[petsc-dev] what's q in snes ex62

[Geoffrey Irving]

On Fri, Nov 15, 2013 at 3:37 PM, Matthew Knepley <knepley at gmail.com> wrote:
> On Fri, Nov 15, 2013 at 5:31 PM, Geoffrey Irving <irving at naml.us> wrote:
>>
>> On Fri, Nov 15, 2013 at 3:21 PM, Matthew Knepley <knepley at gmail.com>
>> wrote:
>> > On Fri, Nov 15, 2013 at 5:09 PM, Geoffrey Irving <irving at naml.us> wrote:
>> >>
>> >> Is the second equation the weak form of incompressibility?  If so, why
>> >> isn't q u?  It's used as q several places throughout the code, so it
>> >> seems more likely to be my misunderstanding than a typo.
>> >>
>> >>   < \nabla v, \nabla u + {\nabla u}^T > - < \nabla\cdot v, p > + < v, f
>> >> > = 0
>> >>   < q, \nabla\cdot v >
>> >> = 0
>> >
>> > u, v \in V and p, q \in P, and hopefully \div V = P for stability.
>>
>> Are you using the same variables as
>>
>>
>> http://fenicsproject.org/documentation/dolfin/dev/python/demo/pde/stokes-iterative/python/documentation.html
>>
>> If so, maybe that line should be
>>
>>     < q, \nabla\cdot u >
>>
>> I.e., "u" instead of "v"?
>
>
> Damn, you are right.

Great, that makes it easier to understand.  I'm trying to reverse
engineer the general definitions of all the fields in PetscFEM.  It's
okay that they're not written down yet; it's probably better for me to
do this myself anyways to get a better feel for everything.

Geoffrey