<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
</head>
<body>
<p><br>
</p>
<p>All,</p>
<p>We tend to take the simpler end of computing for granted. This
may be unwise. The following table shows the results given by six
calculators when performing numerical integration with respect to
x, from 0.1 to 1, of the function sin(2*pi/x):</p>
<p><br>
</p>
<p><font face="DejaVu Sans Mono">Canon F789SGA
1.401636757e-1 <br>
</font></p>
<p><font face="DejaVu Sans Mono">Casio fx-991ES PLUS
1.401636757e-1<br>
</font></p>
<p><font face="DejaVu Sans Mono">Casio fx-991EX
1.401636757e-1<br>
</font></p>
<p><font face="DejaVu Sans Mono">Hewlett-Packard HP50g
1.401636757e-1 (rounded to ten SF)<br>
</font></p>
<p><font face="DejaVu Sans Mono">Sharp EL-W506 </font><font
face="DejaVu Sans Mono">1.390363882e-1 *** WTF?! ***<br>
</font></p>
<p><font face="Courier New, Courier, monospace"><font face="DejaVu
Sans Mono">Sharp EL-W506T 1.390363882e-1 "</font></font><font
size="+1" face="Courier New, Courier, monospace"><br>
</font></p>
<p><font size="+1" face="Courier New, Courier, monospace"><br>
</font></p>
<p>Perhaps surprisingly, the Casio fx-991EX, a top-end,
school-level, non-programmable, non-graphical calculator, was
actually faster than the HP50g, a top-end, professional-level,
programmable, graphical device.</p>
<p><br>
</p>
<p>I hope that neither Boeing engineers nor school students are
using either of the Sharp devices.</p>
<p>Olwen</p>
<p><font size="+1"><br>
</font></p>
<p><font size="+1"><br>
</font></p>
<p><br>
</p>
<p><br>
</p>
<p><br>
</p>
<p><br>
</p>
<p><br>
</p>
</body>
</html>