Path: utzoo!utgpu!jarvis.csri.toronto.edu!clyde.concordia.ca!uunet!crdgw1!control!kassover
From: kassover@control.crd.ge.com (David Kassover)
Newsgroups: comp.graphics
Subject: Re: Fracturing self-intersecting polygons into trapezoids
Keywords: polygon,intersection,double precision, accuracy
Message-ID: <5637@crdgw1.crd.ge.com>
Date: 1 Mar 90 00:01:15 GMT
References: <20408@versatc.versatec.COM>
Sender: news@crdgw1.crd.ge.com
Organization: Aule-Tek, Inc.
Lines: 32

In article <20408@versatc.versatec.COM> ritter@versatc.versatec.COM (Jack Ritter) writes:
...
>
> How can I do double precision fixed-point arithmetic in C?
>-- 
Easy (to say, anyway)
 
you write a bunch of routines (maybe you want them to be
functions) that explicitly do the arithmetic you want to do.
Then you call the routines instead of combine variables with
operators.
 
E.G. for addition, you add the two lower order segments together,
checking for overflow.  Then you add the two higher order
segments, and add 1 if there was overflow from the previous
addition.  Then you glom both parts back into your data
structure.
 
If this were Ada, which it is not  (If I were vulgar...*), you
could write your functions carefully and overload them onto the
operators...
 
I'm not a C guru, so I don't know if integer overflow is
detectable in any standard form of the language...

Or maybe you could ignore overflow, then examine the two lower
order segments carefully to determine if there *should* have been
an overflow?
 
It begins to be not hardly worth the effort.  Perhaps you can get
permission to implement THESE ROUTINES in local assembler and
call them from your main program.