Igor Schein on Fri, 22 Jan 1999 14:34:40 -0500


[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

nffactor() timing


Hi, 

I define functions f and g as follows:

? f(x)=x^16+x^12+x^5+1
? polredabs(f(x))
x^16 - x^11 + x^4 + 1
? g(x)=x^16-x^11+x^4+1

Then I try all 4 possible combinations:

nffactor(nfinit(g(y)),g(x))
nffactor(nfinit(g(y)),f(x))

nffactor(nfinit(f(y)),g(x))
nffactor(nfinit(f(y)),f(x))

The first two return the result in 12 seconds on Ultra-60.
The last two take forever ( I ran for 1.5 hours, then gave up).
"exposant considere" and precision keep doubling, and I don't know if
it'd ever finish.

Any ideas why such discrepancy?  After all, 2 polynomials define the
same number field and their T2 norms are very close.

Thanks
Igor