[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Implementing AbsTol for quadcc
From: |
Marco Caliari |
Subject: |
Re: Implementing AbsTol for quadcc |
Date: |
Sat, 23 Sep 2017 11:44:40 +0200 |
On Fri, 22 Sep 2017 10:31:39 -0700
Rik <address@hidden> wrote:
> How do people feel about making this change? Should I default it to using
> [RelTol AbsTol] so existing code can run unmodified? I'd prefer not to
> since the interface would then differ from the other quad routines. Should
> there be a warning in place for two versions of Octave when the single
> tolerance syntax is used informing the user that this value will now be
> used for AbsTol rather than RelTol?
Yes, maybe with the hint on how to disable that specific warning.
> Should the AbsTol default to something
> more sensible than 0, such as 1e-10 which is what quadgk uses?
Yes, this is the default in integral?, too.
Marco