On Wed, May 27, 2015 at 11:56 AM, Fande Kong <[email protected]>
wrote:

> Hi all,
>
> I was wondering if it is possible to let SNES  continue regardless of the
> divergency of linear search, even though the residual only decreases a
> little bit. In my test, if basic linear search is used, the snes converges
> and  needs more iterations. But, BT does not work well although it
> can guarantee a decrease in the residual.
>

http://www.mcs.anl.gov/petsc/petsc-master/docs/manualpages/SNES/SNESSetMaxNonlinearStepFailures.html

  Matt


> Thanks,
>
> Fande,
>



-- 
What most experimenters take for granted before they begin their
experiments is infinitely more interesting than any results to which their
experiments lead.
-- Norbert Wiener

Reply via email to