From bug-request at octave dot org Tue Nov 22 21:58:25 2005
Subject: axis rounding
From: "John W. Eaton" <jwe at bevo dot che dot wisc dot edu>
To: hwz at hwz dot bv dot TU-Berlin dot DE
Cc: bug at octave dot org
Date: Tue, 22 Nov 2005 22:56:21 -0500

On 22-Nov-2005, hwz at hwz dot bv dot TU-Berlin dot DE wrote:

| I found that the axis-command introduces a large roundoff problem
| when using large values as limits because of the g-format statement
| in axis.m, e.g.
| 
|       if (len > 1)
|         __gnuplot_raw__ (sprintf ("set xrange [%g:%g];\n", ax(1), ax(2)));
|       endif
| 
| This lead to problems, if one likes to scale exactly
| Here is an example:
| 
| octave:17> y1 = 979923450;
| octave:18> y2 = 979923750;
| octave:19> sprintf ("set xrange [%g:%g];\n", y1, y2)
| ans = set xrange [9.79923e+08:9.79924e+08];
| 
| is not what i need (cutting 979923450 to 979923000!), but
| 
| octave:20> sprintf ("set xrange [%20.15g:%20.15g];\n", y1, y2)
| ans = set xrange [           979923450:           979923750];
| 
| does the job right.
| Therefore I'd suggest to change the format to %20.15g (or similar)
| to ensure to send all significant digits to gnuplot.

OK, I used %.16g instead (I don't think there is a need to set the
minimum overall field width here, but maybe I'm missing something).

jwe



-------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.

Octave's home on the web:  http://www.octave.org
How to fund new projects:  http://www.octave.org/funding.html
Subscription information:  http://www.octave.org/archive.html
-------------------------------------------------------------