On Tue, 20 Mar 2012 at 08:33PM -0700, P Purkayastha wrote: > It is simply joining the points. The plot is actually correct. Attached is > the output of > > plot(sqrt(2*x^2-3)/(5*x-2), (x,-10,10),ymin=-5,ymax=5,marker='*') > > which shows that there are no points in the imaginary range.
Ah, that's helpful. The straight line joining the points looks funny on my screen due to some kind of aliasing or something. But yes, it's just connecting the points. On Tue, 20 Mar 2012 at 10:34PM -0500, Benjamin Jones wrote: > This is the same artifact that you run into when plotting over a vertical > asymptote on most graphing calculators, you get a (nearly) vertical line > drawn because the plotter is joining computed points with lines. It would > be nice to add a feature like the `detect_poles` option where the domain > could be detected better. There already is a detect_poles option to plot(), but it doesn't detect this problem. :) It would be nice to detect undefined intervals like I had here. Dan -- --- Dan Drake ----- http://mathsci.kaist.ac.kr/~drake -------
signature.asc
Description: Digital signature
