> < ^ Date: Thu, 28 Jul 1994 14:33:00 +0100 (MET)
> < ^ From: Martin Schoenert <martin.schoenert@math.rwth-aachen.de >
> < ^ Subject: Re: the StabChain function

Leonard Soicher writes in his e-mail message of 1994/07/28

The new StabChain function in Gap 3.4 is very useful for my work on
graphs and geometries, since I often know the size of a given group
acting on a graph. However, sometimes I only know an upper bound, and
was pleased to see that StabChain offers the option of specifying this
upper bound through the limit parameter. I was disappointed, however,
to find that with this option, StabChain appears to take as long as
when given no extra information, even when this upper bound is
achieved. Please tell me how to fix this!

The 'limit' option is currently only honoured by the random method.
The default however is the deterministic method.

The easiest solution is to force using the random method. You do this
with the 'random' option, using any value less than 1000.

gap> C := Combinations( [1..12], 4 );;
gap> G := Operation( SymmetricGroup(12), C, OnSets );;
gap> StabChain( Copy(G), rec( random := 1000, limit := Factorial(12) );;
gap> time;
62353
gap> StabChain( Copy(G), rec( random :=  999, limit := Factorial(12) );;
gap> time;
4867
gap> StabChain( Copy(G), rec( random := 1000, limit := Factorial(13) );;
gap> time;
62524
gap> StabChain( Copy(G), rec( random :=  999, limit := Factorial(13) );;
gap> time;
64566

Of course there is a small chance that the random method will fail.
But with 'random := 999', it is extremly small.

Martin.

-- .- .-. - .. -.  .-.. --- ...- . ...  .- -. -. .. -.- .-
Martin Sch"onert,   Martin.Schoenert@Math.RWTH-Aachen.DE,   +49 241 804551
Lehrstuhl D f"ur Mathematik, Templergraben 64, RWTH, 52056 Aachen, Germany

> < [top]