Xref: utzoo comp.arch:12343 comp.misc:7318
Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!uwm.edu!cs.utexas.edu!tut.cis.ohio-state.edu!pt.cs.cmu.edu!a.gp.cs.cmu.edu!koopman
From: koopman@a.gp.cs.cmu.edu (Philip Koopman)
Newsgroups: comp.arch,comp.misc
Subject: trig function question
Keywords: elementary function implementation
Message-ID: <6909@pt.cs.cmu.edu>
Date: 10 Nov 89 12:04:23 GMT
Organization: Carnegie-Mellon University, CS/RI
Lines: 25

Cody & Waite (1980 edition) for SIN and COS.  I'm getting horrible
losses of precision for inputs > pi/4.  When I tried
deriving the equations myself a few years ago, the
equations I used said that had a valid interval from
0 < |x| < pi/4, but Cody & Waite use a similar equation
for what appears to be 0 < |x| < pi.  (Of course, you use range
reduction if the input is outside that range in either case).
 
Could there be a horrible error in Cody & Waite?  I had been
told that this was "the book" for this kind of thing.  Are
there any other major problems, esp. with atan, log, and exp?
Or am I missing something here? (That's entirely possible,
since I'm just a hardware jock doing this).
 
Thanks for any help or insight.
 
BTW I posted to comp.arch because the kind of people who know
the answer might be reading.  If there is a more appropriate bboard,
please let me know so I can post there.

  Phil Koopman                koopman@greyhound.ece.cmu.edu   Arpanet
  2525A Wexford Run Rd.
  Wexford, PA  15090
Senior Scientist at Harris Semiconductor.
I don't speak for them, and they don't speak for me.