Services & ResourcesWolfram ForumsMathGroup Archive

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

Re: AbsoluteOptions and NDSolve

• To: mathgroup at smc.vnet.net
• Subject: [mg123889] Re: AbsoluteOptions and NDSolve
• From: Gadi Oron <gadior at gmail.com>
• Date: Wed, 28 Dec 2011 05:17:01 -0500 (EST)
• Delivered-to: l-mathgroup@mail-archive0.wolfram.com
• References: <201112262230.RAA20909@smc.vnet.net>

Hi Bob, and thanks for your assistance.

I've tried your idea, and it works for the second variable (t) but not for
the first (x) which is sadly the more important one.

Any idea as to how to push the step monitor one level down the evaluation?

Thanks again

On Tue, Dec 27, 2011 at 9:10 AM, Bob Hanlon <hanlonr357 at gmail.com> wrote:

> The output of NDSolve does not have options. Use StepMonitor
>
> xList = Reap[nds = NDSolve[
>       {y'[x] == y[x] Cos[x + y[x]], y[0] == 1},
>       y, {x, 0, 30}, StepMonitor :> Sow[x]][[1]]][[2, 1]];
>
> steps=Differences[xList];
>
> Take[steps,10]
>
> {0.00017069, 0.00034138, 0.00034138, 0.00034138, 0.0034138, \
> 0.0034138, 0.0034138, 0.0034138, 0.0227572, 0.0227572}
>
>
> Bob Hanlon
>
> On Mon, Dec 26, 2011 at 5:30 PM, Snark <gadior at gmail.com> wrote:
> > Hi,
> >
> > I am perplexed by the behavior of AbsoluteOptions on NDSolve. I tried
> > getting the actual value of StartingStepSize and got something weird.
> >
> > In[]:= AbsoluteOptions[NDSolve[eqSys, {c[x, t], T[x, t]}, {x, 0, xb},
> > {t, 0,1}]],StartingStepSize]
> > Out[]:= {{AbsoluteOptions[c[x,t]->InterpolatingFunction[{{0.,5.*10^-6},
> > {0.,1.}},<>][x,t],StartingStepSize],AbsoluteOptions[T[x,t]-
> >>InterpolatingFunction[{{0.,5.*10^-6},{0.,1.}},<>]
> > [x,t],StartingStepSize]}}
> >
> > Isn't there a way to fetch absolute options from NDSolve?
> >
> > Thanks in advance
> >
> >
>
>
>
> --
> Bob Hanlon
>

• References:
  • AbsoluteOptions and NDSolve
    • From: Snark <gadior@gmail.com>