Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!seismo!sundc!pitstop!sun!amdcad!ames!fxgrp!ljz
From: ljz@fxgrp.UUCP (Lloyd Zusman)
Newsgroups: comp.lang.c
Subject: Re: We need  H E L P (!!!!) with an algorithm.
Message-ID: <119@fxgrp.UUCP>
Date: Tue, 13-Oct-87 23:05:22 EDT
Article-I.D.: fxgrp.119
Posted: Tue Oct 13 23:05:22 1987
Date-Received: Thu, 15-Oct-87 20:52:22 EDT
References: <116@fxgrp.UUCP> <1460@spar.SPAR.SLB.COM>
Reply-To: ljz@fxgrp.UUCP (Lloyd Zusman)
Organization: FX Development Group, Inc., Mountain View, CA
Lines: 30
Summary: Thanks, but ...

In article <1460@spar.SPAR.SLB.COM> hunt@spar.UUCP (Neil Hunt) writes:

>If you are on a UNIX machine, I hope that you looked at MP(3)
>
>  MP(3X)	MISCELLANEOUS FUNCTIONS       MP(3x)

  [ followed by the first few lines of the 'man' page which describes
    these multiple precision routines already available udner Unix     ]


Thank you VERY much!  I feel a little stupid with this stuff sitting right
under my nose.

However, I ran these routines in a benchmark and found them to be VERY
slow compared to the Add, Subtract, and Multiply routines I had already
written (on an order of approximately 20-to-1 or more!).  I presume that
this is because these MP routines are intended for calculations of
arbitrary precision, while my routines are especially written for 64-bit
calculations.

Does anyone know where I can find the source code to these MP routines or
to something similar?  I'd like to borrow the division algorithm and hack
it to work with my 64-bit thingies.

Thank you very much.

-- 

Lloyd Zusman
...!ames!fxgrp!ljz