[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Plotting issue, Windows, Octave 4.0, very basic plots
From: |
Mike Miller |
Subject: |
Re: Plotting issue, Windows, Octave 4.0, very basic plots |
Date: |
Tue, 1 Sep 2015 10:27:44 -0400 |
User-agent: |
Mutt/1.5.23 (2014-03-12) |
On Tue, Sep 01, 2015 at 06:28:53 -0700, JokerOne wrote:
> but I am still not sure, if this is really the effect. I read anywhere, that
> the fltk toolkit fails at number as large as 1e38 (10e-38), but not at
> numbers in the order of 1e6?
This is also true, because the upper bound of single-precision floating
point is around 1e38.
But Panxto's explanation is also correct, the precision of
single-precision is less than the 12 significant digits that your
example tries to use.
octave:1> flintmax ("single")
ans = 16777216
This is the maximum integer value that can be represented exactly in a
single-precision value. That's about 7 significant digits (discounting
the leading 1).
HTH,
--
mike