Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.3 4.3bsd-beta 6/6/85; site ucbvax.ARPA
Path: utzoo!watmath!clyde!cbosgd!ucbvax!huey.udel.edu!kathy
From: kathy@HUEY.UDEL.EDU (Kathy Minnich)
Newsgroups: fa.info-vax
Subject: another fortran problem?
Message-ID: <8509200421.AA05283@UCB-VAX.ARPA>
Date: Thu, 19-Sep-85 20:45:42 EDT
Article-I.D.: UCB-VAX.8509200421.AA05283
Posted: Thu Sep 19 20:45:42 1985
Date-Received: Sat, 21-Sep-85 05:31:53 EDT
Sender: daemon@ucbvax.ARPA
Reply-To: info-vax@ucb-vax.arpa
Organization: The ARPA Internet
Lines: 14

we're running VMS 4.1 with FORTRAN 4.2.

one of my users is complaining about receiving different results from
his fortran program when he runs it with the /nooptimize switch.  he claims
that the program will run slower (about 6 hours instead of 4 hours) and that
the results differ.  (he is working with random numbers, but says that the
difference is very noticable.)  if anyone knows in fact that this a bug with 
the optimization, then please let me know ... my user is now thinking that
he's wasted over a month's worth of computer time because of the varying
results and doesn't quite know which set to trust.

thanks for any comments/information,

kathy