Path: utzoo!attcan!uunet!husc6!mailrus!tut.cis.ohio-state.edu!mandrill!abvax!gfs
From: gfs@abvax.UUCP (Greg F. Shay)
Newsgroups: comp.graphics
Subject: Request for CORDIC algorithm info
Message-ID: <354@abvax.UUCP>
Date: 25 May 88 14:10:11 GMT
Organization: Allen-Bradley Company, Inc; Industrial Computer Division, Highland Heights, OH
Lines: 22

Ok, I'm interested.  In reply to my sin(x) in integer posting, a couple
people mentioned the CORDIC algorithm which apparently performs the whole
coordinate system transform without any multiplies (albeit using a small
lookup table.)  
	My application for the integer (actually fixed point) sin(x) is
indeed a vector display program, which I now have converted completely to
68000 assembler (no c routines left) in the display pipeline (vector multiply,
vector draw,screen clear).  The bottleneck in performance is the vector by 
matrix multiply (six multiplies and six additions per vector).  The CORDIC
algorithm sounds like it will greatly speed up my display pipeline by 
transforming the display vectors without the vector by matrix multiply per
vector.
	Could someone either thumbnail sketch the CORDIC algorithm for me
or give a reference?  Many Thanks.

		Greg Shay

	pyramid  |
	decvax   |..!abvax!gfs
	mandrill |