home *** CD-ROM | disk | FTP | other *** search
/ Light ROM 4 / Light ROM 4 - Disc 1.iso / text / net_news / 1994 / 2.doc / 000228_bdupras@bert.eecs.uic.edu_Sun Dec 11 21:28:14 PST 1994.msg < prev    next >
Text File  |  1995-03-23  |  1KB  |  37 lines
  1. Article: 481 of comp.graphics.packages.lightwave
  2. Xref: netcom.com comp.graphics.packages.lightwave:481
  3. Newsgroups: comp.graphics.packages.lightwave
  4. Path: netcom.com!netcomsv!digdug.pencom.com!uunet!news.ltw.org!newshost.marcam.com!news.mathworks.com!udel!gatech!howland.reston.ans.net!pipex!bnr.co.uk!bcarh8ac.bnr.ca!corpgate!news.utdallas.edu!arco!brian
  5. From: bdupras@bert.eecs.uic.edu (Brian Dupras)
  6. Subject: Re: Pentium Bug WAS Re: LightWave 4.0 Delays
  7. Message-ID: <D0IJBo.5ou@news.arco.com>
  8. Sender: usenet@news.arco.com
  9. Nntp-Posting-Host: 130.201.190.69
  10. Organization: ARCO Oil % Gas Company
  11. X-Newsreader: News Xpress Version 1.0 Beta #1
  12. References: <3bt0b7$5ae@news.cs.tulane.edu> <3c3vlg$8ri@nnrp.ucs.ubc.ca> <3c51l2$e4u@news.cs.tulane.edu> <Pine.SUN.3.91.941207233425.4133A-100000@gaspra.pd.com>
  13. Date: Thu, 8 Dec 1994 21:15:37 GMT
  14. Lines: 20
  15.  
  16. Ernie Wright <ernie@gaspra.pd.com> wrote:
  17.  
  18. [munch]
  19.  
  20. >The worst possible relative error is about 0.00006 (6e-5).  Relative
  21. >errors larger than 0.0000001 (1e-7), approximately epsilon for single
  22. >precision, occur on average once every 65 billion divides.
  23. >
  24. [munch]
  25.  
  26. >- Ernie
  27.  
  28. Really?  Then I must have an error in my Pentium bug.  It seems when I
  29. divide 1048958.75 by 2.999999048, I get an answer that is about 21
  30. off.  The error in the chip is something I'm not too concerned about, but
  31. it's also not just some 7th decimal place screw up.  There are combinations
  32. that produce surprising results.
  33.  
  34. Brian
  35.  
  36.  
  37.