MathGroup Archive 2007

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

Search the Archive

Re: Complexity explosion in linear solve

  • To: mathgroup at smc.vnet.net
  • Subject: [mg80262] Re: Complexity explosion in linear solve
  • From: carlos at colorado.edu
  • Date: Thu, 16 Aug 2007 04:42:28 -0400 (EDT)
  • References: <200708141107.HAA09644@smc.vnet.net><f9ueas$b9n$1@smc.vnet.net>

>
> Try Method->"OneStepRowReduction" (without quotes, if in versions < 6) and
> see if that helps.
>
> I get (after suitably redefining b so it is not self-referential, and
> changing A to mat)
>
> In[8]:= Timing[
>  x = LinearSolve[mat, b, Method -> "OneStepRowReduction"];]
>
> During evaluation of In[8]:= LinearSolve::nosol: Linear equation \
> encountered that has no solution.
>
> Out[8]= {0.15, Null}
>
> Probably not the anticipated solution, but it is quick and short.
>
> Daniel Lichtblau
> Wolfram Research

When I tried that option under version 5.2 for Mac OS X,
the expression returns unevaluated.  I could not find documentation
for it; what does it do?

I had loaded 6.0 in July, from the CU campus license, but discarded it
after showing too unstable - kept freezing, had to kill it manually.






  • Prev by Date: FindRoot and Interpolating function
  • Next by Date: Re: Re: RandomComplex documentation
  • Previous by thread: Re: Complexity explosion in linear solve
  • Next by thread: Re: Complexity explosion in linear solve