Greetings! I believe the output is correct to within double floating
point precision. I don't think that either version is explicitly
simplifying the trigonometry to 0 (though I could be mistaken), so it
is simply an issue that gnuplot autoscales the y axis, whereas the
older native plotter doesn't seem to do so. The newer behavior (if
described correctly above) would appear to be better.
Take care,
Milan Lukic <lmilan@shell.core.com> writes:
> I attempted to verify graphically the following trig identity:
>
> sin(x) + cos(x) = sqrt(2)*cos(x-%pi/4);
>
> so I did:
>
> (%i1) sin(x)+cos(x)-sqrt(2)*cos(%pi/4-x);
> -- skip the output --
> (%i2)
> plot2d(%o1,[x,0,2*%pi]);
> I can send the eps file to anyone interested to see the weird
> picture I got.
>
> Here is the result of build_info();
>
> Maxima version: 5.9.0.9beta2
> Maxima build date: 14:27 8/11/2004
> host type: i686-pc-linux-gnu
> lisp-implementation-type: Kyoto Common Lisp
> lisp-implementation-version: GCL 2.6.4
>
> *****
> Maxima 5.9.0 produced a correct plot. Another observation, the plot
> in maxima-5.9.0 came out quite quickly, while maxima-5.9.0.9beta2
> took a long time to calculate and produce the plot.
> "time(%o2)" shows [13.6].
>
> Milan
>
> _______________________________________________
> Maxima mailing list
> Maxima@www.math.utexas.edu
> http://www.math.utexas.edu/mailman/listinfo/maxima
>
>
--
Camm Maguire camm@enhanced.com
==========================================================================
"The earth is but one country, and mankind its citizens." -- Baha'u'llah