On Mon, Nov 04, 2013 at 07:30:43AM -0800, Nils Bruin wrote:
> On Monday, November 4, 2013 1:29:10 AM UTC-8, Georgi Guninski wrote:
> >
> > Isn't it possible to define the quality only in 
> > terms of the norm and the integer radical, 
> > something like this: 
> >
> > q(a,b,c) = max( norm(a),norm(b),norm(c) ) / 
> >            (log(Delta(K)) + degree(K) * log(radical(norm(a*b*c))) ) 
> >
> > with the restriction the norms of a,b,c to be coprime ?
> >
> 
> Assuming you mean to take log(max(...))):
> 
> With this definition you don't get a quantity that's invariant under 
> scaling: you want to have that
> 
> q(a,b,c)=q(t*a,t*b,t*c)
> 
> If a,b,c,t are integers you can accomplish this by dividing by doing 
> something like:
> 
> log ( max(norm(a),norm(b),norm(c))/norm(ideal(a,b,c)) )
> 
> [in case you're not familiar with ideals in rings of integers: think of it  
> as an analogue to gcd(a,b,c)]

Thanks.

I meant log(max()), sorry for the typo.

Since abc for integers doesn't allow scaling (coprimality),
I tried to avoid scaling too with coprimality of the norms,
but didn't do it properly.

I meant:  If the norms of a,b,c are not coprime integers,
q(a,b,c)=0

otherwise q(a,b,c)=log(max()) / ....


-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" 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-support.
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to