On Feb 9, 2013, at 6:03 PM, Jed Brown <jedbrown at mcs.anl.gov> wrote:
>
> On Sat, Feb 9, 2013 at 5:52 PM, Barry Smith <bsmith at mcs.anl.gov> wrote:
> The new values are no less valid than the old so just overwrite the old
> outputs.
>
> Do we have an explanation for why they changed?
Switching to a very different way of handling DMs and functions in DMs and
many other things.
Yes one could do bi-section and see exactly when the change happened but
look, the old code was not validated, so it is not like we are any less (or
more) certain with the new code. When the error bars are infinite any answer is
equally valid.
Barry
> The change is not small (though it's better, mostly).
