MathGroup Archive 2003

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

Search the Archive

maxsteps, startingstepsize in NDSolve.

  • To: mathgroup at
  • Subject: [mg43425] [mg43425] maxsteps, startingstepsize in NDSolve.
  • From: sean kim <shawn_s_kim at>
  • Date: Tue, 16 Sep 2003 04:36:53 -0400 (EDT)
  • Sender: owner-wri-mathgroup at

hello group. 

it's been sugested that the problem i was having with
my earlier post was due more or less to starting step
size, number of maximum steps.

is it possible that the the weird oscillations seen
and the failures are due to the accumulative numerical

i have tried changing maxstep size and starting
stepsize but it seems do very little.

is there a limit as to how high ( maxsteps) and how
low( starting step size) i can designate? 

if so what are they? 

and what else can i do here to ensure more accurate
results from the NDSolve?  I have tried precision
goals and accuracy goals along with machine precision
but it seems to take too long with the step size i
have specified. ( at the time i was using like
100000000 steps. since then i have tried it with
smaller number of steps but the speed seems to be an

thanks in advance for all helpful comments as always. 


Do you Yahoo!?
The New Yahoo! Search - Faster. Easier. Bingo.

  • Prev by Date: maxsteps, startingstepsize in NDSolve.
  • Next by Date: Re: Mathematica refuses to evaluate?
  • Previous by thread: maxsteps, startingstepsize in NDSolve.
  • Next by thread: Re: Pick up coordinates along different contours.