whats happening ?

[Karl Hasselström]

--82I3+IH0IqGh5yIs
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On 2002-10-22 00:26:30 +0100, Jason Moxham wrote:
>=20
> I done some preliminary work on a few algorithms ,and even in a completel=
y=20
> un-optimized form have got some good results for some of the following=20
> functions , other results , not so good :(  =20

> mpz_tdiv_q  using  varible precision , division free newton method
>=20
> Before I go ahead with anymore tweaking , is anyone else working on any o=
f=20
> these fn's or plans to ?

I'm working on inverting numbers with newton division, and using those
inverted numbers to do division. So what you describe is pretty much
what I am doing, yes. :-)

I'm pretty much done with the actual Newton code, apart from (a lot
of) polishing; currently, I'm trying to build the last pieces of the
routine that multiplies A with 2^k/B and spits out a quotient and a
remainder (ideally, I'd like to say quotient and/or remainder, but
that's for later).

If you're interested, we could exchange code and ideas. But beware,
I'm doing this as my master's project, so no matter how much better
your code is, I'll keep trying ... (Actually, what I'm trying to do is
to evaluate whether Newton division is a viable alternative for
reasonably small numbers, so by building a fast implementation you're
actually going to help me.)

--=20
Kalle Hasselstr=F6m, kha@treskal.com
      www.treskal.com/kalle

--82I3+IH0IqGh5yIs
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQE9tOnEN7iPzXSoOQoRAmKUAKCEtc94ZchTNGbYYblk2w76ENIO1gCfXQbq
SIMrkZhR3HLIjWHOAD79o7I=
=gBP4
-----END PGP SIGNATURE-----

--82I3+IH0IqGh5yIs--