MathGroup Archive 2011

[Date Index] [Thread Index] [Author Index]

Search the Archive

plot error messages are completelly missing

  • To: mathgroup at smc.vnet.net
  • Subject: [mg116890] plot error messages are completelly missing
  • From: john <j.f.b.payne at blueyonder.co.uk>
  • Date: Thu, 3 Mar 2011 05:59:47 -0500 (EST)

Hello,

Apologies if this isn't the proper way to submit, it seemed to be from
http://smc.vnet.net/MathGroup.html

This is a late reply on the subject which came up Sep 10 2007!

Using Mathematica 8 recently, I was working with a complicated function, which
I'd Plot[]ted.  None of the results I was getting made sense.  Eventually, I
discovered that where the function appeared to be zero in the plot it was
actually imaginary, and everything fell into place.  I hadn't considered this
possibility because I've been used to the Plot[] behaviour of version 5.2 (and
previous versions) where error messages reported the function not being real.

So, I strongly agree with Szabolcs comment that there should be a message for
complex numbers/infinity.  Personally, I think error messages should be on by
default.  I want to know when things are going wrong, and not be left believing
misleading information.  I suppose this is because I do research, so I don't
just want a noise-free picture of things I already know.

Anyway, thanks to Jason Martinez of Wolfram Technical Support for the following
information.  You can get error messages for complex numbers as follows:

On[General::plnr]

Plot[Sqrt[1 - x^2]^7, {x, 0, 2},
 Method -> {"SuppressMessages" -> False}]

This is still not default behaviour, but default behaviour is not a big issue,
because of course, you can include in the init.m file

On[General::plnr]

SetOptions[Plot, Method->{"SuppressMessages"->False}]

Normally Method option for Plot[] is Automatic.  I don't know whether there are
any side effects of changing it.


John Payne



  • Prev by Date: Re: problem to define a complex quantity
  • Next by Date: Re: Array from for-loop iterations
  • Previous by thread: Re: Patterns with conditions
  • Next by thread: Re: plot error messages are completelly missing