home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.sgi
- Path: sparky!uunet!usc!sdd.hp.com!ncr-sd!crash!photon.photon.com!wmc
- From: wmc@photon.com (Bill Cornette)
- Subject: f77 compiler bug?
- Message-ID: <1992Dec15.001050.5454@photon.com>
- Sender: usenet@photon.com (Usenet dummy users)
- Organization: Photon Research Associates, Inc.
- X-Newsreader: TIN [version 1.1 PL6]
- Date: Tue, 15 Dec 1992 00:10:50 GMT
- Lines: 35
-
- I seem to have encountered a bug in our f77 compiler. The following
- code:
-
-
- real tref,temp,xk
- double precision qrot
- c
- data tref /273./, temp /304./, xk /1.50/
- c
- qrot=dprod(tref,1./temp)**xk
- print *,qrot
- c
- qrot=dprod(tref,1./temp)
- qrot=qrot**xk
- print *,qrot
- c
- stop
- end
-
- produces the results:
-
-
- 0.0000000
- 0.85..... (I don't recall the exact number, but it is the corret one)
-
- I don't know of any reason why I can't raise the INTRINSIC function
- dprod to a power directly. Anyone have any comments?
- ---------------------------------------------------------------------------
- ---------------------------------------------------------------------------
- William M. Cornette Photon Research Associates, Inc.
- (619) 455-9741 9393 Towne Centre Drive, Suite 200
- FAX (619) 455-0658 San Diego, California 92121
- wmc@photon.com
- ---------------------------------------------------------------------------
- T.A.N.S.T.A.A.F.L.
-