MathGroup Archive 2003

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

Search the Archive

RE: Re: Evaluation of AiryAi on the complex plane

  • To: mathgroup at smc.vnet.net
  • Subject: [mg44986] RE: [mg44959] Re: Evaluation of AiryAi on the complex plane
  • From: "Wolf, Hartmut" <Hartmut.Wolf at t-systems.com>
  • Date: Fri, 12 Dec 2003 04:41:24 -0500 (EST)
  • Sender: owner-wri-mathgroup at wolfram.com

>-----Original Message-----
>From: Jens-Peer Kuska [mailto:kuska at informatik.uni-leipzig.de]
To: mathgroup at smc.vnet.net
>Sent: Thursday, December 11, 2003 11:28 AM
>To: mathgroup at smc.vnet.net
>Subject: [mg44986] [mg44959] Re: Evaluation of AiryAi on the complex plane
>
>
>Hi,
>
>you know what you are doing ??
>The Airy function decay exponential with
>Exp[- 2 z^(3/2)/3] for Re[z]>0
>
>for 11.18 this gives a factor of Exp[-25]
>and you will need hig precision arithmetic
>to evaluate that.
>
>Regards
>  Jens
>

Yes, of course. I might add to Mauricio: observe

x0 = SetPrecision[11.18, 20]

Plot[Evaluate[Through[{Re, Im}[AiryAi[x0 + x \[ImaginaryI]]]]], {x, -10^-1, 
    10^-1}, PlotStyle -> {Hue[2/3], Hue[0]}, Evaluate[opt]]

and you'll see the problems of precision (entering through machine precision
x values).


You may overcome these

$MaxExtraPrecision = 100.

Plot[{Re@AiryAi[x0 + SetPrecision[x, 20] \[ImaginaryI]], 
    Im@AiryAi[x0 + SetPrecision[x, 20] \[ImaginaryI]]}, {x, -10^-1, 10^-1}, 
  Compiled -> False, PlotStyle -> {Hue[2/3], Hue[0]}]



For your examples:

Plot[Abs[AiryAi[x0 + SetPrecision[x, 20]]], {x, 0, 10^-100}, 
  PlotStyle -> Hue[0], PlotRange -> {All, {0, 10^-11}}, Compiled -> False]

Plot[Abs[AiryAi[x0 + SetPrecision[x, 20] I]], {x, 0, 10^-100}, 
  PlotStyle -> Hue[0], PlotRange -> {All, {0, 10^-11}}, Compiled -> False]

(Also, don't Compile the expression.)

--
Hartmut Wolf



>Mauricio Bedoya wrote:
>> 
>> Hello:
>> 
>> I'm in trouble evaluating the AiryAi function. I need to evaluate it
>> around the point (11.18,0) or (11.18 + I 0.0). When I sweep the real
>> axis there is no problem [1], it has a value of 2 10^-12 on all the
>> range.
>> 
>> But when I choose a range on a complex line [2], the 
>function throws a
>> 0 value on all the range.
>> 
>> when I evaluate the function [3] I obtain 0 in the complex plane
>> around 11.18 and 2.31151*10^-12 in 11.18, no matter how close I am
>> from 11.18 + 0.0 I.
>> 
>> This doesn't happen around the points 10.03 + 0.0 I and 
>23.52 + 0.0 I.
>> 
>> I work in mathemathica 4.1.2.0.
>> 
>> I have tried to evaluate the Ai function in the version 3 and the
>> result is 0 in the complex plane, but when I plot the function with
>> complex values I obtain values different from 0 and also the absolute
>> value of the function is continuous. I think that there is a problem
>> with the evaluation of the function around 11.18 because I 
>plotted the
>> function in mathematica 3 with no problems but I couldn't 
>evaluate it.
>> In the version 4 I couldn't plot or evaluate it.
>> 
>> I would be grateful if someone could help me with this problem.
>> 
>> Mauricio Bedoya
>> 
>> List of commands
>> [1] Plot[{AiryAi[11.18 + x]}, {x, 0, 10^-100}, PlotRange -> {All, {0,
>> 10^-11}}]
>> [2] Plot[{AiryAi[11.18 + x I]}, {x, 0, 10^-100}, PlotRange -> {All,
>> {0, 10^-11}}]
>> [3] Abs[ai[11.18]]
>>     Abs[ai[11.18 + 0 10^-11 I]]
>


  • Prev by Date: Immediate or Delayed Definitions in NDSolve?
  • Next by Date: Save as HTML and missing graphics
  • Previous by thread: Re: Evaluation of AiryAi on the complex plane
  • Next by thread: Compile