reduce your time step or you increase the number of maximum iterations.

3) you have Nan values during the computation, you should investigate to understand why. Maybe because of the solver non-convergence.

4) it seems that you request 1 parallel processor, it means that you are running the model with the executable compiled with parallel libraries but over 1 processor. Try with "PARALLEL PROCESSORS = 0"

5) if for 3 minutes of computation you have a 40Go result file, try to reduce the frequency of outputs so as to see if your file is readable.

Laurent