0
$\begingroup$

Consider a complex path expressed in one real parameter, such as $$f(\alpha)=\frac{1}{e^{i \alpha} + 1},$$ which is a Möbius transformation of the unit circle. I know I can expand it all out and use the parametric plot, but it is cumbersome and not very easy (sometimes practically impossible) to do so.

Therefore, I want to know if there is some plotting function specifically designed for this. (I searched for it in the documentation but did not find it; sorry if it's all obvious.)

Improve this question
$\endgroup$
5
  • $\begingroup$ The first argument of ParametricPlot should be {Re@#, Im@#} &@f[x] -- not so cumbersome, I think. $\endgroup$
    – C. E. ♦
    Commented Sep 29, 2013 at 22:47
  • $\begingroup$ @Anon Thank you for the answer, I never thought of that nice solution. Anyway, I find it hard to get the code working. I've tried a couple of possibilities like ParametricPlot[{Re@#, Im@#} & (1 / (Cos[u] + I Sin[u] + 1)), {u, 0, 2 Pi}], ParametricPlot[{Re@#, Im@#} & (1 / (Exp[I u] + 1)), {u, 0, 2 Pi}] and ParametricPlot[{Re[1/(Cos[u] + I Sin[u] + 1)], Im[1/(Cos[u] + I Sin[u] + 1)]}, {u, 0, 2 Pi}]. None of them worked. The only thing that produced a result was ParametricPlot[{Re[Cos[u] + I Sin[u]], Im[Cos[u] + I Sin[u]]}, {u, 0, 2 Pi}]. $\endgroup$
    – 4ae1e1
    Commented Sep 29, 2013 at 23:07
  • $\begingroup$ It's because you've omitted @. $\endgroup$
    – C. E. ♦
    Commented Sep 29, 2013 at 23:14
  • $\begingroup$ @Anon Oops sorry, those are typos. Inserting @ does not seem to help though, and note there's simply no @ ParametricPlot[{Re[1/(Cos[u] + I Sin[u] + 1)], Im[1/(Cos[u] + I Sin[u] + 1)]}, {u, 0, 2 Pi}], it is explicit. Maybe you could open up a Mathematica session and give them a try? At least they don't work in my 9.0.1.0 student edition. $\endgroup$
    – 4ae1e1
    Commented Sep 29, 2013 at 23:41
  • $\begingroup$ I mean, the problem is with the complexity of the function, not the syntax. Since the very simple ParametricPlot[{Re[Cos[u] + I Sin[u]], Im[Cos[u] + I Sin[u]]}, {u, 0, 2 Pi}] does work. $\endgroup$
    – 4ae1e1
    Commented Sep 29, 2013 at 23:43

1 Answer 1

Reset to default
2
$\begingroup$

The problem is not with the complexity of the function, but with the fact that your function diverges. Consider:

f[u_, a_] := 1./(Exp[I u] + a);
Manipulate[ParametricPlot[{Re@#, Im@#} &@f[u, a], {u, 0, 2 Pi}, PlotRange -> All], 
          {a, 0, 2}]

As long as you stay away from values of $a$ near 1, the plot is reasonable. As $a$ approaches 1, the function diverges to +/-$\infty$.

Improve this answer
$\endgroup$
2
  • $\begingroup$ So I was trapped by my stupid example, into which I didn't give much thinking... Okay. $\endgroup$
    – 4ae1e1
    Commented Sep 29, 2013 at 23:57
  • $\begingroup$ Well the real part is a constant 1/2 so you get a vertical line that goes to infinity at Pi or -Pi. Otherwise Presentations has a ComplexCurve[f[t], {tmin, tmax}] routine for plotting complex curves, which you would have to email me to obtain more information. $\endgroup$
    – David Park
    Commented Sep 30, 2013 at 2:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.