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!burl!ulysses!cbosgd!ucbvax!seismo.css.gov!mcvax!stl!dww
From: dww@stl.UUCP
Newsgroups: fa.info-vax
Subject: Maths problems with VMS 4.1 FORTRAN
Message-ID: <8509190858.AA12218@stl.UUCP>
Date: Thu, 19-Sep-85 04:58:06 EDT
Article-I.D.: stl.8509190858.AA12218
Posted: Thu Sep 19 04:58:06 1985
Date-Received: Sat, 21-Sep-85 03:23:46 EDT
Sender: daemon@ucbvax.ARPA
Reply-To: info-vax@ucb-vax.arpa
Organization: The ARPA Internet
Lines: 13

One of our users has complained that his Fortran program is now giving
different results from 3 months ago, with identical input data, the
only change being that we've gone from v3.5 to v4.1 and he's
recompiled/relinked.  There's been some discussion on the net about v4.2
problems - are there also problems in v4.1?   

As I never touch FORTRAN myself I can't advise him; as it's a large
package written elsewhere he doesn't know which particular function is
returning a different result (nor even whether v3 or v4 is "right"). 
I guess I should make him do an SPR, but there's too little to go on at
present to identify what's wrong.

If the difference is well known we can just tell him to wait for 4.3!