I'm just using the sage on sage.math, which presumably William built.
Perhaps he has been doing some experimenting or perhaps there is
something wrong with the build for that system. Just speculation on my
part though.
Bill.
On 8 Oct, 03:28, "Ondrej Certik" <[EMAIL PROTECTED]> wrote:
> > Sorry,
You are right. It's in the powering code I think, where powers of 2
are taken out before powering. This is at least evidence that Magma
uses the binomial expansion after all.
Actually, Magma is still about 4 times quicker for the original
problem at the start of this thread. I had some errors in
> Sorry, I meant sage 2.8.6. Typo. So this looks like a new problem
> rather than an old one.
I also tried SAGE 2.8.5 and it works fine. I didn't yet try 2.8.6,
since there are no binaries yet, but the sympy in there can be either
0.5.3 or 0.5.4, and I tried both versions (alone) and it works, so
On Oct 7, 2007, at 10:02 PM, Bill Hart wrote:
> Because of the truncated FFT in FLINT, this turns out to be way more
> efficient. I believe Magma gets around this by using classical
> multiplication when the length of one of the operands is less than 10
> (this is something SAGE could do too).
Bill Hart wrote:
> I've just been playing around with polynomial powering in sage, using
> magma, ZZ['x'] and Fmpz_poly objects.
>
> I've noticed that if I raise 743*x+423 to the power 2000, magma takes
> about 0.1s, Fmpz_poly takes about 3s and ZZ['x'] takes about 1s!!
>
> This sort of thing onl
I figured it outpartly.
I was being very stupid with the binary powering algorithm in FLINT.
If I wanted to raise a polynomial f of length n to the power of 7 I
would do:
out = f
pow = f
pow = pow*pow // n x n
out *= pow // n x 2n-1
pow = pow*pow 2n-1 x 2n-1
out *= pow // 3n-2 x 4n-3
Instea
P.S: how do I get the source code for f^2000 from the command line in
sage?
Bill.
On 8 Oct, 01:03, Bill Hart <[EMAIL PROTECTED]> wrote:
> I've just been playing around with polynomial powering in sage, using
> magma, ZZ['x'] and Fmpz_poly objects.
>
> I've noticed that if I raise 743*x+423 to th
I've just been playing around with polynomial powering in sage, using
magma, ZZ['x'] and Fmpz_poly objects.
I've noticed that if I raise 743*x+423 to the power 2000, magma takes
about 0.1s, Fmpz_poly takes about 3s and ZZ['x'] takes about 1s!!
This sort of thing only happens for polynomials of d
Sorry, I meant sage 2.8.6. Typo. So this looks like a new problem
rather than an old one.
Bill.
On 7 Oct, 23:55, "Ondrej Certik" <[EMAIL PROTECTED]> wrote:
> On 10/8/07, Bill Hart <[EMAIL PROTECTED]> wrote:
>
>
>
>
>
> > Just trying out 2.6 on sage.math. I typed "help()" then "modules" and
> > i
On 10/8/07, Bill Hart <[EMAIL PROTECTED]> wrote:
>
> Just trying out 2.6 on sage.math. I typed "help()" then "modules" and
> it said:
>
> Ignoring redefinition of Function: 'sympy.core.function.Function'> defined earlier than 'sympy.core.functions.Function'>
> Ignoring redefinition of Derivative
Just trying out 2.6 on sage.math. I typed "help()" then "modules" and
it said:
Ignoring redefinition of Function: defined earlier than
Ignoring redefinition of Derivative: defined earlier than
Ignoring redefinition of Mul: defined
earlier than
Ignoring redefinition of Add: defined
earlier
Indeed! Next time this comes up, I will take care of it myself and
send you a patch. :)
JV
--~--~-~--~~~---~--~~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, v
On Oct 7, 12:42 pm, Chris Chiasson <[EMAIL PROTECTED]> wrote:
Hello Chris,
> What about a virtualized system for each user?
Sure, that works, but seems to be an extraordinary amount of work to
fix the problem. Down the road at least the public notebook on
sage.math will migrate to its own har
What about a virtualized system for each user?
On Oct 6, 3:43 pm, mabshoff <[EMAIL PROTECTED]
dortmund.de> wrote:
> On Oct 6, 9:27 pm, Michel <[EMAIL PROTECTED]> wrote:
>
> > > I helps a little, but getting from non-privileged shell to root shell
> > > provided you have compilers isn't very hard.
14 matches
Mail list logo
