[petsc-users] Information on SNES

[Matthew Knepley]

On Tue, Sep 29, 2020 at 9:21 AM Oliver Zacharias <olz at freenet.de> wrote:

> Dear Sir or Madam,
>
> regarding your SNES-Tool one can specifiy at least three types of
> accuracy which force the code to terminate. I would like know how you
> have defined (and implemented) the following quantities:
>
> a) "rtol"
>

This means a relative tolerance, or bound on ||F(x_n)||/||F(x_0)||


> b) "abstol"
>

This mean an absolute tolerance, or a bound on ||F(x_n)||

  Thanks,

      Matt


> The definition of the quantity "stol" is already described on one of
> your manual pages. Would you also like to present a cite where to find
> those definitions?
> Thanks in advance,
> Oliver Zacharias
>
>

-- 
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

https://www.cse.buffalo.edu/~knepley/ <http://www.cse.buffalo.edu/~knepley/>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20200929/f0c2600d/attachment.html>