[GAP Forum] computing automorphism groups

[Alexander Hulpke]

Dear GAP Forum,

Rober Heffernan wrot:

> I'm computing the automorphism groups of finite groups  
> (particularly,finite perfect groups) but GAP seems to be taking an  
> awful long timeto do this.
> For example, the following has been running for several hours now:
> A:=AutomorphismGroup(PerfectGroup(168,1));
> The machine I am running this on is quite decently specced.  Also,  
> Istarted gap with the following:gap -o 1000Mto give GAP plenty of  
> memory to work with.

PerfectGroup returns by default a finitely presented group. An  
automorphism group calculation for these is likely to be very  
inefficient. If you use a permutation representation instead you are  
likely to get a much better performance:

A:=AutomorphismGroup(PerfectGroup(IsPermGroup,168,1));

Best wishes,

    Alexander Hulpke