On 2013-01-25, Dima Pasechnik <[email protected]> wrote:
> On 2013-01-25, Volker Braun <[email protected]> wrote:
>> ------=_Part_681_2503209.1359120527992
>> Content-Type: text/plain; charset=ISO-8859-1
>>
>> On Friday, January 25, 2013 1:11:15 PM UTC, Dima Pasechnik wrote:
>>
>>> If I read what Volker noticed right, in this code the bulk of time is 
>>> spent computing the order of a group of 2x2 matrices by converting it 
>>> into a permutation group, via GAP. And the order of the group is never 
>>> more than 342 here. A naive algorithm, enumerating the 
>>> elements of the matrix group directly, would speed this up hugely, IMHO. 
>>
>>
>> Or port the matrix groups to libgap...
> no, that's won't help, as that's what GAP does internally!
> you can do profiling of GAP on your example:
> http://www.gap-system.org/Manuals/doc/ref/chap7.html#X7FDF923D7D2937A1
> and see what happens...
> (just use ProfileGlobalFunctions(True) before running the thing to be
> profiled, and ProfileGlobalFunctions() to see the result)
>
on your example on http://tracker.gap-system.org/issues/221
it's actually about 4 times faster to call Size(Elements(G))
(this just naively enumerates the elements)
than Order(G)
(this constructs a permutation group along the way)




-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sage-devel?hl=en.


Reply via email to