I seem to be having trouble with the cells_first option. I want cell
integration to be done after ALL face/boundary integration. But this does
not seem to be happening. Here is what I encounter.

I call integration_loop with cells_first set to false.

In face/boundary function, I update some variable, say, shock_indicator.

In the cell function, I overwrite this variable. But when I visualize the
shock_indicator by saving it to file, I dont see what I had written in cell
function.

Then, I put a "return" statement at beginning of face/boundary function, so
that those functions dont change shock_indicator. In this case, I see the
correct values.

>From this, it seems to be that some face/boundary integration is done after
the cell integration.

Any ideas what could be going wrong ?

praveen

On Mon, Apr 4, 2011 at 6:39 PM, Bärbel Janssen <
[email protected]> wrote:

> Yes, this does exactly what you described.
>
>
> On Mon, 4 Apr 2011, Praveen C wrote:
>
>  Hello all,
>> In integration_loop, the last argument is
>>
>> bool cells_first
>>
>> In my problem, I need to do all the face integrations first, computing
>> some jump terms, and adding these to the two cells sharing the face. Once
>> ALL the faces
>> are done, I want to do the cell integration. If I set cells_first=false,
>> does this serve my purpose ?
>>
>> Thanks
>> praveen
>>
>>
_______________________________________________
dealii mailing list http://poisson.dealii.org/mailman/listinfo/dealii

Reply via email to